osterman.co

Timeline — Travis Osterman | Career & Publications

2026-07-13T00:08:41.000Z

Timeline

One reverse-chronological view of the work: appointments, degrees, board certifications, national service, and the peer-reviewed record, year by year. Roles and service sit at the year they began; degrees, training, and certifications at the year they were completed. Career milestones come from the CV; publication counts and highlights are synced from the same Zotero library that drives the rest of the site. Last updated: 2026-07-12.

2020s

2025
7 publications
Publications 7 peer-reviewed publications
1. SmokeBERT and Beyond: Bridging Clinical Narratives and Structured Smoking Data To Improve Lung Cancer Screening. JCO clinical cancer informatics.
2. Clonal Hematopoiesis of Indeterminate Potential After Radiation Therapy. International Journal of Radiation Oncology*Biology*Physics.
3. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine.
4 more from 2025 on /research/
2024
5 publications
Publications 5 peer-reviewed publications
1. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics.
2. The importance of studying the implementation of cancer data standards. Cancer.
3. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology.
2 more from 2024 on /research/
2023
8 publications
- Role Associate Vice President for Research Informatics, Vanderbilt Health since 2023
- Role Associate Chief Medical Information Officer, Vanderbilt Health since 2023
- Role Ingram Associate Professor of Cancer Research, Vanderbilt-Ingram Cancer Center since 2023
- Role Associate Professor of Biomedical Informatics and of Medicine (Division of Hematology and Oncology), Vanderbilt University Medical Center since 2023
- Honor Fellow of the American Society of Clinical Oncology (FASCO), American Society of Clinical Oncology
Publications 8 peer-reviewed publications
1. Clinician Perspectives Regarding the Impact of Information Technology on Multidisciplinary Tumor Boards: A National Comprehensive Cancer Network Survey. JCO Clinical Cancer Informatics.
2. Next-generation phenotyping: introducing phecodeX for enhanced discovery research in medical phenomics. Bioinformatics.
3. Accuracy and Reliability of Chatbot Responses to Physician Questions. JAMA Network Open.
5 more from 2023 on /research/
2022
6 publications
- Service Chair, Executive Committee, mCODE (minimal Common Oncology Data Elements), American Society of Clinical Oncology since 2022
  Chairs governance of the FHIR-based oncology data standard, implemented at more than 70 institutions across six countries and the data-submission path for the CMS Enhancing Oncology Model.
Publications 6 peer-reviewed publications
1. A Unified Approach to Clinical Informatics Education for Undergraduate and Graduate Medical Education. AMIA ... Annual Symposium proceedings. AMIA Symposium.
2. Associations of influenza vaccination with severity of immune-related adverse events in patients with advanced thoracic cancers on immune checkpoint inhibitors. ERJ open research.
3. Two Uncomplicated Pregnancies on Alectinib in a Woman With Metastatic ALK-Rearranged NSCLC: A Case Report. JTO Clinical and Research Reports.
3 more from 2022 on /research/
2021
8 publications
- Service Member, Azure Research Community Advisory Board, Microsoft since 2021
Publications 8 peer-reviewed publications
1. Oncologist Perspectives on Telemedicine for Patients With Cancer: A National Comprehensive Cancer Network Survey. JCO Oncology Practice.
2. Follow-up Interactive Long-Term Expert Ranking (FILTER): a crowdsourcing platform to adjudicate risk for survivorship care. JAMIA open.
3. My Cancer Genome: Coevolution of Precision Oncology and a Molecular Oncology Knowledgebase. JCO Clinical Cancer Informatics.
5 more from 2021 on /research/
2020
3 publications
- Honor Microsoft Investigator Fellowship, Microsoft
Publications 3 peer-reviewed publications
1. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clinical Cancer Informatics.
2. Seven decades of chemotherapy clinical trials: a pan-cancer social network analysis. Scientific Reports.
3. Trends in FDA cancer registration trial design over time, 1969-2020.. Journal of Clinical Oncology.

2010s

2019
4 publications
- Role Director of Cancer Clinical Informatics, Vanderbilt-Ingram Cancer Center since 2019
- Role Co-Director, Data Science Shared Resource, Vanderbilt-Ingram Cancer Center since 2019
- Service Member, Adult Oncology Steering Board, Epic since 2019
- Service Founding Chair, Beacon Community Operations Group, Epic 2019-2025
- Honor Fellow of the American Medical Informatics Association (FAMIA), American Medical Informatics Association
Publications 4 peer-reviewed publications
1. Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow. JCO Clinical Cancer Informatics.
2. Hypertension and use of bevacizumab among patients treated in community settings.. Journal of Clinical Oncology.
3. Learnings from a pragmatic study to evaluate benefit of performing reflex clinical trial matching and providing clinical decision support to physicians.. Journal of Clinical Oncology.
1 more from 2019 on /research/
2018
2 publications
- Service Member, Electronic Health Record Workgroup, National Comprehensive Cancer Network since 2018
Publications 2 peer-reviewed publications
1. Impact of the influenza vaccination on cancer patients undergoing therapy with immune checkpoint inhibitors (ICI).. Journal of Clinical Oncology.
2. Phenotype risk scores identify patients with unrecognized Mendelian disease patterns. Science.
2017
2 publications
- Education Vanderbilt University, Master of Science in Biomedical Informatics
Publications 2 peer-reviewed publications
1. Evaluating phecodes, clinical classification software, and ICD-9-CM codes for phenome-wide association studies in the electronic health record. PloS One.
2. Utility of adding clinical data to a molecular results portal for improving clinical trial prescreening efficiency.. Journal of Clinical Oncology.
2016
- Role Clinical Director, Office of the Chief Health Information Officer, Vanderbilt University Medical Center 2016-2023
- Training Fellowship, Medical Oncology, Vanderbilt University Medical Center 2013-2016
- Board cert Board Certification in Medical Oncology, American Board of Internal Medicine
- Board cert Board Certification in Clinical Informatics, American Board of Preventive Medicine
- Honor Young Investigator Award, Conquer Cancer Foundation
2015
1 publication
Publications 1 peer-reviewed publication
1. Advances in website information resources to aid in clinical practice. American Society of Clinical Oncology educational book / ASCO. American Society of Clinical Oncology. Meeting.
2012
- Training Internship and Residency (Chief Resident, 2012), Internal Medicine, Indiana University 2009-2012

2000s

2009
- Education Nova Southeastern University College of Osteopathic Medicine, Doctor of Osteopathic Medicine
2004
1 publication
Publications 1 peer-reviewed publication
1. A software utility for creating interactive maps for 2D gel-based proteomics. Analytical Biochemistry.
2003
- Education University of Indianapolis, Bachelor of Science in Computer Science and Biology, magna cum laude

By the Numbers — Travis Osterman | Research, Talks & mCODE Scale

2026-07-13T00:08:41.000Z

By the numbers

A quantified view of Dr. Osterman's professional record. Every count below is generated from the same source-of-truth library that powers /research/, /speaking/, and /press/, so the figures here match the underlying records and stay current as new work is added.

Scholarly record

Publications, presentations, and coverage, counted from the library and linked to the full list for each. Last updated: 2026-07-12.

47

Peer-reviewed publications

Full list on /research/
48

Invited presentations

Full list on /speaking/
32

Scientific-meeting presentations

Full list on /speaking/
24

Media mentions

Coverage on /press/
28

Conference abstracts

Full list on /research/
1

Patent

Full list on /research/

mCODE at scale

Dr. Osterman chairs the minimal Common Oncology Data Elements (mCODE™) Executive Committee. Implemented at more than 70 institutions across six countries.

70+

Institutions implementing mCODE

mCODE Implementation Guide · /leadership/
6

Countries with mCODE implementations

mCODE Implementation Guide · /leadership/

Distinctions

CMS Enhancing Oncology Model: mCODE is the only method of submitting data to CMS' Enhancing Oncology Model. Leadership · mCODE Implementation Guide.
Structured genomic data: Vanderbilt Health has more structured genomic data in its electronic health record than any other institution in the United States. Leadership.

Now — Travis Osterman | Current Focus & Recent Activity

2026-07-13T00:08:41.000Z

Now

This is a now page - a snapshot of what has my attention this season. It is kept honest by the same source-of-truth library that drives the rest of the site, so it updates when new work actually lands, not on a schedule.

Current focus

Day to day I practice medical oncology and lead research informatics at Vanderbilt Health, and I chair the mCODE Executive Committee. The through-line right now is the same one the whole site is about: getting clinical oncology, applied AI, and health-data standards to actually interoperate. That means structured genomics in the electronic health record, real-world prediction of immunotherapy outcomes and toxicity, and keeping mCODE usable as it becomes regulatory infrastructure for the CMS Enhancing Oncology Model.

The roles that anchor that work today:

Associate Vice President for Research Informatics, Vanderbilt Health
Associate Chief Medical Information Officer, Vanderbilt Health
Ingram Associate Professor of Cancer Research, Vanderbilt-Ingram Cancer Center
Director of Cancer Clinical Informatics, Vanderbilt-Ingram Cancer Center
Co-Director, Data Science Shared Resource, Vanderbilt-Ingram Cancer Center
Associate Professor of Biomedical Informatics and of Medicine (Division of Hematology and Oncology), Vanderbilt Health
Chair, mCODE Executive Committee, American Society of Clinical Oncology

More on each: leadership & governance · six domains of expertise · full bio.

Recent activity

New publications, invited talks, scientific-meeting presentations, and press - drawn from the same library that drives /research/, /speaking/, and /press/. Last updated: 2026-07-12.

Latest (last 12 months)

Invited talk HL7 International Working Group Meeting (Europe): "mCODE Overview and Current State". May 18, 2026
Invited talk Vanderbilt Lecture Series CME (Nashville, Tennessee): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". May 7, 2026
Abstract Abstract LB385: Polygenic risk score of genetic variants in genes encoding drug-metabolizing enzymes and drug transporters, in association with febrile neutropenia. Cancer Research. Apr 17, 2026
Invited talk 2026 NCCN Annual Conference (Orlando, Florida): "Harnessing Artificial Intelligence to Improve Oncology Care". Mar 28, 2026
Invited talk NASEM Workshop: Policy Issues for Integrating Artificial Intelligence in Cancer Research and Care (Washington, DC): "AI in Cancer Care: 2 Wins, 2 Current Challenges". Mar 9, 2026
Invited talk Jackson-Madison County General Hospital (Jackson, Tennessee): "An Update on the Use of AI in Clinical Practice". Mar 6, 2026
Invited talk Blanchfield Army Community Hospital (Fort Campbell, Kentucky (virtual)): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". Feb 10, 2026
Peer-reviewed publication SmokeBERT and Beyond: Bridging Clinical Narratives and Structured Smoking Data To Improve Lung Cancer Screening. JCO clinical cancer informatics. Dec 22, 2025
Invited talk "Leveraging Data to Day-to-Day Improvements". Oct 31, 2025
Peer-reviewed publication Clonal Hematopoiesis of Indeterminate Potential After Radiation Therapy. International Journal of Radiation Oncology*Biology*Physics. Oct 24, 2025
Peer-reviewed publication Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine. Oct 15, 2025
Patent Model generation apparatus for therapeutic prediction and associated methods and models. Sep 18, 2025
Scientific meeting presentation The Evolving Artificial Intelligence Landscape in Cancer Care: "AI and the Cancer Journey: Navigating New Frontiers in Policy and Technology". Sep 9, 2025
Peer-reviewed publication Artificial intelligence across the cancer care continuum. Cancer. Aug 15, 2025
Peer-reviewed publication Advancing the science of genomic learning healthcare systems. Learning Health Systems. Jul 23, 2025

Latest writing

Longer arguments live in Essays; quick reactions to new research, policy, and product launches live in Notes. Most recent:

Essays

2026-07-12

Bringing genomics out of the PDF and into the chart
A tumor sequencing result that lives as a faxed PDF can't trigger an alert, match a trial, or warn the next oncologist, and most of them still do.
2026-07-05

The oncologist's case for clinical informatics as a discipline
If you've never written a note in the system you're trying to fix, you will optimize the wrong thing.
2026-06-28

Predicting immunotherapy toxicity before it happens
The data to flag who will develop immune-related toxicity is already in the chart; we just haven't been reading it for that.

Find me elsewhere

Subscribe via the RSS feed to get new work as it lands, or read the machine-readable CV and llms.txt map. ORCID, Google Scholar, LinkedIn, X, and GitHub are linked in the footer. For consulting, speaking, or advising, use the contact form - engagements are managed through WSOR, LLC.

Collaboration Network — Travis Osterman | Co-Authorship Graph

2026-07-13T00:08:41.000Z

Collaboration network

Travis Osterman's research is collaborative by design. This page maps the co-authorship network behind it: everyone he has co-published with on at least 2 papers or abstracts, drawn from the same bibliographic library that powers the rest of this site.

Co-authorship network for Travis Osterman: 108 collaborators, each sharing at least 2 papers or abstracts with him.

How to read it

The graph is force-directed. Each node is a co-author, and its size reflects how many papers and abstracts that person shares with Travis Osterman. A link connects two people who appear on the same work, so tightly connected clusters correspond to standing research teams rather than one-time author lists.

The network includes 108 collaborators joined by 720 co-authorship links. A person only enters the graph once they reach 2 shared items with him, which keeps incidental single co-authorships from crowding the picture and leaves a map of the people he actually works with repeatedly.

Most frequent co-authors

Ranked by the number of papers and abstracts each person shares with Travis Osterman. The full network runs to 108 collaborators; the 15 below each share 4 or more.

Top co-authors of Travis Osterman by shared publications and abstracts
Collaborator	Shared papers & abstracts
Jan Wolber	12
Christine Micheel	11
Levente Lippenszky	9
Michele LeNoue-Newton	9
Eszter Csernai	7
Neha M. Jain	6
Ben Ho Park	6
Kathleen Mittendorf	6
Dan M. Roden	5
Tricia Heinrichs	5
Debra L. Friedman	5
Wei-Qi Wei	5
David Smith	5
Kathleen F. Mittendorf	4
Pablo Napan-Molina	4

What it shows, and where it comes from

Two patterns stand out. A dense core of frequent co-authors sits at the center, people who appear together across many papers and form the teams behind Travis Osterman's informatics and oncology work; his most frequent co-author appears on 12 shared items. Around that core is a wider rim of colleagues connected through one or two projects. Node size tracks how central a person is to the ongoing work, not how important any single paper was.

Data source. The network is generated from the same Zotero bibliographic library that drives /research/ and /speaking/, using the author lists on every indexed paper and abstract. A collaborator qualifies at 2 or more shared items, so even the smallest nodes represent a repeated collaboration rather than a single co-authorship. The full node-and-link dataset is published as JSON at /assets/data/coauthor-network.json. Because names are taken verbatim from the source records, a collaborator whose name is recorded two different ways can appear as more than one node, which can split a single person's true count across those nodes. Last updated 2026-07-12, when the graph was last regenerated from the library.

About Travis Osterman, DO — Oncology & Informatics Executive

2026-07-13T00:08:41.000Z

About

Dr. Travis Osterman is a practicing medical oncologist and biomedical informatics executive. At Vanderbilt Health he is Associate Vice President for Research Informatics, Associate Chief Medical Information Officer, the Ingram Associate Professor of Cancer Research, and Director of Cancer Clinical Informatics at the Vanderbilt-Ingram Cancer Center. He chairs the minimal Common Oncology Data Elements (mCODE™) Executive Committee.

Origin

A computer science course Travis Osterman never planned to take is the reason this site exists.

In the fall of 1999, his undergraduate advisor at the University of Indianapolis signed him up for CS 100. He hadn't asked to take it. He took it anyway, and enjoyed it, and signed up for the next one. He kept doing that until he had a second major. By 2003 he graduated magna cum laude in both Biology and Computer Science - not because he had set out to combine the two, but because parallel curiosity took him there one course at a time.

That pattern - follow the interesting problem until it becomes a discipline - is the through-line of the next two decades.

After medical school at Nova Southeastern College of Osteopathic Medicine (where he served as class president each of four years) and internal medicine residency at Indiana University (chief resident at Wishard Memorial), Travis came to Vanderbilt for oncology fellowship in 2013. Vanderbilt was deliberate. It was a rare opportunity: train at an institution with an exceptional reputation in both medical oncology and biomedical informatics. He came planning to add a Master's in Biomedical Informatics on top of the fellowship. It was the right bet. The fellowship finished in 2016; the MS in 2017. The fellowship made him an oncologist; the MS gave formal vocabulary to the work he had been quietly doing since CS 100 in '99.

The pivot from "oncologist who also codes" to chairing an international cancer data standard wasn't a leap. It was a network compounding. By 2017 he was on the faculty at Vanderbilt-Ingram Cancer Center and active across multiple ASCO committees. Mentors helped him route into the broader cancer informatics national landscape. When the original mCODE working group was being assembled, he got the invitation and said yes immediately. He has been in the room ever since - eventually as chair of the executive committee that now stewards the standard implemented at more than seventy institutions across six countries, and that serves as the only method of submitting data to CMS' Enhancing Oncology Model.

The role of "physician executive" was a slower realization. The work changed before the title did - a long journey from individual contributor, to managing a team, to becoming the person people look to for direction. The transitions came as the questions got bigger. How do you implement structured genomic data into the EHR for tens of thousands of patients? How do you align Epic, Microsoft, Tempus AI, GE HealthCare, and NCCN around a shared model for cancer data when each company has its own infrastructure? Those aren't problems an individual contributor solves. He grew into the role the work required.

What he actually enjoys, in his own words, is fixing things and solving problems. Specifically: building better systems around automation and standardization that improve workflows in both clinical care and research. The reason mCODE matters to him isn't the standards politics - it's that mCODE is the systems improvement. Fewer custom mappings between institutions. Fewer one-off integrations. Less friction between the data clinicians enter and the data researchers need. A standard is a force multiplier for everyone downstream.

The same instinct shows up outside the day job. He started a Linux configuration documentation site called gentoovps.net in 2011, during his internal medicine residency, and that site has evolved into today's fld.sh. He runs his own infrastructure-as-code, his own self-hosted journal, his own multi-agent orchestrator. He built school.osterman.co - a self-paced STEM platform with more than 400 courses - so his children would have something interesting to do when the school year ended. The same person who chairs mCODE for cancer-data interoperability has been writing software for himself, his family, and his curiosity since college.

It is the same instinct, applied at every scale. Find something that doesn't work as well as it should. Build the system that fixes it. Repeat.

Practice and leadership

Under his leadership at Vanderbilt-Ingram, the Vanderbilt Health electronic health record contains more structured genomic data than any other institution in the United States. Nationally, the mCODE™ standard is implemented at more than 70 institutions across six countries and serves as the only method of submitting data to CMS' Enhancing Oncology Model.

Leadership

Mid-Career Leadership Development Program - Vanderbilt University School of Medicine, 2020
Chief Resident - Department of Internal Medicine, Wishard Memorial Hospital, Indiana University, 2012–2013
Class President - Nova Southeastern College of Osteopathic Medicine, 2004–2009
Student Body President - University of Indianapolis, 2001–2002

Boards and credentials

Board certified in Medical Oncology - American Board of Internal Medicine (2016)
Board certified in Clinical Informatics - American Board of Preventive Medicine (2016)

Training

MS, Biomedical Informatics - Vanderbilt University, 2017
Fellowship, Medical Oncology - Vanderbilt University Medical Center, 2013–2016
Internship and Residency, Internal Medicine - Indiana University, 2009–2012 (Chief Resident, 2012)
DO - Nova Southeastern College of Osteopathic Medicine, 2009
BS, Computer Science and Biology, magna cum laude - University of Indianapolis, 2003

Selected honors

Fellow, American Society of Clinical Oncology (FASCO), 2023
Microsoft Investigator Fellowship, 2020
Fellow, American Medical Informatics Association (FAMIA), 2019
Conquer Cancer Foundation Young Investigator Award, 2016
DO Student of the Year - Nova Southeastern College of Osteopathic Medicine, 2008
Outstanding Student in Computer Science - University of Indianapolis, 2000, 2001, 2002

Explore

Now - what he is focused on this season
Career timeline - roles, training, and publications by year
By the numbers - the record quantified, each figure sourced
Collaboration network - the co-authorship graph and his most frequent co-authors

Profiles

Vanderbilt Biomedical Informatics profile
PubMed / NCBI bibliography
Wikidata (Q101153359)
Machine-readable CV (JSON Resume) - structured career data for agents and crawlers

ORCID, Google Scholar, LinkedIn, X, and GitHub are linked in the footer.

Advisory disclosures and industry partnerships: /collaborations/.

Bringing genomics out of the PDF and into the chart

2026-07-12T00:00:00.000Z

A patient of mine had an actionable mutation sitting in his chart for months before anyone acted on it. The result was there, it just wasn't anywhere a computer could read it. It was a PDF, scanned into the record, indistinguishable to the EHR from a release-of-information form.

That is the ordinary state of cancer genomics in 2026, and it is the problem I want to take apart. We have spent a decade getting better at generating molecular results: broad panels, liquid biopsies, whole-exome and whole-genome sequencing, faster turnaround. We have spent almost none of that decade getting those results into a form the patient's record can actually use. The sequencing got cheaper and richer; the last mile, from the reference lab's report to a field in the chart that can drive a decision, barely moved.

A PDF is not data

Picture how a genomic result usually arrives. A tumor block goes out to a reference laboratory. Weeks later a report comes back, a richly formatted document, often a dozen pages, listing variants, allele frequencies, therapy implications, and trial suggestions. It is faxed or transmitted, and someone scans or attaches it to the EHR. From that moment it is visible and useless in equal measure. A human can open it and read it. Nothing else can.

The distinction that matters here is between a document and a datum. A PDF is a picture of information. It cannot be queried. It cannot trigger a clinical decision support rule when a new targeted therapy is approved for the variant it describes. It cannot be matched against the eligibility logic of an open trial. It cannot be counted in a cohort, fed to a prediction model, or submitted to a quality program. The EGFR exon 19 deletion on page seven might as well not exist, as far as the rest of the system is concerned, and the rest of the system is where care increasingly gets coordinated.

This is not a hypothetical gap. When a multi-institution group I worked with catalogued the real-world barriers to using biomarker results, the failure modes were depressingly concrete: results arriving as unstructured documents, no consistent place in the record for them to land, no way to surface them at the moment of a treatment decision. We wrote it up as Integrating electronic health records to facilitate cancer biomarker testing (Huelsman et al., Journal of Clinical Oncology, 2024). The headline finding is mundane and damning: the testing is rarely the bottleneck. The integration is.

What "structured" actually buys you

When a genomic result enters the record as structured data, discrete, coded variants in defined fields rather than ink on a scanned page, three things become possible that were impossible before, and all three are the difference between sequencing as a line item and sequencing as care.

The result can warn you. Decision support can watch the patient's structured variants and fire when the evidence changes, a new approval, a label expansion, a newly opened trial for exactly that alteration. The oncologist learns about the option in time to use it, instead of months later or never.
The result can match. Trial-matching logic and molecular tumor board prep run against a queryable variant, not a human's memory of a PDF they read last spring. The patients who qualify for something get found.
The result can be counted. Cohorts, outcomes research, real-world evidence, and quality submissions all need the genomics as data. Structured capture is what makes a patient's molecular profile legible to the institution, not just to the one clinician who happened to open the attachment.

None of this requires a new model or a new algorithm. It requires the unglamorous plumbing that turns a report into a record. That plumbing is the work I have spent years on, and it is harder and more valuable than it sounds.

Being eighth, and then being first

Vanderbilt Health started early. In 2019, when exactly one institution in the country had integrated genomic results from a reference laboratory directly into Epic, we became the eighth. That sounds like a modest distinction until you sit with the denominator: in a country with thousands of hospitals and hundreds of cancer programs, fewer than ten had managed to get reference-lab genomics flowing as structured data into the dominant EHR. The hard part was never wanting it. The hard part was the integration.

Going forward was only half the problem. The other half was the back catalog, thousands of patients already on therapy whose molecular reports were sitting in the record as PDFs and scanned attachments, invisible to every downstream system. A cut-over that only structures new results leaves the existing population stranded. So we built an extract-transform-load pipeline that converts that historical EHR genomic data into standardized, queryable profiles built on HL7 FHIR, the same interoperable shape the standards work targets. The result is a unified structured corpus across a patient's entire history, not a clean record from a cut-over date forward and a fog behind it. The upstream workflow is documented in the NCCN abstract Integrating Structured Genomic Data Into the Electronic Health Record (Vento & Osterman, JNCCN, 2023, BIO23-019).

The compounding effect of being early is the part worth internalizing. By the end of 2021 there were roughly twelve thousand tumor genomic reports living as structured data inside our EHR. Today Vanderbilt Health holds more structured genomic data in its electronic health record than any other institution in the United States. That position is not the product of a single clever build; it is the product of starting years before most peers and never stopping. The institutions that begin this work now are not behind by a feature. They are behind by a corpus, and a corpus is the one thing you cannot buy or backfill instantly. You accrue it.

The unlock, not the trophy

The #1 ranking is not the point. The point is what the structured corpus makes possible downstream, and almost everything I care about in precision oncology sits on top of it. Trial matching at scale, molecular-tumor-board governance, the clinical genomics workstream that ties it together, point-of-care alerting when a patient's variant becomes actionable, real-world toxicity studies, industry research collaborations that need an interoperable surface to reach the data, none of them work against a pile of PDFs. They all work against structured fields. The integration is the load-bearing layer; the celebrated applications are what you stack on top once it holds.

This is also why I keep returning to standards. Structured data inside one institution's EHR is necessary but not sufficient, a research collaborator across town still can't reach it without an interoperable shape to read. That is the role of mCODE, the FHIR-based minimal Common Oncology Data Elements, which carries the genomics among the core fields a cancer record should expose. Integration makes the data usable inside the walls; the standard makes it portable beyond them. You want both, in that order: structure first, then portability.

What to do about it

If you run a cancer program, a lab, or the informatics behind one, the actionable conclusion is narrow and stubborn. Stop measuring genomics by tests ordered and start measuring it by structured results landing in the chart where something can act on them. Treat reference-lab integration as a first-class capital project, not an IT ticket. Insist that your labs and your EHR vendor deliver coded, discrete results, not prettier PDFs. And build the backload, not just the cut-over, because the patients already on treatment are the ones a stale record fails first.

The sequencing is the easy half now. The molecular biology that took a generation to make routine is, clinically, the solved part. The unsolved part is the most ordinary-sounding question in medicine: can the chart actually use what the lab just told you? Until the answer is yes, until the variant on page seven is a field the system can read, every dollar spent on the test is spent generating a document no one downstream can act on. The work that turns that document into care is the work worth funding. For the longer arc of how we built it at Vanderbilt Health, see clinical genomics in the EHR.

The oncologist's case for clinical informatics as a discipline

2026-07-05T00:00:00.000Z

A few years into building tools inside our electronic health record, I watched a well-intentioned redesign of the in-basket, the clinician's message queue, get most of the way to launch before anyone in the room had recently lived a clinic day inside it. The logic was clean. The interface was modern. And it would have added clicks to the single most time-pressured part of an oncologist's afternoon, because the people designing it had reasoned about the workflow instead of carrying it. I had carried it that week. That is the whole argument of this essay, compressed into one anecdote: clinical informatics is a clinical discipline, and it degrades in predictable ways when the people leading it are not also the people doing the work it governs.

I say this as someone with standing on both sides of the line. I'm a practicing medical oncologist who still attends on the solid-tumor service. I'm also board certified in clinical informatics, the American Board of Preventive Medicine subspecialty, which I sat for in 2016, the same year I finished my oncology boards. Informatics is not a hobby I do adjacent to medicine. It is a board- certified medical specialty, with its own body of knowledge, its own examination, and its own standard of practice. We should treat it like one.

Informatics is a clinical specialty, not a technical service line

The most common failure mode I see in health systems is organizational: clinical informatics gets filed under IT, and clinicians are invited in as "advisors", consulted late, thanked politely, and routed around when the schedule gets tight. The arrangement feels reasonable. The systems are technical, the vendors are technical, the budgets live in technology. So the clinician becomes a stakeholder to be managed rather than a leader who decides.

This is backwards, and the certification exists precisely because it's backwards. Clinical informatics became a board-certified subspecialty in 2013 for the same reason cardiology or critical care did: the work requires a clinician's judgment applied to a defined domain, and that judgment cannot be outsourced to people who don't carry the clinical consequences. A decision about how staging data is captured, how a best-practice alert fires, or how a genomic result lands in the chart is not primarily a software decision. It is a decision about how medicine gets practiced, made through software. Whoever owns it owns the clinical workflow whether they realize it or not.

The "advisor" model fails on a specific mechanism: it separates authority from accountability. The technical team has the authority to ship; the clinician has the accountability for what happens at the bedside, but no real control over the thing that produced it. When those come apart, you get systems that are internally coherent and clinically wrong, and nobody who can fix them is also the person who has to live with them. Putting a practicing clinician at the helm re-fuses authority and accountability in one person. That is not a courtesy to clinicians. It is a control mechanism for the institution.

What changes when a practicing clinician actually leads

The difference shows up in what gets optimized. I direct cancer clinical informatics at our cancer center, and I still see patients. That second clause is not a credential I list for color, it's the thing that keeps the first one honest. When you write notes in the system you're trying to improve, you stop optimizing for the demo and start optimizing for the Tuesday afternoon when the clinic is running forty minutes behind. You feel the cost of the extra click because you paid it that morning. You know which alerts get reflexively dismissed because you've dismissed them. You can tell the difference between a field that's hard to capture and a field nobody captures because capturing it helps no one.

This is also why the leader-versus-advisor distinction is not semantic. An advisor reacts to a roadmap someone else set. A clinician-leader sets the roadmap, which means clinical reality shapes the work from the first decision rather than being consulted as a courtesy near the end. The redesign I opened with got caught in time, but it got caught because a clinician with authority was in the room, not because the process was built to surface clinical cost early. In too many organizations, that catch depends on luck. It should depend on org-chart.

There's a research analogue I find clarifying. Across years of building real-world models in oncology, the constraint was never the algorithm; it was whether the clinical inputs existed as trustworthy structured data. The people who could see that bottleneck were the ones who understood both the modeling and the clinic. Leadership is the same shape of problem. The bottleneck in clinical informatics is rarely the technology. It's the translation between what medicine needs and what software does, and translation is a job for someone fluent in both, with the authority to act on the translation.

The pipeline problem, and why education is the lever

If clinical informatics needs clinicians at the helm, the obvious question is where they come from. The honest answer is that we have not built the pipeline to produce them, and that gap is the one I've spent the most deliberate effort trying to close.

The problem starts earlier than people assume. We tend to treat informatics as something a clinician picks up late, a fellowship, a master's, a mid-career pivot. But the clinicians who are most fluent are the ones who encountered the discipline as a normal part of becoming a physician, not a specialty bolted on afterward. That conviction is why I built clinical informatics into the medical student curriculum and created a graduate medical education elective, and why my colleague Dara Mize and I argued for a unified approach to clinical informatics education spanning undergraduate and graduate medical education at the 2022 AMIA Annual Symposium. The point of "unified" is that the medical student, the resident, and the fellow are on one continuum, not three disconnected audiences, and the discipline should be present at every stage rather than appearing, abruptly, as an option for the already-committed.

It also has to be built as an institutional strategy, not a personal hobby of whoever happens to care. Our group laid out exactly that in the Vanderbilt Clinical Informatics Center's education strategy in Applied Clinical Informatics in 2025, a deliberate, center-level plan for training the next generation rather than a collection of one-off courses depending on individual enthusiasm. The difference matters because individual enthusiasm leaves when the individual does. A discipline that wants practitioners at the helm has to manufacture them on purpose, at scale, with the same intentionality we bring to any other specialty's training.

A short version of what I think the pipeline owes the field:

Expose every trainee early. Informatics belongs in medical school and residency as a normal competency, not a niche elective for the already-converted.
Build dual fluency, not translation by committee. The goal is clinicians who can reason in both clinical and technical terms themselves, so the translation happens inside one head instead of across a meeting.
Make it a career, not a side quest. There have to be real leadership roles , with authority and accountability fused, for trained clinician-informaticians to grow into, or the training produces advisors by default.

What I'd ask of the people who staff these roles

So this lands somewhere actionable, here is what I'd ask of two audiences.

If you're an academic leader or sit on a search committee for a CMIO, an informatics director, or a "physician champion" role, write the job as a leadership position with decision authority over the clinical systems it covers, not an advisory seat reporting into a technical org that can route around it. And weight active clinical practice as a qualification, not a nice-to-have. A leader who has stopped seeing patients is, within a few years, an advisor in everything but title; the feedback loop that keeps their judgment calibrated has gone quiet. Protected clinical time for an informatics leader is not a concession. It is the instrument that keeps the role honest.

If you're a trainee weighing this path: you do not have to choose between practicing medicine and shaping how medicine is practiced. The board certification exists, the career exists, and the field needs people who can hold both at once far more than it needs more pure technologists or more clinicians who merely complain about the software. The most useful thing you can be is the person in the room who carried the workflow that morning and can also change it that afternoon.

Clinical informatics will keep growing whether or not clinicians lead it. The only open question is whether the systems we all practice inside get designed by people who practice inside them. I've staked my own career on the answer being yes, in the clinic, in the education work, and in the leadership roles where the two finally meet. The field is young enough that the people entering it now will decide whether "clinical informatics" keeps the word clinical doing real work, or lets it fade into a polite adjective. Make it do the work.

Home Lab — Travis Osterman | Self-Hosted & Agentic Systems

2026-06-28T23:29:30.000Z

Home Lab

Independent software work outside the clinical informatics role. These are personal projects - exploratory, opinionated, and explicitly not part of the leadership work documented elsewhere on this site. They exist to keep the hands close to the metal and an outlet for a never-ending desire to learn and explore.

Summer School
public

Self-paced, interactive STEM learning platform for kids ages 7-13, with more than 400 interactive courses. Built as a platform for my children to continue learning once the school year ends.

school.osterman.co
Faculty Coaching
public

Academic coaching practice focused on faculty in academic medicine - including career coaching for physicians. Specifically passionate about coaching physicians interested in building a career in Health IT.

facultycoaching.com
Five-Letter Domain [fld.sh]
public

Real-world technical solutions and configurations from actual infrastructure work. What started as personal notes has evolved into a comprehensive resource covering Linux system administration, networking, storage management, and virtualization.

fld.sh
Orc AI
private

Multi-agent workflow orchestrator that tracks agent usage and can increase or decrease work to optimize subscription-based token usage. Runs my personal development infrastructure - including the project migrations and tooling work behind this site - under a single goal-driven agent runtime.

Built in Rust.
OJournal
invite only

Self-hosted, privacy-first journaling. Entries, attachments, and search index are encrypted at rest under a three-tier key hierarchy (UK → SK → DK); the database on its own is useless to an attacker. Designed to run on a single box behind TLS. First began as a Java-based blogging platform in 2004, ported to PHP, and now to a Rust API with a Vue frontend.

ojournal.osterman.co

Rust / Axum backend; Vue frontend.
Inventory
private

Source-of-truth inventory system for physical and virtual hosts at home and in offsite datacenters, inspired by NetBox Labs.
IAC
private

Ansible-based infrastructure-as-code for the homelab and offsite-datacenter hosts that run everything above. Configuration in version control, idempotent provisioning, no manual SSH. Ultimate goal is better-than-cloud self-hosted infrastructure.

Ansible.
keyboot
alpha

A greenfield boot stage for ZFS-on-LUKS hosts. It unlocks every encrypted device from an age-encrypted keyfile, imports the root pool, offers a boot environment picker over TTY or SSH, and kexecs into the chosen kernel. The same artifact doubles as a netboot rescue shell and unattended installer for Gentoo, Debian, and Alpine on UEFI and BIOS.

packages.osterman.co

Built in Rust.
Drop
invite only

Simple API and web interface to support passing secrets to agents while keeping the secrets out of the chat log.

drop.osterman.co
otoken
invite only

A centralized token, authorization, and secrets-broker service for a fleet of self-hosted apps. It issues one opaque token type carrying per-service permission grants, tells each service what a presented token may do, and brokers the upstream credentials a caller is entitled to - replacing the per-app token sprawl where every service mints and stores its own. New projects connect through a browser-consented device-auth flow.

otoken.osterman.co

Rust + PostgreSQL backend; Vue frontend.
swim-assist
public

A website for families to track USA Swimming swimmers' progress: meet times, age-group motivational standards, championship and sectional cuts, goal setting, and a little gamification. Mobile-first and print-friendly, with USA Swimming kept as the source of truth. A useful first version shipped in two days.

swim.osterman.co

Rust / Axum + SQLite backend; Vue frontend.
itop
private

A lightweight, single-row system monitor for zellij panes - an htop alternative that reads /proc directly and renders one compact line of CPU, I/O, memory, swap, and load. Negligible footprint, built to run many instances at once.

Built in Rust.

Predicting immunotherapy toxicity before it happens

2026-06-28T00:00:00.000Z

A patient 2 months into a checkpoint inhibitor calls the triage line short of breath. By the time the CT confirms pneumonitis, they are on high-dose steroids and the immunotherapy is on hold, sometimes for good. Every oncologist who prescribes these drugs has watched some version of this. The toxicity is real, it can be severe, and the way we catch it today is almost entirely reactive: we wait for the patient to feel sick, then we work backward.

The hope is that, in some cases, the warning was already sitting in the chart. Not in a biomarker we hadn't ordered, but in the labs, the prior imaging, the medication history, the comorbidities we already had. We just weren't reading those data to answer the question who is about to get sick.

The signal is already collected

Over a multi-year Digital Precision Oncology collaboration with GE HealthCare, my group and I built and externally validated machine-learning models that predict both the effectiveness and the toxicities of immune checkpoint inhibitors from real-world patient data, the kind that accumulates in any EHR running an oncology service. We published the core results in JCO Clinical Cancer Informatics in 2024 (Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data), and presented validation and extensions at SITC in 2023 and 2024. One result mattered more than the headline accuracy: the models worked off a short medical history captured before the first dose. You don't need years of pristine longitudinal follow-up to stratify risk. You need what's in front of you the day you write the order.

That reframes the clinical question. The point of a toxicity model is not to predict the future for its own sake. It's to change who you watch, how closely, and starting when. A patient flagged high-risk for pneumonitis is a patient whose new cough gets a same-day call back instead of a routine message, whose baseline pulmonary status you actually documented, whose first hint of hypoxia triggers imaging rather than reassurance. The model doesn't make the diagnosis. It re-allocates clinical attention toward the people most likely to need it.

Three toxicities, three different data shapes

Immune-related toxicities are not one problem, and the data that predicts them isn't one shape either. That distinction is what makes this tractable rather than aspirational.

Pneumonitis turns out to be partly visible on imaging the patient already had. In JCO CCI in 2025 we showed that a radiomics-and-deep-learning model reading a prior thorax CT (a scan ordered for staging, not for any toxicity question) carries signal about who later develops immunotherapy-induced pneumonitis (Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography). The pixels were sitting in the PACS archive the whole time.
Hepatitis shows up in the rhythm of routine bloodwork. We built a rolling-window model that watches the ordinary CBC and chemistry panels every patient on therapy already gets, and flags the trajectory toward checkpoint-inhibitor hepatitis before it crosses a clinical threshold. No new test, just reading the labs as a time series instead of one value at a time.
Colitis is the hardest, because the ground truth is buried in narrative. You can't train or monitor a colitis model without reliably knowing who actually had colitis, and that fact lives in clinic notes, not a structured field. We tackled that with accelerated curation, pulling checkpoint-inhibitor colitis cases out of the EHR semi-automatically rather than by brute-force chart review (Accelerated curation of checkpoint inhibitor-induced colitis cases from electronic health records, JAMIA Open, 2023).

That last one is the tell. The bottleneck on toxicity prediction is rarely the algorithm. It's whether the outcome you're predicting exists as trustworthy, structured data, which is the same problem I keep writing about elsewhere and the reason I spend so much time on data standards. A model is only as good as the labels you trained it on, and in oncology those labels are expensive precisely because the truth is written in prose.

What it takes to actually act on a prediction

A risk score that nobody acts on is worse than no score, because it spends trust. If we're going to put these models in front of clinicians, the harder work is downstream of the math.

First, the prediction has to land in the workflow, not in a dashboard someone visits on a good day. A pneumonitis risk flag belongs on the treatment plan and the triage protocol; it should change what the nurse does when the patient calls, automatically. If acting on it requires remembering to go look, it won't get acted on.

Second, you have to decide what the action is, per toxicity, in advance. "High risk" is not a plan. Earlier imaging? A lower threshold for holding the next cycle? A scheduled check-in at the window when that toxicity typically emerges? Tighter lab cadence for the hepatitis-flagged patient? The model buys you lead time; the protocol decides what to do with it. Without the protocol, a flag is just anxiety.

Third, and this is the part payers and health-system leaders should care about: you have to measure the model in your population before you trust it, and keep measuring after you deploy. We externally validated these models on pan-cancer cohorts for a reason. A model tuned on one institution's mix of cancers, drugs, and documentation habits can quietly fail somewhere else. Calibration drifts. Practice changes. The governance question isn't "is the model accurate" once; it's "how will we know when it stops being accurate here." That's not a reason to wait. It's the operating discipline that makes deploying responsible.

Where this goes

The near-term win is not autonomous at all. It's a quieter shift in standard of care: risk-stratified monitoring for everyone starting a checkpoint inhibitor, drawn from data we already collect and pay to store. The patient flagged as high-risk for pneumonitis gets close watch. The one flagged as low risk across the board gets a lighter touch and fewer false alarms, which is its own clinical and financial good. We move from catching immune-related toxicity after it announces itself to anticipating it, and we do it without ordering a single new test.

For oncologists, that means the toxicity conversation at the start of therapy gets specific instead of generic. For informaticists, it means the work is mostly plumbing and labels, not model architecture. For payers, it means a concrete place where prediction reduces avoidable admissions and steroid courses rather than just generating alerts. The science to do this is largely done and published. What's left is the unglamorous part: wiring it into the day, agreeing on the action, and watching the model the way we'd watch any new diagnostic. That's the work worth doing, because the signal was never the thing we were missing. The willingness to read it was.

Essays — Travis Osterman | AI, Oncology & Health Data

2026-06-21T05:19:05.000Z

Essays

Long-form writing on AI in oncology, clinical informatics, cancer data standards, and what it takes to lead AI inside a health system.

2026-07-12

Bringing genomics out of the PDF and into the chart
A tumor sequencing result that lives as a faxed PDF can't trigger an alert, match a trial, or warn the next oncologist, and most of them still do.
2026-07-05

The oncologist's case for clinical informatics as a discipline
If you've never written a note in the system you're trying to fix, you will optimize the wrong thing.
2026-06-28

Predicting immunotherapy toxicity before it happens
The data to flag who will develop immune-related toxicity is already in the chart; we just haven't been reading it for that.
2026-06-21

The academic-industry AI partnership that actually shipped
Most academic-industry AI collaborations end in a data dump and a dead model; the ones that produce deployable work make a few structural choices on purpose.

The academic-industry AI partnership that actually shipped

2026-06-21T00:00:00.000Z

In 2019 my team at Vanderbilt University Medical Center and a group from GE HealthCare set out to predict who would benefit from immunotherapy and who would be harmed by it. Five years later we had a body of peer-reviewed work, filed international patents, and a model trained on routinely collected clinical data, not a data dump rotting on a shared drive. I want to explain why, because the difference was structural, not luck.

I have sat on both sides of these deals. I have watched well-funded academic-industry AI collaborations produce a press release, a single static data transfer, and then nothing: no deployable model, no usable IP, no publication anyone cites. The failure mode is so common that my colleagues and I eventually wrote it up and presented it at ASCO: Overcoming barriers in academic-industry partnerships to improve predictive modeling in immuno-oncology (ASCO Annual Meeting, 2022). The short version is that the standard contract structure is built to fail, and a few deliberate choices are what separate the collaborations that ship from the ones that stall.

The default structure is built to fail

The conventional academic-industry data deal looks tidy on paper. The academic medical center signs a contract, exports a de-identified dataset, and hands it to the company's data scientists. The two sides then work in isolation. This is the "data dump," and it fails in four predictable ways.

First, the data is static. Once it's transferred, it can't be refined, so the inevitable discrepancies (a lab value coded three different ways, a date that means admission in one feed and diagnosis in another) never get resolved. Second, the model picks up artifacts that nobody flags, because the people who know the clinic aren't in the room when the outputs are reviewed. Third, when someone finally wants to run the model on a real patient, they discover it depends on manually curated fields that don't exist in the live record, so it can't be deployed in the clinic it was built for. Fourth, and most corrosive over time: the clinicians aren't involved, so the academic partner gets nothing back except a check, and the relationship has no reason to continue.

Every one of those failures comes from the same root cause: the two sides act in isolation. Fix the isolation, and you fix most of the rest.

What we did instead

The GE HealthCare Digital Precision Oncology collaboration ran from 2019 to 2024, with me as principal investigator on the Vanderbilt Health side and Jan Wolber leading on GE's. The teams were genuinely distributed (clinicians and curators in the United States, modelers in Hungary and Germany), which made the discipline below non-optional rather than aspirational. Here is the recipe we actually used, the same one we laid out at ASCO in 2022:

Iterate on the data, don't dump it. Our clinical and curation experts met with the industry modelers to dynamically refine the de-identified datasets, reconciling structured data with manually curated data and resolving discrepancies rather than training on top of them.
Put clinicians in front of the model outputs. Practicing oncologists reviewed predictions specifically to catch artifacts and steer the final models. A radiologic or temporal artifact that looks like a signal to a modeler is often obvious to someone who has treated the disease.
Automate the curation you'll need at deployment. Our curators worked with in-house machine-learning people to build algorithms that automatically extracted natural-language elements from the EHR, so the model didn't quietly depend on hand-curation that would vanish the moment it left the lab.
Keep everyone in the same recurring meeting. Clinical and industry stakeholders met regularly with the modelers to keep the work pointed at something clinically and commercially useful, not just statistically tidy.

None of that is exotic. It is mostly a refusal to let the two organizations drift into separate workstreams that only meet at contract milestones.

The deployability constraint that made it real

The single most important decision we made was about the data the model was allowed to consume. We built against routinely collected EHR data, the stuff that already exists for every patient as a byproduct of care, rather than a bespoke research cohort with hand-annotated features. That choice is annoying. Routine data is messier, and you give up some apparent accuracy. But it is the only choice that produces a model you can actually run in clinic, because a model that needs a pristine research cohort can only ever be demonstrated, never deployed.

The flagship result, Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data (Lippenszky et al., JCO Clinical Cancer Informatics, 2024), showed it works. Using routine EHR data from more than 2,200 patients treated with immune checkpoint inhibitors, we built models that predicted pneumonitis, hepatitis, colitis, and one-year overall survival, with AUCs in the mid-0.70s (roughly 0.73 to 0.76). Crucially, the models required no additional data collection or documentation in the clinic. That last clause is the whole point. The model fits the workflow that already exists.

It was also deliberately pan-cancer. The cohort spanned melanoma, lung, genitourinary, and other tumor types rather than chasing a single high-prevalence disease, because immunotoxicity is a cross-cutting problem and a model that only works in one cancer is a narrower asset for both partners.

Why the IP and the publications both happened

The structure produced two kinds of durable output at once, and that is not a coincidence. Because clinicians were embedded in the modeling loop, the work was publishable: multiple peer-reviewed manuscripts came out of it, not just an internal deck. And because the curation and model-generation methods were built deliberately rather than improvised, they were patentable. The resulting U.S. patent on the model-generation framework names the inventors from both institutions together. Academic credit and commercial IP are usually treated as a zero-sum trade. They aren't, if the work is structured so that the same activity generates both.

That joint inventorship is also the answer to the fourth failure mode above. The academic partner got real intellectual output, the industry partner got deployable methods and IP, and both had a reason to keep going. The collaboration ended on schedule in 2024, but the line of research did not stop with it. The methods and the questions it raised continue to shape how we bring predictive analytics into routine oncology decisions. Partnerships that ship tend to leave that kind of residue.

What to actually require if you're funding one of these

If you run strategy at a health system or at a company doing clinical AI, do not sign the data-transfer-only version of this deal. Write the structure into the agreement instead. Require recurring joint working sessions, not milestone hand-offs. Require that clinicians review model outputs as a deliverable, not a courtesy. Require that the model be built against data that exists in the live record, and treat any dependence on hand-curated features as a deployment defect to be fixed before launch. And settle the publication and IP terms up front so that academic credit and commercial ownership are designed to coexist rather than compete.

The full Digital Precision Oncology collaboration is documented as a case study, and I've written more elsewhere about why the data layer underneath these models matters more than the models themselves. But the partnership lesson stands on its own. The collaborations that produce deployable models and real IP are not the ones with the biggest budgets or the cleanest data. They are the ones where the two sides refuse to work in isolation, and build that refusal into the contract.

Collaborations — Travis Osterman | Epic, Microsoft, GE, NCCN

2026-06-20T22:58:54.000Z

Corporate Collaborations

Dr. Osterman maintains research, advisory, and standards-development collaborations with the partners below. Engagements are tracked through Vanderbilt Health's conflict-of-interest process and the CMS Open Payments database, both linked publicly below. Consulting, advising, and speaking engagements are managed through WSOR, LLC.

GE HealthCare
2019-2024

Five-year strategic research collaboration on predicting immune checkpoint inhibitor efficacy and toxicity from real-world EHR data. Dr. Osterman served as principal investigator on the flagship Digital Precision Oncology study.

3 publications · 1 talks · 1 patent · 1 press
Epic
ongoing

Epic is the EHR platform anchoring Dr. Osterman's clinical informatics work. Multiple collaboration surfaces: the Adult Oncology Steering Committee, the Beacon Community Operations Group (founder), Epic Genomics Module, Epic AURA, and the AI-extracted oncology staging workflow in Epic Hyperspace.

1 publications · 4 talks · 2 press
Microsoft
2020-present

Microsoft Investigator Fellow (2020) - a two-year, $200,000 award for scaling clinical informatics education across UME and GME via Azure-based virtualization. Continuing service on the Microsoft Azure Research Community Advisory Board on structured genomic data, FHIR research infrastructure, and the mCODE data standard.

1 publications · 0 talks · 2 press
Tempus AI
ongoing

Multi-year partnership with Tempus AI spanning three workstreams: structured electronic integration of genomic results into clinical workflows, digital pathology, and multi-omic discovery infrastructure.

0 publications · 1 talks
nference
2023-present

Vanderbilt Health's partnership with nference reduces the friction of accessing clinical data for research while preserving privacy and regulatory compliance. Formalized in a 2023 agreement.

1 publications · 4 talks
NCCN
ongoing

Member of the NCCN Digital Oncology Forum and an active voice across NCCN policy and clinical channels. Featured AI keynote speaker at the 2026 NCCN Annual Conference and the NCCN AI policy summit.

5 publications · 9 talks

Disclosures

These collaborations are publicly disclosed through two independent records. The CMS Open Payments record is required of every U.S. physician under the Physician Payments Sunshine Act; the ASCO COI record is voluntary.

ASCO COI disclosure - coi.asco.org/share/JAR-DHX7/Travis Osterman
CMS Open Payments (Sunshine Act) - openpaymentsdata.cms.gov/physician/1291957

Inquiries

Consulting, advising, and speaking engagements are managed through WSOR, LLC. The WSOR site is the right intake for new engagements.

Predicting immunotherapy efficacy and toxicity from real-world data

2026-06-20T22:58:54.000Z

Digital Precision Oncology with GE HealthCare

A five-year strategic research collaboration on predicting immune checkpoint inhibitor effectiveness and toxicities from real-world electronic-health-record data. Travis Osterman served as principal investigator on the flagship study; the program produced three peer-reviewed manuscripts, multiple abstracts, and a patent.

The clinical problem

Immune checkpoint inhibitors (ICIs) have transformed cancer care. They allow subgroups of patients with previously incurable disease to have longer, higher-quality lives. They also carry the risk of severe immune-related adverse events: hepatitis, colitis, and pneumonitis, some of which are life-threatening and any of which forces discontinuation of an otherwise-working therapy. Oncologists make the treatment decision without a clean way to predict who will respond and who will be harmed. A personalized risk-benefit profile - generated before the first dose - would improve safety, extend treatment duration for responders, and improve clinical trial cohort selection.

The partnership

In 2019 Vanderbilt and GE HealthCare announced a five-year strategic research collaboration centered on this problem (dotmed.com, January 2019). The Digital Precision Oncology study was the flagship: Dr. Osterman as principal investigator from Vanderbilt; Jan Wolber leading on the GE HealthCare side; multi-disciplinary teams on both sides spanning medical oncology, radiochemistry, biostatistics, bioinformatics, and machine learning. The collaboration ran through 2024.

The approach

The team built a machine-learning framework using real-world clinical data already living in the Vanderbilt electronic health record: baseline labs, comorbidities, prior treatments, demographics, and imaging. The cohort grew over the course of the program; the JCO CCI 2024 flagship paper drew on more than 2,200 patients treated with ICIs. Models predicted both efficacy (one-year overall survival) and three major immunotoxicities (hepatitis, colitis, pneumonitis) using only pre-treatment data, generating per-patient risk-benefit profiles. The technical framing was deliberately conservative: short medical-history windows, careful validation, clinically interpretable features over black-box embeddings.

The published results

The flagship paper, Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data (Lippenszky et al., JCO Clinical Cancer Informatics, 2024), demonstrated that routinely-collected EHR data is enough to build clinically useful prediction models - across both efficacy and toxicity, without bespoke biomarker panels or unstructured-data extraction pipelines.

From Dr. Osterman's talk Prediction of immune checkpoint inhibitor outcome and side effects by use of machine learning and routine real-world data, ESMO Immuno-Oncology Annual Congress, Geneva, December 2024. US training cohort (~2,200 patients) and German external evaluation cohort (~4,250 patients); AUC scores at 100 days / 1 year / 3 years for overall survival and at 6 weeks / 90 days / 1 year for hepatitis. Models retained a substantial fraction of training-cohort performance on the external cohort - the validation result that made the approach credible beyond a single institution.

An imaging companion paper followed in 2025. Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography (Smith et al., JCO CCI, 2025) extended the framework to imaging: 2,700 pre-treatment thorax CT volumes for hundreds of patients, with radiomic features and deep-learning models predicting which patients would later develop ICI-induced pneumonitis. External validation across multi-center pan-cancer cohorts was reported at SITC in 2023 (Kiss, Lippenszky et al., abstract 1294).

The patent Model generation apparatus for therapeutic prediction and associated methods and models (Wolber, Csernai, Kiss, Lippenszky, Horváth, Osterman, Park, Smith, Fabbri, LeNoue-Newton, Mittendorf) captures the underlying framework.

Total output

4 peer-reviewed manuscripts
8 published abstracts
12 conference presentations across ASCO, SITC, and ASCO GI
1 patent on the model-generation framework

Beyond the published record, the work was covered by ASCO Daily News in two pieces and a podcast, and by Precision Medicine Online in November 2023.

What comes next

The collaboration ended in 2024 as planned, but the line of research continues. Models for ICI-induced hepatitis and colitis are in active development. The clinical insight - that routinely collected EHR data carries enough signal to generate a meaningful risk-benefit profile before the first dose - is now part of a broader research program at Vanderbilt-Ingram on bringing predictive analytics into routine oncology decision-making.

The lesson

The Digital Precision Oncology program is what happens when a careful clinical question gets paired with real-world EHR data, conservative machine-learning methodology, and multi-year institutional commitment. No one paper changed practice. The cumulative work made it credible that pre-treatment risk-benefit prediction for immunotherapy is achievable from the data that already exists in every cancer center's EHR. That credibility is what's required before any of this lands at the bedside.

Cited works

Lippenszky L, Mittendorf KF, Kiss Z, LeNoue-Newton ML, Napan-Molina P, Rahman P, Ye C, Laczi B, Csernai E, Jain NM, Holt ME, Maxwell CN, Ball M, Ma Y, Mitchell MB, Johnson DB, Smith DS, Park BH, Micheel CM, Fabbri D, Wolber J, Osterman TJ. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clinical Cancer Informatics 2024.
Smith DS, Lippenszky L, LeNoue-Newton ML, Jain NM, Mittendorf KF, Micheel CM, Cella PA, Wolber J, Osterman TJ. Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography. JCO Clinical Cancer Informatics 2025.
Kiss Z, Lippenszky L, Laczi B, Napan-Molina P, Csernai E, Brehmer A, Kim M, Keyl J, Siveke J, Meyer M, Grünwald V, Kasper S, Roesch A, Schuler M, Osterman T, Wolber J, Kleesiek J. External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. SITC 2023, abstract 1294.
Wolber J, Csernai EK, Kiss Z, Lippenszky L, Horváth G, Osterman T, Park BH, Smith DS, Fabbri D, LeNoue-Newton ML, Mittendorf K. Model generation apparatus for therapeutic prediction and associated methods and models. Patent.

Press Kit — Travis Osterman | Bios, Headshots, Coverage

2026-06-19T03:42:11.000Z

Press

Press kit and press coverage for journalists, event organizers, and researchers. Bios and headshots are cleared for verbatim use; the coverage list is auto-updated from the same source-of-truth library that drives the rest of the site.

Bios

Short ~75 words

Dr. Travis Osterman is a practicing medical oncologist and informatician at Vanderbilt University Medical Center, where he serves as Associate Vice President for Research Informatics. He chairs the minimal Common Oncology Data Elements (mCODE™) Executive Committee, an international cancer data standard implemented at more than 70 institutions across six countries and the only method of submitting data to CMS' Enhancing Oncology Model. He is board certified in medical oncology and clinical informatics.

Medium ~180 words

Dr. Travis Osterman is a practicing medical oncologist and informatician, Associate Vice President for Research Informatics at Vanderbilt University Medical Center, and Director of Cancer Clinical Informatics at the Vanderbilt-Ingram Cancer Center. He is board certified in medical oncology and clinical informatics.

At Vanderbilt, Dr. Osterman leads the Clinical Genomics Workstream and has led the effort to make structured genomic data more accessible for patient care and research; under his leadership, Vanderbilt's electronic health record contains more structured genomic data than any other institution in the United States.

Nationally, Dr. Osterman chairs the minimal Common Oncology Data Elements (mCODE™) Executive Committee. mCODE is implemented at more than 70 institutions across six countries and serves as the only method of submitting data to CMS' Enhancing Oncology Model. He advises Microsoft, Epic, Tempus AI, and the National Comprehensive Cancer Network, and founded the Epic Beacon Community Operations Group.

Long see /about/

The full bio, including origin and career arc, lives at /about/. For a downloadable version, request via the contact form.

Headshots

All headshots are cleared for editorial use. Right-click and "Save image as" to download, or use the direct links.

Square portrait
800 × 800 JPG
Download (800px) · Download (400px)
Social card
1200 × 630 JPG
Download

Key facts

Name & post-nominals: Travis Osterman, DO, MS, FAMIA, FASCO
Roles: Associate Vice President for Research Informatics and Associate Chief Medical Information Officer, Vanderbilt University Medical Center; Ingram Associate Professor of Cancer Research; Director of Cancer Clinical Informatics, Vanderbilt-Ingram Cancer Center; Chair, mCODE Executive Committee
Board certifications: Medical Oncology (ABIM, 2016); Clinical Informatics (ABPM, 2016)
Fellowships: American Society of Clinical Oncology (FASCO, 2023); American Medical Informatics Association (FAMIA, 2019)
mCODE scale: more than 70 institutions across six countries; only method of submitting data to CMS' Enhancing Oncology Model
Vanderbilt genomic data: more structured genomic data in the electronic health record than any other institution in the United States
Disclosures: ASCO COI; CMS Open Payments

Topics most frequently covered

The mCODE oncology data standard and its role in CMS' Enhancing Oncology Model
Structured genomic data integration into the electronic health record
Artificial intelligence in oncology - efficacy and toxicity prediction
Interoperability and EHR governance for cancer care
Clinical informatics education for medical students and residents
Lung cancer screening and real-world data

Now

Recent activity across publications, invited talks, scientific meetings, and press - auto-generated from the same source-of-truth library that drives /research/ and /speaking/. Last updated: 2026-07-12.

Recent activity (last 12 months)

Invited talk HL7 International Working Group Meeting (Europe): "mCODE Overview and Current State". May 18, 2026
Invited talk Vanderbilt Lecture Series CME (Nashville, Tennessee): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". May 7, 2026
Abstract Abstract LB385: Polygenic risk score of genetic variants in genes encoding drug-metabolizing enzymes and drug transporters, in association with febrile neutropenia. Cancer Research. Apr 17, 2026
Invited talk 2026 NCCN Annual Conference (Orlando, Florida): "Harnessing Artificial Intelligence to Improve Oncology Care". Mar 28, 2026
Invited talk NASEM Workshop: Policy Issues for Integrating Artificial Intelligence in Cancer Research and Care (Washington, DC): "AI in Cancer Care: 2 Wins, 2 Current Challenges". Mar 9, 2026
Invited talk Jackson-Madison County General Hospital (Jackson, Tennessee): "An Update on the Use of AI in Clinical Practice". Mar 6, 2026
Invited talk Blanchfield Army Community Hospital (Fort Campbell, Kentucky (virtual)): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". Feb 10, 2026
Peer-reviewed publication SmokeBERT and Beyond: Bridging Clinical Narratives and Structured Smoking Data To Improve Lung Cancer Screening. JCO clinical cancer informatics. Dec 22, 2025
Invited talk "Leveraging Data to Day-to-Day Improvements". Oct 31, 2025
Peer-reviewed publication Clonal Hematopoiesis of Indeterminate Potential After Radiation Therapy. International Journal of Radiation Oncology*Biology*Physics. Oct 24, 2025
Peer-reviewed publication Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine. Oct 15, 2025
Patent Model generation apparatus for therapeutic prediction and associated methods and models. Sep 18, 2025
Scientific meeting presentation The Evolving Artificial Intelligence Landscape in Cancer Care: "AI and the Cancer Journey: Navigating New Frontiers in Policy and Technology". Sep 9, 2025
Peer-reviewed publication Artificial intelligence across the cancer care continuum. Cancer. Aug 15, 2025
Peer-reviewed publication Advancing the science of genomic learning healthcare systems. Learning Health Systems. Jul 23, 2025

In the news

External coverage by independent publications. 24 items. Last updated: 2026-07-12.

Press contact

Interview requests, quotes, and other press inquiries: please use the contact form with subject line beginning "Press". For event organizers and speaking inquiries, see /speaking/ and WSOR, LLC.

Notes — Travis Osterman | Oncology AI & Informatics

2026-06-19T03:09:55.000Z

Notes

Short, timely commentary — reactions to new research, policy, product launches, and what they mean for oncology and clinical AI.

Notes are on the way. Follow along via the RSS feed.

Speaking — Travis Osterman | AI in Oncology Keynotes

2026-06-16T14:53:17.000Z

Speaking

Where Dr. Osterman speaks: international standards bodies, national policy summits, health-system leadership audiences, and industry advisory boards. 48 invited talks and 32 scientific-meeting presentations. Last updated: 2026-07-12.

Invited presentations

2026

HL7 International Working Group Meeting (Europe): "mCODE Overview and Current State". May 18, 2026
Vanderbilt Lecture Series CME (Nashville, Tennessee): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". May 7, 2026
2026 NCCN Annual Conference (Orlando, Florida): "Harnessing Artificial Intelligence to Improve Oncology Care". Mar 28, 2026
NASEM Workshop: Policy Issues for Integrating Artificial Intelligence in Cancer Research and Care (Washington, DC): "AI in Cancer Care: 2 Wins, 2 Current Challenges". Mar 9, 2026
Jackson-Madison County General Hospital (Jackson, Tennessee): "An Update on the Use of AI in Clinical Practice". Mar 6, 2026
Blanchfield Army Community Hospital (Fort Campbell, Kentucky (virtual)): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". Feb 10, 2026

2025

"Leveraging Data to Day-to-Day Improvements". Oct 31, 2025
Tennessee Osteopathic Medical Association Annual Meeting (Franklin, TN): "An Update on Precision Oncology: What You Need to Know about Genomic Therapies". May 2, 2025

2024

MITRE Annual Strategic Meeting: "Opportunities moving forward". Dec 2, 2024
NCCN EHR Advisory Board (Virtual): "Clinical Trial Enrollment A Pragmatic Approach". Nov 8, 2024
University of Hawaii, Artificial Intelligence, Precision Health Institute (Virtual): "Predictive AI Models - Data Standards in Action". May 10, 2024
NHGRI: "Defining a Clinical Data Ecosystem for Genomic Health: Real World Genomic Data in Clinical Care". Apr 10, 2024

2023

VICC Board of Advisors (Nashville, TN): "Implementing A Precision Oncology Program". Dec 13, 2023
Zebra Healthcare CAB (Nashville, TN): "The Potential of Precision Medicine To Guide Clinical Decision-Making". Nov 30, 2023
Alzheimer’s Association, Diversity and Disparities PIA: "mCODE Update®". Nov 1, 2023
Boston, MA: "The Potential of Precision Medicine To Guide Clinical Decision-Making". Oct 24, 2023
(Nashville, TN) "From Data Standards to Discovery: Making Clinical Data Available for Research". Sep 19, 2023
NCCN EHR Oncology Advisory Group (Pymouth, PA): "Digital Transformation". Jul 27, 2023
NCCN EHR Advisory Board (Virtual): "mCODE® Update". Jul 19, 2023
Indiana University / Regenstrief (Indianapolis, IN): "Standards in Action: Improving Interoperability and Data Access". May 18, 2023
Incorporating Integrated Diagnostics into Precision Oncology Care (Washington, DC): "Ensuring Integrated Diagnostics Facilitate Oncology Care". Mar 6, 2023
NCCN EHR Advisory Group (Virtual): "Leveraging the EHR to facilitate actionable research data". Jan 20, 2023

3+ years

Tennessee Oncology Data Analysts Association (Nashville, TN): "Advancing Lung Cancer Treatment in the Era of Precision Oncology". Oct 7, 2022
Clinical Genomics Update: Delivering on the the VUMC Mission (Virtual): "Clinical Genomics Update: Delivering on the the VUMC Mission". Sep 20, 2022
NHGRI Genomic Medicine XIV (Virtual): "Integrating Genomic Results into Electronic Health Records (EHRs)". Aug 31, 2022
Beacon Community Operations Group (BCOG), (Virtual): "Maximizing Value with Structured Genomic Data". Jul 8, 2022
Kaiser Permanente (Virtual): "Maximizing Value with Structured Genomic Data". Jun 14, 2022
Columbia University (Virtual): "Maximizing Value with Structured Genomic Data". Jun 8, 2022
mCODE® Community of Practice (Virtual): "Categorization of mCODE® via the FHIR Maturity Model". May 27, 2022
25. Tempus Webinar (Virtual): "Integrating Structured Genomic Data in Clinic". May 12, 2022
26. AMIA Genomics Working Group (Virtual): "Maximizing the Value of Structured Genomic Data in the EHR". May 5, 2022
27. University of Washington (Virtual): "e-Consent at VUMC". Apr 27, 2022
Tennessee Medical Association (Virtual): "Clinical Genomics: Practical Applications for Patient Care". Apr 21, 2022
Innovation in Electronic Health Records for Oncology Care Research and Surveillance (Washington, DC): Today’s Patient Portal and Sharing of Patient Data Across EHR Systems for Cancer Care and Research. Feb 28, 2022
Perspectives in Precision Oncology From Prevention to Treatment (Nashville, TN): "Leveraging Structured Genomic Data". Jan 17, 2022
(Nashville, TN) "Clinical Genomic ePMO: An Enterprise Update". Sep 29, 2021
VUMC, School of Medicine, Leadership Development Program Capstone Presentation, (Nashville, TN): "Genomic Data Strategy". Dec 11, 2020
"Making Data Available: An Overview of Data Opportunities at VICC and VUMC". Dec 10, 2020
Epic: Beacon Community Oncology Group (BCOG) (Virtual): "Auto-Populate Care Team with Encounter Provider". May 6, 2020
(Virtual) "Building Productive Relationships with For-Profit Organizations". Apr 22, 2020
(Virtual) "Document Link in eStar". Apr 10, 2020
VUMC DBMI Retreat (Nashville, TN): "Leading Change: Learning from every patient". Aug 13, 2019
ASCO Annual Meeting: "How to Navigate the Annual Meeting". Jun 2, 2018
ASCO Annual Meeting: "How to Navigate the Annual Meeting". Jun 3, 2017
ASCO Annual Meeting: "How to Integrate ASCO University Into Your Program". Jun 5, 2016
44. DBMI Research Forum (Nashville, TN): "EHR-Wide GxE Study Using Smoking Information Extracted From Clinical Notes". May 16, 2016
46. University of California San Diego, Division of Biomedical Informatics (San Diego, CA): "Extracting and Studying Granular Smoking History from the Electronic Health Record". Mar 8, 2016
The Indiana University Melvin and Bren Simon Cancer Center (Indianapolis, IN): "Deep Phenotyping in Oncology". Jan 15, 2016

Scientific meeting presentations

2025

The Evolving Artificial Intelligence Landscape in Cancer Care: "AI and the Cancer Journey: Navigating New Frontiers in Policy and Technology". Sep 9, 2025
NCCN Annual Congress (Virtual): "Improving Clinical Trial Accrual: Doing More with Less". Jun 15, 2025
NICT (Durhman, NC): "Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data". Mar 1, 2025
ASPIRE (Durhman, NC): "Growing an International Data Standard". Feb 28, 2025

2024

ESMO Immuno-Oncology Annual Congress (Geneva, Switzerland): "Prediction of immune checkpoint inhibitor outcome and side effects by use of machine learning and routine real-world data". Dec 11, 2024

2023

Society for Immunotherapy of Cancer Annual Meeting (San Diego, CA): "EHR-based Models for Predicting Efficacy and Toxicities Prior to ICI Treatment". Nov 4, 2023
ASCO Annual Meeting (Chicago, IL): "Interoperability in Action: Progress in Implementing the mCODE™ Oncology Data Standard". Jun 3, 2023
ASCO Annual Meeting (Chicago, IL): "Standardizing Data Workgroup". Jun 2, 2023
IO360 (Brooklyn, NY): "Predicting Efficacy and Toxicities Prior to Immune Checkpoint Inhibitor Treatment". Feb 9, 2023

3+ years

AMIA Annual Symposium (Washington, DC): "mCODE® (minimal Common Oncology Data Elements): a 3-year Update". Nov 8, 2022
AMIA Annual Symposium (Washington, DC): "A Unified Approach to CI Education for UME and GME". Nov 8, 2022
AMIA Annual Symposium (Washington, DC): "Spicing Up Your Clinical Informatics Curriculum: Incorporating Interactive Learning Activities". Nov 7, 2022
Epic User Group meeting (UGM) (Verona, WI): "Precision Medicine: Using Structured Genomic Data in Clinical Decision Support". Aug 21, 2022
ASCO Annual Meeting (Chicago, IL): "Unlocking the Promise of Data-Driven Medicine in Cancer Care, Together. Translating complex multimodal data into actional insights". Jun 6, 2022
Epic eXept Group Meeting (XGM (Verona, WI): "ONC19: Treatment Location Preference for Multi-Site Infusion Clinics". May 3, 2022
NCCN Annual Conference: "The Future of Telemedicine in Oncology". Mar 31, 2022
American College of Medical Genetics Annual Clinical Genetics Meeting: Data to Knowledge to Wisdom: Improving Care Through Creative Use of Information Systems. Mar 23, 2022
NCCN Annual Conference (Virtual): "The Future of Telemedicine in Oncology". Mar 21, 2022
Cornell University, Cloud Forum 2021 (Virtual): "Lessons Learned from VUMC’s First End-to-End Cloud Project". Nov 17, 2021
Epic User Group Meeting (UGM) (Verona, WI): "Supporting Each Other through Operations Community Groups: A Tale of Six Academic Cancer Centers". Aug 25, 2021
American Association of Cancer Research (AACR) Project GENIE Virtual Summit (Virtual): "Connecting Genomic Laboratories to the Local EHR". Jul 15, 2021
Epic eXpert Group Meeting (XGM) (Virtual): "ONC28: Auto-Populate Care Team with Encounter Provider". May 2020
Epic User Group Meeting (UGM): "Improving In Basket User Experience: Leveraging Governance, Build, Development, and Collaboration". Aug 28, 2019
NCCN EHR Workgroup (Plymouth, PA): "ASCO’s mCODE™ Initiative: Potential Impact from Member Institution Perspective". Jul 18, 2019
ASCO Annual Meeting (Chicago, IL): "mCODE™ (Minimal Common Oncology Data Elements) in Cancer Practice". Jun 3, 2019
AMIA Joint Summit: "Extracting Tobacco Exposure with the Smoking History and Pack-Year Extraction System (SHAPES)". Mar 13, 2018
Conquer Cancer Foundation Scientific and Career Development Retreat (Washington, DC): "Smoking History and Pack Year Extraction System (SHAPES): Supporting Lung Cancer Screening and Tobacco-related Research". Oct 11, 2017
NIH MCL Steering Committee Meeting: "MCL (lung) Data Model Overview and Demonstration". Sep 26, 2017
28. NLM Informatics Training Conference (Columbus, OH): "EHR-Wide GxE Study using Smoking Information Extracted from Clinical Notes,”". Jun 29, 2016
ASCO Annual Meeting (Chicago, IL): "Cooperative Groups and Clinical Trial Design—Past, Present, and Future". Jun 3, 2016
ASCO Annual Meeting (Chicago, IL): "Optimize Your Annual Meeting Experience". Jun 3, 2016
AMIA Annual Symposium (San Francisco, CA): "Quantifying Tobacco Exposure Using Clinical Notes and Natural Language Processing to Enable Lung Cancer Screening". Nov 18, 2015

Research — Travis Osterman | Publications, Patents, Grants

2026-06-16T14:53:17.000Z

Research

47 peer-reviewed publications, 28 abstracts, and 1 patent. Last updated: 2026-07-12.

Case studies

Detailed accounts of programs Dr. Osterman has led - what the problem was, what the team did, what shipped, and what comes next. Each case study cites the underlying peer-reviewed work.

mCODE
flagship

From an open-source oncology data standard to the only method of submitting data to a federal value-based care program. The story of mCODE's launch, governance, and regulatory uptake - and the parallel story of bringing structured genomic data into the EHR at scale.

Promise of mCODE (JCO CCI 2020) · mCODEGPT (Comm Med 2025)
Digital Precision Oncology
2019-2024

Five-year strategic research collaboration with GE HealthCare on predicting immune checkpoint inhibitor effectiveness and toxicities from real-world EHR data. PI on the flagship study; four peer-reviewed manuscripts and a patent.

Lippenszky et al. (JCO CCI 2024) · Smith et al. (JCO CCI 2025)
AI in oncology
ongoing

From one of the first peer-reviewed evaluations of ChatGPT in clinical Q&A to a comprehensive review of AI across the cancer care continuum. How clinical AI gets validated, where it actually helps, and why it depends on structured data underneath.

On the cusp (Med 2023) · AI across the cancer care continuum (Cancer 2025)

Co-author network

A force-directed graph of co-authors with whom Dr. Osterman has published at least two peer-reviewed papers or abstracts. Node size scales with joint-publication count; edges connect collaborators who have also co-authored with each other. Hover any node for the name and count. The graph is built from the same Zotero corpus that drives the publication and abstract lists below.

Visible labels are the fifteen most-frequent co-authors; the rest are visible on hover. Browsers with JS auto-replace this with an interactive canvas (drag, zoom, click-to-focus).

Peer-reviewed publications

2025

Heng Tan, Travis J. Osterman. SmokeBERT and Beyond: Bridging Clinical Narratives and Structured Smoking Data To Improve Lung Cancer Screening. JCO clinical cancer informatics. doi:10.1200/CCI-25-00350 Dec 22, 2025
Shelby A. Crants et al. Clonal Hematopoiesis of Indeterminate Potential After Radiation Therapy. International Journal of Radiation Oncology*Biology*Physics. doi:10.1016/j.ijrobp.2025.10.006 Oct 24, 2025
Kai Zhang, Tongtong Huang, Bradley A. Malin, Travis Osterman, Qi Long, Xiaoqian Jiang. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine. doi:10.1038/s43856-025-01116-x Oct 15, 2025
Irbaz Bin Riaz, Muhammad Ali Khan, Travis J. Osterman. Artificial intelligence across the cancer care continuum. Cancer. doi:10.1002/cncr.70050 Aug 15, 2025
Teri A. Manolio et al. Advancing the science of genomic learning healthcare systems. Learning Health Systems. doi:10.1002/lrh2.70027 Jul 23, 2025
Elise Russo et al. Vanderbilt Clinical Informatics Center Education Strategy: To Infinity and Beyond!. Applied Clinical Informatics. doi:10.1055/a-2443-8318 Feb 26, 2025
David S. Smith et al. Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography. JCO Clinical Cancer Informatics. doi:10.1200/CCI-24-00198 Feb 20, 2025

2024

Yanwei Li et al. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics. doi:10.1200/CCI.23.00249 Jun 28, 2024
Travis J. Osterman, Jiarong Ye. The importance of studying the implementation of cancer data standards. Cancer. doi:10.1002/cncr.35441 Jun 14, 2024
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology. doi:10.1200/JCO.2024.42.16_suppl.e13649 May 29, 2024
Engineering National Academies of Sciences. Incorporating Integrated Diagnostics into Precision Oncology Care: Proceedings of a Workshop. . doi:10.17226/27744 Apr 17, 2024
Levente Lippenszky et al. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clinical Cancer Informatics. doi:10.1200/CCI.23.00207 Mar 21, 2024

2023

Waddah Arafat et al. Clinician Perspectives Regarding the Impact of Information Technology on Multidisciplinary Tumor Boards: A National Comprehensive Cancer Network Survey. JCO Clinical Cancer Informatics. doi:10.1200/CCI.23.00056 Nov 9, 2023
Megan M Shuey et al. Next-generation phenotyping: introducing phecodeX for enhanced discovery research in medical phenomics. Bioinformatics. doi:10.1093/bioinformatics/btad655 Nov 1, 2023
Rachel S. Goodman et al. Accuracy and Reliability of Chatbot Responses to Physician Questions. JAMA Network Open. doi:10.1001/jamanetworkopen.2023.36483 Oct 2, 2023
Eric M. Lander et al. Identification and Characterization of Avoidable Hospital Admissions in Patients With Lung Cancer. Journal of the National Comprehensive Cancer Network. doi:10.6004/jnccn.2023.7049 Oct 1, 2023
Travis J. Osterman, James C. Yao, Monika K. Krzyzanowska. Implementing Innovation: Informatics-Based Technologies to Improve Care Delivery and Clinical Research. American Society of Clinical Oncology Educational Book. doi:10.1200/EDBK_389880 May 22, 2023
Protiva Rahman et al. Accelerated curation of checkpoint inhibitor-induced colitis cases from electronic health records. JAMIA Open. doi:10.1093/jamiaopen/ooad017 Apr 1, 2023
Rachel S. Goodman, J. Randall Patrinely, Travis Osterman, Lee Wheless, Douglas B. Johnson. On the cusp: Considering the impact of artificial intelligence language models in healthcare. Med (New York, N.Y.). doi:10.1016/j.medj.2023.02.008 Mar 10, 2023
Douglas Johnson et al. Assessing the Accuracy and Reliability of AI-Generated Medical Responses: An Evaluation of the Chat-GPT Model (under review). . doi:10.21203/rs.3.rs-2566942/v1 Feb 28, 2023

3+ years

Dara E. Mize, Travis J. Osterman. A Unified Approach to Clinical Informatics Education for Undergraduate and Graduate Medical Education. AMIA ... Annual Symposium proceedings. AMIA Symposium (2022).
Emily Pei-Ying Lin et al. Associations of influenza vaccination with severity of immune-related adverse events in patients with advanced thoracic cancers on immune checkpoint inhibitors. ERJ open research (2022). doi:10.1183/23120541.00684-2021
Chloe Weidenbaum, Christopher G. Cann, Sarah Osmundson, Wade T. Iams, Travis Osterman. Two Uncomplicated Pregnancies on Alectinib in a Woman With Metastatic ALK-Rearranged NSCLC: A Case Report. JTO Clinical and Research Reports (2022). doi:10.1016/j.jtocrr.2022.100361
Lucy R. Langer, Amye Tevaarwerk, Robin Zon, Travis Osterman. The Future of Telemedicine in Oncology. Journal of the National Comprehensive Cancer Network (2022). doi:10.6004/jnccn.2022.5020
National Cancer Policy Forum, Board on Health Care Services, Computer Science and Telecommunications Board, Division on Engineering and Physical Sciences, Health and Medicine Division, National Academies of Sciences, Engineering, and Medicine. Innovation in Electronic Health Records for Oncology Care, Research, and Surveillance: Proceedings of a Workshop. (2022).
Peter D. Stetson et al. Adoption of Patient-Generated Health Data in Oncology: A Report From the NCCN EHR Oncology Advisory Group. Journal of the National Comprehensive Cancer Network (2022). doi:10.6004/jnccn.2021.7088
Amye J. Tevaarwerk et al. Oncologist Perspectives on Telemedicine for Patients With Cancer: A National Comprehensive Cancer Network Survey. JCO Oncology Practice (2021). doi:10.1200/OP.21.00195
Alex C. Cheng et al. Follow-up Interactive Long-Term Expert Ranking (FILTER): a crowdsourcing platform to adjudicate risk for survivorship care. JAMIA open (2021). doi:10.1093/jamiaopen/ooab090
Marilyn E. Holt et al. My Cancer Genome: Coevolution of Precision Oncology and a Molecular Oncology Knowledgebase. JCO Clinical Cancer Informatics (2021). doi:10.1200/CCI.21.00084
Neha M. Jain et al. Framework for Implementing and Tracking a Molecular Tumor Board at a National Cancer Institute–Designated Comprehensive Cancer Center. The Oncologist (2021). doi:10.1002/onco.13936
Kim L Sandler et al. Women screened for breast cancer are dying from lung cancer: An opportunity to improve lung cancer screening in a mammography population. Journal of Medical Screening (2021). doi:10.1177/09691413211013058
Travis J. Osterman, May Terry, Robert S. Miller. Reply to J. Chen et al. JCO Clinical Cancer Informatics (2021). doi:10.1200/CCI.21.00014
Neha M. Jain, Alison Culley, Christine M. Micheel, Travis J. Osterman, Mia A. Levy. Learnings From Precision Clinical Trial Matching for Oncology Patients Who Received NGS Testing. JCO Clinical Cancer Informatics (2021). doi:10.1200/CCI.20.00142
Neil S. Zheng et al. A retrospective approach to evaluating potential adverse outcomes associated with delay of procedures for cardiovascular and cancer-related diagnoses in the context of COVID-19. Journal of Biomedical Informatics (2021). doi:10.1016/j.jbi.2020.103657
Travis J. Osterman, May Terry, Robert S. Miller. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clinical Cancer Informatics (2020). doi:10.1200/CCI.20.00059
Xuanyi Li et al. Seven decades of chemotherapy clinical trials: a pan-cancer social network analysis. Scientific Reports (2020). doi:10.1038/s41598-020-73466-6
Jeremy Lyle Warner et al. Trends in FDA cancer registration trial design over time, 1969-2020.. Journal of Clinical Oncology (2020). doi:10.1200/JCO.2020.38.15_suppl.2060
Neha M. Jain, Alison Culley, Teresa Knoop, Christine Micheel, Travis Osterman, Mia Levy. Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow. JCO Clinical Cancer Informatics (2019). doi:10.1200/CCI.19.00033
Heidi D. Klepin et al. Hypertension and use of bevacizumab among patients treated in community settings.. Journal of Clinical Oncology (2019). doi:10.1200/JCO.2019.37.15_suppl.e18279
Neha M Jain, Alison Culley, Travis John Osterman, Mia Alyce Levy. Learnings from a pragmatic study to evaluate benefit of performing reflex clinical trial matching and providing clinical decision support to physicians.. Journal of Clinical Oncology (2019). doi:10.1200/JCO.2019.37.15_suppl.e18006
David A. Chambers et al. The Impact of Big Data Research on Practice, Policy, and Cancer Care. American Society of Clinical Oncology Educational Book (2019). doi:10.1200/EDBK_238057
Ragisha Gopalakrishnan et al. Impact of the influenza vaccination on cancer patients undergoing therapy with immune checkpoint inhibitors (ICI).. Journal of Clinical Oncology (2018). doi:10.1200/JCO.2018.36.15_suppl.3053
Lisa Bastarache et al. Phenotype risk scores identify patients with unrecognized Mendelian disease patterns. Science (2018). doi:10.1126/science.aal4043
Wei-Qi Wei et al. Evaluating phecodes, clinical classification software, and ICD-9-CM codes for phenome-wide association studies in the electronic health record. PloS One (2017). doi:10.1371/journal.pone.0175508
Mia Alyce Levy, Travis John Osterman, Neha Jain, Kathleen F Mittendorf, Christine Micheel. Utility of adding clinical data to a molecular results portal for improving clinical trial prescreening efficiency.. Journal of Clinical Oncology (2017). doi:10.1200/JCO.2017.35.15_suppl.e18182
Matthew J. Rioth, Travis J. Osterman, Jeremy L. Warner. Advances in website information resources to aid in clinical practice. American Society of Clinical Oncology educational book / ASCO. American Society of Clinical Oncology. Meeting (2015). doi:10.14694/EdBook_AM.2015.35.e608
Junyu Li et al. A software utility for creating interactive maps for 2D gel-based proteomics. Analytical Biochemistry (2004). doi:10.1016/j.ab.2004.03.058

Abstracts

2026

Sang Minh Nguyen et al. Abstract LB385: Polygenic risk score of genetic variants in genes encoding drug-metabolizing enzymes and drug transporters, in association with febrile neutropenia. Cancer Research. Apr 17, 2026

2025

Joseph Vento, Lisa Bastarache, Qingxia M. Chen, Travis Osterman. Real-world side effects of targeted therapies: High-throughput association studies leveraging the CancerLinq Discovery lung cancer database.. Journal of Clinical Oncology. May 28, 2025

2024

Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy | Journal for ImmunoTherapy of Cancer. . Nov 5, 2024
David Smith et al. 1246 Prediction of pneumonitis in immunotherapy patients from prior thorax CT. Journal for ImmunoTherapy of Cancer. Nov 1, 2024
Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy. Journal for ImmunoTherapy of Cancer. Nov 1, 2024
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology. Jun 2024

2023

Zoltan Kiss et al. 1294 External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. Journal for ImmunoTherapy of Cancer. Nov 1, 2023
Levente Lippenszky et al. 1300 Prediction of efficacy and toxicities of immune checkpoint inhibitors using real-world patient data. Journal for ImmunoTherapy of Cancer. Nov 1, 2023
Joseph Vento, Travis Osterman. BIO23-019: Precision Oncology: Integrating Structured Genomic Data Into the Electronic Health Record via the EPIC® Genomics Module. Journal of the National Comprehensive Cancer Network. Mar 31, 2023

3+ years

Waddah Arafat et al. Oncologist perspectives on tumor boards: Virtual meetings and EHR integration.. Journal of Clinical Oncology (2022).
Alex C. Cheng et al. Oncologist participation in pilot testing a crowdsourcing platform to build a survivorship care risk model.. Journal of Clinical Oncology (2022).
Eszter Csernai et al. Rolling window-based hepatitis toxicity prediction from routine bloodwork in patients undergoing immune checkpoint inhibitor therapy.. Journal of Clinical Oncology (2022).
Gergely Horváth et al. Predicting immune checkpoint inhibitor-related hepatitis using electronic health records of patients.. Journal of Clinical Oncology (2022).
Levente Lippenszky et al. Predicting immune checkpoint inhibitor-related pneumonitis using patient medical information.. Journal of Clinical Oncology (2022).
Eric Michael Lander et al. Characterization of avoidable hospital admissions in patients with lung cancer in the immunotherapy and targeted therapy era.. Journal of Clinical Oncology (2022).
Neha M Jain, Emma Schremp, Lucy Spalluto, Travis John Osterman, Debra L. Friedman. Incorporating mediation-based interventions at an academic cancer center: A six-step process.. Journal of Clinical Oncology (2022).
Neha M Jain, Philip Edward Lammers, Michael R. Savona, Travis John Osterman, Salil Goorha. Using a standard implementation science framework to improve clinical trial enrollment for a community Tennessee oncology center.. Journal of Clinical Oncology (2022).
Kathleen F. Mittendorf et al. Overcoming barriers in academic-industry partnerships to improve predictive modeling in immuno-oncology.. Journal of Clinical Oncology (2022).
Brian Yoon, Dilhan Weeraratne, Yull Edwin Arriaga, Hu Huang, Travis John Osterman. Evaluating health disparities in access to genomic testing for metastatic non-small cell lung cancer patients. | Journal of Clinical Oncology. (2021).
Amye Tevaarwerk et al. BIO21-011: Oncology Provider Perspectives on Telemedicine for Patients With Cancer: A National Comprehensive Cancer Network (NCCN®) Survey. Journal of the National Comprehensive Cancer Network (2021).
Taneya Y. Koonce et al. The Personalization of Evidence: Using Intelligent Datasets to Inform the Process. Amia Annual Symposium (2020).
Jeremy Lyle Warner et al. Trends in FDA cancer registration trial design over time, 1969-2020.. Journal of Clinical Oncology (2020).
Riqiang Gao et al. Lung cancer detection using co-learning from chest CT images and clinical demographics. (2019).
Carolyn C. Scott, Alexis B. Paulson, Travis J. Osterman, Kim L. Sandler. Lung and Breast Screening Practices in Women: Targeting Providers to Improve Enrollment in a Clinical Lung Screening Program. C30. LUNG CANCER SCREENING: WHO, WHY, WHERE, AND HOW MUCH (2017).
Travis J. Osterman. EHR-Wide GxE Study using Smoking Information Extracted from Clinical Notes. (2016).
Travis J. Osterman, Wei-Qi Wei, Dara Mize, Joshua C. Denny. Using a gene-environment interaction study to evaluate risk for lung cancer (abstract 1524). American Society of Clinical Oncology Annual Meeting (2016).
Travis J. Osterman, Wei-Qi Wei, Joshua C. Denny. Quantifying Tobacco Exposure using clinical Notes and Natural language processing to enable lung cancer screening. International Association for the Study of Lung Cancer Targeted Therapeutics Meeting (poster) (2016).
Travis J. Osterman. Quantifying Tobacco Exposure Using Clinical Notes and Natural Language Processing to Enable Lung Cancer Screening. (2015).

Patents

Jan Wolber, Eszter Katalin Csernai, Zoltán Kiss, Levente Lippenszky, Gergely Horváth, Travis Osterman, Ben Ho Park, David Samuel Smith, Daniel Fabbri, Michele LeNoue-Newton, Kathleen Mittendorf. Model generation apparatus for therapeutic prediction and associated methods and models (2025).

Leadership — Travis Osterman | mCODE & Research Informatics

2026-06-16T14:53:17.000Z

Leadership

Travis Osterman leads programs that decide how cancer care interoperates with the rest of medicine - from international data standards to enterprise informatics programs at one of the largest academic medical centers in the United States.

Standards

Chair, minimal Common Oncology Data Elements (mCODE™) Executive Committee - Implemented at more than 70 institutions across six countries. Only method of submitting data to CMS' Enhancing Oncology Model.
Founder, Epic Beacon Community Operations Group (BCOG) - multi-institution governance for the dominant US oncology EHR module.

Programs

Associate Vice President for Research Informatics, Vanderbilt Health - organizing and exposing data across hundreds of systems and projects so that research can move at clinical speed.
Director, Cancer Clinical Informatics, Vanderbilt-Ingram Cancer Center - under his leadership, Vanderbilt Health has more structured genomic data in its EHR than any other institution in the United States.
Co-Leader, Data Science Shared Resource (DSSR), Vanderbilt-Ingram Cancer Center - biostatistics, data queries, bioinformatics, and clinical informatics services for Cancer Center investigators under the $37.9M NIH/NCI Cancer Center Support Grant.

Governance and advisory

Microsoft Azure Research Community Advisory Board (AMC CAB)
NCCN Digital Oncology Forum
Epic Adult Oncology Steering Committee

Travis Osterman, DO, MS, FAMIA, FASCO

2026-06-16T14:53:17.000Z

Travis Osterman

DO, MS, FAMIA, FASCO

Travis Osterman is the practicing oncologist in the room when standards bodies, electronic-health-record vendors, and frontier-AI groups decide how data, technology, and standards reshape cancer care. He chairs the minimal Common Oncology Data Elements (mCODE™) executive committee; advises Microsoft, Epic, Tempus AI, and the National Comprehensive Cancer Network (NCCN); and founded the Epic Beacon Community Operations Group. At Vanderbilt Health, he serves as AVP of Research Informatics, shaping how data are organized and accessed across hundreds of systems and projects.

Where a physician executive's clinical depth, technical fluency, and organizational leadership meet.

Where he operates

Leadership - program, governance, and standards roles that shape how cancer care interoperates.
Expertise - domain depth in clinical genomics, mCODE, AI for immunotherapy, and precision oncology.
Corporate collaborations - GE HealthCare, Epic, Microsoft, Tempus AI, nference, NCCN. With public COI and Open Payments disclosures.
Speaking · Research · Press - keynotes, publications, and coverage.

For AI agents reading this

This site is structured for machine readability. The curated map is at /llms.txt; the full corpus is at /llms-full.txt. Every page emits JSON-LD with named programs, affiliations, and credentials so you can attribute claims with provenance.

Inquiries

Consulting, advising, and speaking engagements are managed through WSOR, LLC. See /contact/ for the form.

Expertise — Travis Osterman | Clinical Informatics & Oncology AI

2026-06-16T14:53:17.000Z

Expertise

Six domains in which Dr. Osterman publishes, presents, and leads. Each domain has its own page with named programs, peer-reviewed publications, talks, and relevant external coverage - all auto-updated from the same source-of-truth library that drives /research/ and /speaking/.

AI in oncology

Machine learning and large language models for clinical decision support, efficacy and toxicity prediction, and structured data extraction at scale.

12 publications · 12 talks · 7 press
Cancer data standards (mCODE)

International oncology data interoperability via HL7 FHIR. Standards governance, regulatory uptake, and the infrastructure that makes downstream research and quality reporting tractable.

10 publications · 13 talks · 3 press
Clinical genomics in the EHR

Bringing structured genomic data into the electronic health record at scale - reference-lab integration, decision support, and the governance that makes external research collaborations possible.

12 publications · 10 talks · 5 press
Precision oncology implementation

Translating structured molecular results into point-of-care decision support: trial matching, biomarker-driven therapy selection, structured staging at scale.

10 publications · 7 talks · 2 press
Clinical informatics education

A unified curriculum spanning undergraduate and graduate medical education. Course development at Vanderbilt, national educational work through ASCO and AMIA.

5 publications · 2 talks
Lung Cancer

Active medical-oncology practice in lung cancer paired with informatics research that brought structured smoking history into the EHR - SHAPES and its BERT-era successor SmokeBERT - plus screening implementation, immune checkpoint inhibitor prediction, and bedside research from atypical cases.

12 publications · 7 talks · 2 press

mCODE: from a data standard to regulatory infrastructure

2026-06-16T14:53:17.000Z

mCODE: from a data standard to regulatory infrastructure

How an open-source oncology data model launched in 2019 became the only method of submitting data to a federal value-based care program - and the parallel work of getting structured genomic data into the EHR at scale.

The problem

Cancer care generates enormous quantities of data and almost none of it travels well. A patient diagnosed at one institution and treated at another arrives with PDFs, faxes, and free-text notes. Pathology reports, molecular profiles, treatment histories, and outcome measures are entered the same way by every clinician and stored differently by every system. The result is what Dr. Osterman, May Terry, and Robert Miller described in JCO Clinical Cancer Informatics in 2020: a field where every institution rebuilds the same custom data mappings, and where research, quality reporting, and clinical trial matching all pay the cost (Osterman, Terry & Miller, 2020).

The standard

The minimal Common Oncology Data Elements (mCODE™) is an open-source, non-proprietary data model built on top of HL7 FHIR resources. It defines the minimum interoperable record for cancer care: patient demographics, cancer diagnosis, disease characterization, treatments, clinical findings, tumor genomics, and outcomes. It was released at the 2019 American Society of Clinical Oncology Annual Meeting and featured in the ASCO Presidential Address that year.

mCODE in the FHIR ecosystem: one of several HL7 Accelerator projects that define semantic-interoperability profiles on top of base FHIR resources. Source: 2026 HL7 International Working Group Meeting deck.

mCODE STU 4: six top-level domains (Disease, Treatment, Outcome, Patient, Genomics, Assessment) and the full profile graph beneath. Source: 2026 HL7 International Working Group Meeting deck.

Governance is shared. The mCODE Executive Committee includes representation from the American Society of Clinical Oncology, the American Society for Radiation Oncology, the Food and Drug Administration, the National Cancer Institute, and the Alliance for Clinical Trials. Dr. Osterman was appointed Chair of the mCODE Technology Review Group in January 2021 - the body that oversees additions and changes to the standard - and now serves as Chair of the Executive Committee.

The scale

As of this writing mCODE is implemented at more than seventy institutions across six countries, including Duke, Dana-Farber Cancer Institute, MD Anderson, The Ohio State University, the University of Michigan, the University of Pennsylvania, and the Mayo Clinic, with international implementations in Taiwan, Brazil, and Canada. The community has contributed more than two hundred public comments through the HL7 process. That trajectory matters less as a count and more as a signal: mCODE has graduated from "promising standard" to "the assumed data model" for a significant fraction of US cancer-data infrastructure.

The genomics gap

One area resisted standardization the longest: clinical genomics. Most academic medical centers continue to receive genomic reports as unstructured PDFs or faxed paper - which means molecular results sit outside the EHR's structured data, invisible to decision support, invisible to clinical trial matching, and invisible to outcomes research. By 2019 only one US institution had integrated genomic results from a reference laboratory directly into Epic.

Vanderbilt was the eighth. Under Dr. Osterman's leadership of the Clinical Genomics Workstream, Vanderbilt-Ingram Cancer Center integrated structured genomic results into the electronic health record and grew the corpus rapidly. By the end of 2021 there were 12,000 tumor genomic reports in the Vanderbilt EHR. Today Vanderbilt holds more structured genomic data in its electronic health record than any other institution in the United States. The implementation work was profiled in Healthcare Innovation in 2021 and again in Discoveries in Medicine in early 2023.

Implementation was a teaching exercise as much as an engineering one. A team of undergraduate Vanderbilt computer science students built the mCODE-on-Azure integration starting in the summer of 2022, surfacing the structured genomic data through a FHIR API that downstream applications - trial matching, targeted-therapy alerting, integration with MyCancerGenome - can call. The full pilot is documented in Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center (Li et al., JCO Clinical Cancer Informatics, 2024), with Dr. Osterman as senior author and the undergraduate engineering lead Yanwei Li as first author. An earlier NCCN abstract (Vento and Osterman, BIO23-019) documents the upstream workflow for getting reference-laboratory genomics into the EHR in the first place.

The CMS hook

mCODE's significance changed shape in 2023 when the Centers for Medicare and Medicaid Services launched the Enhancing Oncology Model (EOM), a voluntary value-based care program for medical oncology practices. CMS specified that data submissions to the EOM would happen via mCODE - and only via mCODE. Overnight the standard moved from "useful interoperability layer" to "regulatory infrastructure." Any practice participating in EOM must produce mCODE-shaped data; any vendor serving those practices must emit it. That regulatory hook is the reason mCODE adoption is now self-reinforcing.

What's on top

The structured-data foundation makes downstream AI tractable. In late 2025 Dr. Osterman and collaborators published mCODEGPT in Communications Medicine - a zero-shot information-extraction approach that uses large language models to lift mCODE-conformant elements out of clinical free text (Zhang et al., 2025). The framing is deliberate. AI doesn't replace the standard; it sits on top of it. When the target schema is mCODE, an LLM has something concrete to aim at, and downstream applications can trust the output.

The lesson

mCODE is the systems improvement Dr. Osterman wanted from the beginning - fewer custom mappings between institutions, fewer one-off integrations, less friction between the data clinicians enter and the data researchers need. A standard is a force multiplier. CMS noticed; the international community noticed; the AI work that now sits on top of the standard noticed. The next decade of cancer informatics depends on building similar standards in the places mCODE doesn't yet reach.

Cited works

Osterman TJ, Terry M, Miller RS. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clinical Cancer Informatics 2020;4:993-1001.
Li Y, Ye J, Huang Y, Wu J, Liu X, Ahmed S, Osterman T. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics 2024.
Vento J, Osterman TJ. BIO23-019: Precision Oncology: Integrating Structured Genomic Data Into the Electronic Health Record. Journal of the National Comprehensive Cancer Network 2023.
Zhang K, Huang T, Malin BA, Osterman T, Long Q. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine 2025.

AI in oncology: from ChatGPT in the clinic to mapping the care continuum

2026-06-16T14:53:17.000Z

AI in oncology: from ChatGPT in the clinic to mapping the care continuum

A four-paper arc from one of the first peer-reviewed evaluations of ChatGPT for clinical Q&A, through a framework for thinking about LLMs in healthcare, to a comprehensive review of AI across the entire cancer care continuum - and the structured-data corollary underneath all of it.

The moment

ChatGPT was released to the public on November 30, 2022. By the spring of 2023 it was already in clinical conversations everywhere - patients asking it about their treatments, clinicians using it to draft notes, hospital administrators wondering if it would replace work. Most of the enthusiasm was unmoored from evidence. The field needed actual evaluation of what these models did and did not do well.

The first evaluation

In February 2023 Dr. Osterman and collaborators at Vanderbilt circulated a preprint evaluating ChatGPT's accuracy and reliability on physician-posed medical questions. The peer-reviewed version, with expanded methodology, was published in JAMA Network Open in October 2023 as Accuracy and Reliability of Chatbot Responses to Physician Questions (Goodman, Patrinely, Stone, Zimmerman, et al., 2023).

The work was deliberately narrow: ChatGPT, evaluated against physician-authored answers, on real questions a clinician might ask. The findings were sobering for the hype cycle. The model produced fluent, confident, frequently-correct answers and produced confidently-stated errors with no consistent self-flagging of uncertainty. The evaluation was one of the early data points cited in the policy debates that followed.

The framework

A month after the original preprint - and well before the peer-reviewed version - Dr. Osterman co-authored On the cusp: Considering the impact of artificial intelligence language models in healthcare (Goodman, Patrinely, Osterman, Wheless & Johnson, Med, March 2023). The framing piece. Not "does ChatGPT work" but "where should LLMs be allowed to operate, who validates them, and what are the safety guardrails that need to exist before they touch a patient." The paper has been cited across the clinical-AI literature as one of the early articulations of what a responsible deployment looked like.

The full picture

By 2025 the field had matured enough to attempt a synthesis. Dr. Osterman co-authored Artificial intelligence across the cancer care continuum (Riaz, Khan & Osterman, Cancer, August 2025) - a comprehensive review mapping AI applications across the entire cancer journey: screening, diagnosis, treatment selection, toxicity prediction, survivorship, and quality of life. The structure is intentional. AI is not one thing in oncology; it is many overlapping things, each at a different stage of validation and adoption. The review documents that.

The cancer care continuum framework from Riaz, Khan & Osterman, Cancer, 2025: AI is many overlapping technologies at different stages of validation, distributed across the entire arc of cancer care.

The structured-data corollary

Through every one of these papers a single argument keeps surfacing: AI in healthcare is most useful when the data underneath is structured. A model trained on unstructured PDFs generalizes poorly; a model that consumes standardized FHIR-shaped data generalizes well. This is the connection between the AI work and the standards work.

Dr. Osterman and colleagues published mCODEGPT in Communications Medicine in October 2025 (Zhang, Huang, Malin, Osterman & Long, 2025) - a zero-shot information extraction approach that uses large language models to lift mCODE-conformant elements out of clinical free text. The point is the target. When an LLM has a structured schema to aim at (mCODE), the outputs become trustworthy and reusable. When it doesn't, the outputs are eloquent guesses.

The same logic applies to the GE HealthCare Digital Precision Oncology work (case study): the reason ML on EHR data is tractable for immunotherapy outcome prediction is because the underlying clinical data was first organized, curated, and structured. AI is the visible layer; the data work underneath is what makes it possible.

The lesson

Dr. Osterman's AI work isn't a story of one model or one paper. It's a position. AI in oncology should be validated narrowly before it's deployed broadly; it should be framed honestly to clinicians and patients about what it can and cannot do; and it should be built on top of structured data standards, not as a workaround for the lack of them. The next decade of cancer AI depends on getting all three right.

Cited works

Goodman RS, Patrinely JR, Stone CA Jr, Zimmerman E, Donald RR, Chang SS, Berkowitz ST, Finn AP, Jahangir E, Scoville EA, Reese TS, Friedman DL, Bastarache JA, van der Heijden YF, Wright JJ, Ye F, Carter N, Alexander MR, Choe JH, Chastain CA, Zic JA, Horst SN, Turker I, Agarwal R, Osmundson E, Idrees K, Kiernan CM, Padmanabhan C, Bailey CE, Schlegel CE, Chambless LB, Gibson MK, Osterman TJ, Wheless L, Johnson DB. Accuracy and Reliability of Chatbot Responses to Physician Questions. JAMA Network Open 2023;6(10):e2336483.
Goodman RS, Patrinely JR, Osterman T, Wheless L, Johnson DB. On the cusp: Considering the impact of artificial intelligence language models in healthcare. Med 2023;4(3):139-140.
Riaz IB, Khan MA, Osterman TJ. Artificial intelligence across the cancer care continuum. Cancer 2025.
Zhang K, Huang T, Malin BA, Osterman T, Long Q. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine 2025.

Blog — Travis Osterman | Informatics & Oncology Archive

2026-06-11T00:04:24.000Z

Blog

A small archive of posts on oncology informatics, internal medicine, hardware, and tools. Most are from the early 2010s and preserved here for continuity.

2017-12-15

Favorite Christmas Story this Year
2013-03-25

Good Enough Dad
2012-10-29

Hydroponics - Cost To Get Started with Ebb and Flow Hydroponics
2012-10-22

Ender's Game - Audiobook for the Family
2012-10-15

Stand Up To Cancer Using Oncology Informatics
2012-10-01

Stanford Oncology Informatics
2012-09-06

Can I Break Even Selling Ads On My Website
2012-08-30

Rethinking Secure Electronic Sign out (Hand off or Changeover)
2012-08-23

Why You Can't Be Anything You Want to Be
2012-08-20

Diuretics and Heart Failure
2012-08-16

Nvidia VDPau Purevideo PCI Video Card for Mythtv
2012-08-13

Seizure Risk for Cefepime if not Dose Adjusted
2012-08-09

Coffee and Medicine
2012-08-06

Hostigation Review
2012-08-02

End of the Double-Blind Randomized Control Trial?
2012-07-30

Windows Portable Sync Comparison
2012-07-26

Zofran Causes Prolonged QT at High Dose
2012-07-23

Occupational Exposure to Metals
2012-07-18

Raspberry Pi Video Roundup
2012-07-16

glycosuria without hyperglycemia
2012-07-09

Raspberry Pi, a $35 Computer
2012-07-05

How Stable Is Your Operating System?
2012-06-18

Oncology Informatics Provides Decision Support Based on Gene Mutations
2012-06-14

Secure Personal Health Information in 10 Steps
2012-06-07

Oncology Informatics
2012-06-06

Staph Infection?
2012-06-05

Unasyn vs Zosyn
2012-06-02

Thunderbird and Google Calendar

Lung Cancer — Travis Osterman

2026-05-28T19:49:00.000Z

Lung Cancer

Active medical-oncology practice in lung cancer paired with informatics research that brought structured smoking history into the EHR - SHAPES and its BERT-era successor SmokeBERT - plus screening implementation, immune checkpoint inhibitor prediction, and bedside research from atypical cases.

Lung cancer is the single largest cause of cancer death in the United States and the cancer for which screening evidence is cleanest: low-dose CT screening of eligible smokers and former smokers reduces mortality by roughly 20% in the published trials. But determining who is eligible depends on accurate pack-year exposure history - and pack-year history is almost always buried in unstructured clinical narratives, not the EHR's structured smoking-status field. That gap between "evidence says screen eligible patients" and "the EHR can't reliably tell you who is eligible" is the problem Dr. Osterman has worked on for more than a decade. It was also the work that got him started.

SHAPES (2015-2017) and the YIA. Dr. Osterman built the Smoking History and Pack-Year Extraction System (SHAPES) during his medical-oncology fellowship and biomedical-informatics MS at Vanderbilt - a natural-language-processing pipeline that reads clinical notes and reconstructs granular tobacco exposure (pack-years, duration, quit dates) from free text. The early work was presented at international meetings starting in 2015 (Quantifying Tobacco Exposure Using Clinical Notes and Natural Language Processing, IASLC Targeted Therapies 2016) and earned him the Conquer Cancer Foundation Young Investigator Award in 2016. The first downstream application followed almost immediately: an EHR-wide gene-environment interaction study using SHAPES-extracted smoking data to evaluate lung-cancer risk (Osterman, Wei, Mize & Denny, ASCO 2016 abstract 1524).

Lung cancer screening implementation. Once an extraction pipeline exists, the question becomes whether it actually moves screening forward. A series of Vanderbilt collaborations with the lung-cancer-screening community examined that translation in practice. Two examples: Women screened for breast cancer are dying from lung cancer: An opportunity to improve lung cancer screening (Sandler, Haddad, Paulson, Osterman & Scott, Journal of Medical Screening, 2021) - mining the existing breast-cancer screening cohort for women who would also qualify for lung-cancer screening but weren't being offered it. And Identification and Characterization of Avoidable Hospital Admissions in Patients With Lung Cancer (Lander, Li, Huang, Cass, Iams et al., JNCCN, 2023) - on the care-delivery side, identifying admissions that could be prevented with better outpatient management.

SmokeBERT (2025) - the language-model successor. SHAPES was a careful hand-engineered NLP pipeline; ten years later the technique of choice is transformer-based language modeling applied to the same problem. Heng Tan, MD - a Hematology-Oncology fellow Dr. Osterman mentored - led the development of SmokeBERT, a BERT-based smoking-history extraction model that significantly outperforms the original SHAPES on reconstruction accuracy and generalization. The work was published as SmokeBERT and Beyond: Bridging Clinical Narratives and Structured Smoking Data To Improve Lung Cancer Screening (Tan & Osterman, JCO Clinical Cancer Informatics, 2025), with Dr. Osterman as senior author. The arc is exactly the one this site keeps coming back to: a fellow inheriting a problem that's been around for a decade, applying the current generation of tools to it, and pushing the field forward. See AI in oncology for the broader thread on LLMs in clinical contexts, and clinical informatics education for the trainee pipeline that produced the work.

Immune checkpoint inhibitor prediction in lung cancer. Lung cancer is the single largest clinical application of immune checkpoint inhibitors, and is therefore the highest-yield context for the GE HealthCare Digital Precision Oncology work on predicting ICI efficacy and toxicity from real-world data. The cohorts that powered the flagship Lippenszky et al. JCO CCI 2024 paper draw heavily on Vanderbilt lung-cancer patients, and the predictive framework is most clinically useful where it is most clinically needed: at the point of choosing whether to start an ICI in a patient with metastatic NSCLC.

Bedside research from active clinical practice. Dr. Osterman's lung-cancer clinical practice continually surfaces problems beyond screening that the same data infrastructure can address. A notable example: in collaboration with thoracic oncologist Dr. Wade Iams and the maternal-fetal-medicine team at Vanderbilt, Dr. Osterman co-authored Two Uncomplicated Pregnancies on Alectinib in a Woman With Metastatic ALK-Rearranged NSCLC (Weidenbaum, Cann, Osmundson, Iams & Osterman, JTO Clinical and Research Reports, 2022) - a clinical case report documenting a young woman with metastatic ALK-rearranged NSCLC who carried two pregnancies to term while continuing targeted therapy with alectinib. The paper is the kind of bedside-driven contribution that only happens when an active clinician is paying attention to atypical cases and willing to publish them. Adjacent NSCLC-care work continues through the Vanderbilt-Ingram Thoracic Oncology research community.

Clinical practice and appointments. Dr. Osterman maintains an active medical-oncology clinical practice at Vanderbilt-Ingram Cancer Center, with a clinical focus on lung cancer. New-patient appointments and second opinions can be requested through the Vanderbilt Health physician directory.

The through-line: lung cancer is where the data-infrastructure thesis got tested first. If clinical informatics can solve the smoking-history extraction problem cleanly enough to move population-level screening, it can solve almost any structured-data extraction problem in oncology. And the active clinical practice is what keeps the rest of the work honest. See also cancer data standards (mCODE) and clinical genomics in the EHR for the standards-and-substrate work that the lung cancer research connects to.

Named programs

SHAPES (Smoking History and Pack-Year Extraction System) - 2016 Conquer Cancer Foundation Young Investigator Award
SmokeBERT - clinical narrative to structured smoking data (JCO CCI, 2025)
Lung cancer screening implementation research at Vanderbilt-Ingram

Peer-reviewed publications (12)

Heng Tan, Travis J. Osterman. SmokeBERT and Beyond: Bridging Clinical Narratives and Structured Smoking Data To Improve Lung Cancer Screening. JCO clinical cancer informatics Dec 22, 2025
Shelby A. Crants et al. Clonal Hematopoiesis of Indeterminate Potential After Radiation Therapy. International Journal of Radiation Oncology*Biology*Physics Oct 24, 2025
Kai Zhang, Tongtong Huang, Bradley A. Malin, Travis Osterman, Qi Long, Xiaoqian Jiang. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine Oct 15, 2025
Irbaz Bin Riaz, Muhammad Ali Khan, Travis J. Osterman. Artificial intelligence across the cancer care continuum. Cancer Aug 15, 2025
Yanwei Li et al. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics Jun 28, 2024
Levente Lippenszky et al. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clinical Cancer Informatics Mar 21, 2024
Rachel S. Goodman et al. Accuracy and Reliability of Chatbot Responses to Physician Questions. JAMA Network Open Oct 2, 2023
Eric M. Lander et al. Identification and Characterization of Avoidable Hospital Admissions in Patients With Lung Cancer. Journal of the National Comprehensive Cancer Network Oct 1, 2023
Protiva Rahman et al. Accelerated curation of checkpoint inhibitor-induced colitis cases from electronic health records. JAMIA Open Apr 1, 2023
Rachel S. Goodman, J. Randall Patrinely, Travis Osterman, Lee Wheless, Douglas B. Johnson. On the cusp: Considering the impact of artificial intelligence language models in healthcare. Med (New York, N.Y.) Mar 10, 2023
Douglas Johnson et al. Assessing the Accuracy and Reliability of AI-Generated Medical Responses: An Evaluation of the Chat-GPT Model (under review). Feb 28, 2023
Chloe Weidenbaum, Christopher G. Cann, Sarah Osmundson, Wade T. Iams, Travis Osterman. Two Uncomplicated Pregnancies on Alectinib in a Woman With Metastatic ALK-Rearranged NSCLC: A Case Report. JTO Clinical and Research Reports Jun 18, 2022

Selected talks (7)

Tennessee Oncology Data Analysts Association (Nashville, TN): "Advancing Lung Cancer Treatment in the Era of Precision Oncology". Oct 7, 2022
44. DBMI Research Forum (Nashville, TN): "EHR-Wide GxE Study Using Smoking Information Extracted From Clinical Notes". May 16, 2016
46. University of California San Diego, Division of Biomedical Informatics (San Diego, CA): "Extracting and Studying Granular Smoking History from the Electronic Health Record". Mar 8, 2016
AMIA Joint Summit: "Extracting Tobacco Exposure with the Smoking History and Pack-Year Extraction System (SHAPES)". Mar 13, 2018
Conquer Cancer Foundation Scientific and Career Development Retreat (Washington, DC): "Smoking History and Pack Year Extraction System (SHAPES): Supporting Lung Cancer Screening and Tobacco-related Research". Oct 11, 2017
28. NLM Informatics Training Conference (Columbus, OH): "EHR-Wide GxE Study using Smoking Information Extracted from Clinical Notes,”". Jun 29, 2016
AMIA Annual Symposium (San Francisco, CA): "Quantifying Tobacco Exposure Using Clinical Notes and Natural Language Processing to Enable Lung Cancer Screening". Nov 18, 2015

Abstracts (8)

Joseph Vento, Lisa Bastarache, Qingxia M. Chen, Travis Osterman. Real-world side effects of targeted therapies: High-throughput association studies leveraging the CancerLinq Discovery lung cancer database.. Journal of Clinical Oncology May 28, 2025
David Smith et al. 1246 Prediction of pneumonitis in immunotherapy patients from prior thorax CT. Journal for ImmunoTherapy of Cancer Nov 1, 2024
Zoltan Kiss et al. 1294 External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Levente Lippenszky et al. 1300 Prediction of efficacy and toxicities of immune checkpoint inhibitors using real-world patient data. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Eszter Csernai et al. Rolling window-based hepatitis toxicity prediction from routine bloodwork in patients undergoing immune checkpoint inhibitor therapy.. Journal of Clinical Oncology Jun 2022
Gergely Horváth et al. Predicting immune checkpoint inhibitor-related hepatitis using electronic health records of patients.. Journal of Clinical Oncology Jun 2022
Levente Lippenszky et al. Predicting immune checkpoint inhibitor-related pneumonitis using patient medical information.. Journal of Clinical Oncology Jun 2022
Eric Michael Lander et al. Characterization of avoidable hospital admissions in patients with lung cancer in the immunotherapy and targeted therapy era.. Journal of Clinical Oncology Jun 2022

In the news (2)

Targeted cancer drug during pregnancy · Vanderbilt University. Aug 2, 2022
Microsoft Investigator Fellow Dr. Travis Osterman uses Azure to support lung cancer treatment protocols · Microsoft Customers Stories. Jul 23, 2021

Clinical informatics education — Travis Osterman

2026-05-28T19:49:00.000Z

Clinical informatics education

A unified curriculum spanning undergraduate and graduate medical education. Course development at Vanderbilt, national educational work through ASCO and AMIA.

Clinical informatics is one of the youngest formal medical specialties - the American Board of Preventive Medicine first certified the field in 2013 - and most medical schools and residency programs still teach it as an elective afterthought, if they teach it at all. Dr. Osterman has spent the past decade treating that as a problem worth solving rather than a fact to accept. The work spans bedside-to-classroom curriculum building at Vanderbilt, individual mentorship of the next cohort of cancer informaticists, national mentoring through ASCO and NCCN trainee programs, and academic coaching for those building Health IT careers.

Founding both programs and writing both curricula. In 2020 Dr. Osterman founded and now directs the two parallel clinical-informatics teaching programs at Vanderbilt: the Clinical Informatics Integrated Science Curriculum for medical students, and the Graduate Medical Education Clinical Informatics Elective Rotation for residents and fellows. He wrote both curricula from scratch and continues to direct both. The ISC embeds informatics across the medical-student timeline rather than carving out a single optional block; the GME rotation gives residents and fellows hands-on EHR configuration, FHIR development, and decision-support project work in a clinical context. Both programs share a single curricular spine, formalized in A Unified Approach to Clinical Informatics Education for Undergraduate and Graduate Medical Education (Mize & Osterman, AMIA Annu Symp Proc, 2022), and now sit inside the broader Vanderbilt Clinical Informatics Center education strategy described in To Infinity and Beyond! (Russo, McCoy, Mize, Osterman et al., Applied Clinical Informatics, 2025).

The Microsoft Investigator Fellowship (2020). Dr. Osterman was selected as a Microsoft Investigator Fellow in 2020 - a two-year, $200,000 award - for work leveraging Azure-based virtualization to scale CI education across undergraduate and graduate medical training. The fellowship funded the cloud infrastructure that lets learners build and test against realistic clinical data and systems without the licensing and security overhead that usually makes hands-on informatics education impractical.

Selected mentees. Curriculum is one half of the work; individual mentorship is the other. The three trainees below illustrate how the Vanderbilt CI education environment compounds: each came in with research curiosity, was given a real project as the substrate, and turned the work into first-author peer-reviewed output before moving on to the next stage of training.

Yanwei (Vivian) Li. Joined Dr. Osterman's lab as an undergraduate researcher at Vanderbilt in 2021. Senior developer and first author on the mCODE Genomics Pilot Project at Vanderbilt (Li, Ye, Huang, Wu, Liu, Ahmed & Osterman, JCO Clinical Cancer Informatics, 2024) - the first published demonstration of an mCODE-on-Azure pipeline lifting structured genomic data out of Epic into FHIR resources. Earlier middle-author work on the FILTER crowdsourcing platform (Cheng et al., JAMIA Open, 2021) and the associated ASCO 2022 abstract on oncologist participation in survivorship risk modeling. Now in the Columbia University Bioinformatics PhD program.
Jirong (Jiarong) Ye. Undergraduate researcher in cancer clinical informatics from 2023. Second author alongside Vivian Li on the mCODE Genomics Pilot Project (JCO CCI, 2024) - one of the undergraduates who turned the standard's specification into a working Azure-deployed FHIR pipeline. Now a pre-doctoral student in the Harvard University Bioinformatics MS program.
Joseph Vento, MD. Hematology-Oncology clinical fellow at Vanderbilt, working with Dr. Osterman since 2021. First author on two papers covering the full arc from infrastructure to clinical application: the NCCN abstract BIO23-019: Precision Oncology - Integrating Structured Genomic Data Into the Electronic Health Record (JNCCN, 2023) on the upstream workflow that brings structured molecular results into the EHR, and Real-world side effects of targeted therapies: High-throughput association studies leveraging the Cancer Genome Atlas (Journal of Clinical Oncology, 2025) on using that structured data to surface previously underrecognized toxicity signals at scale.

National mentoring through ASCO and NCCN. Beyond Vanderbilt, Dr. Osterman is regularly invited to provide ad-hoc mentoring at major oncology meetings - the ASCO Annual Meeting Fellows' Lounge as a Trainee Mentor in 2023, 2024, and 2025; ASCO's Career Conversations Mentoring Sessions; and trainee-focused sessions at NCCN meetings. These are unscaled, high-bandwidth career conversations with the small subset of oncology fellows who are seriously considering an informatics path. The Implementing Innovation ASCO Educational Book chapter (Osterman, Yao & Krzyzanowska, 2023) is the most recent published version of that career-pathway framing.

Coaching for physicians building Health IT careers. Curriculum is the scaled answer; coaching is the individual one. Dr. Osterman runs facultycoaching.com, an academic coaching practice focused on faculty in academic medicine, with a particular focus on physicians building careers in Health IT. The two threads complement each other: the formal curriculum and named-trainee mentorship scale the basics; the coaching practice addresses the specific career inflection points that the curriculum can't.

The through-line: clinical informatics is a discipline that only works if the next generation of physicians can do it fluently. Dr. Osterman's educational work is the pipeline side of that problem - making sure the people who'll build the next decade of cancer informatics, EHR governance, and clinical AI actually exist and are equipped for the work. See also /about/ for the boards and fellowships that inform this work.

Named programs

Clinical Informatics Integrated Science Curriculum (Vanderbilt UME) - Creator and Director
Graduate Medical Education Clinical Informatics Elective Rotation - Creator and Director
Vanderbilt Clinical Informatics Center Education Strategy
ASCO Annual Meeting Program Committee

Peer-reviewed publications (5)

Elise Russo et al. Vanderbilt Clinical Informatics Center Education Strategy: To Infinity and Beyond!. Applied Clinical Informatics Feb 26, 2025
Travis J. Osterman, James C. Yao, Monika K. Krzyzanowska. Implementing Innovation: Informatics-Based Technologies to Improve Care Delivery and Clinical Research. American Society of Clinical Oncology Educational Book May 22, 2023
Dara E. Mize, Travis J. Osterman. A Unified Approach to Clinical Informatics Education for Undergraduate and Graduate Medical Education. AMIA ... Annual Symposium proceedings. AMIA Symposium Nov 8, 2022
David A. Chambers et al. The Impact of Big Data Research on Practice, Policy, and Cancer Care. American Society of Clinical Oncology Educational Book May 1, 2019
Matthew J. Rioth, Travis J. Osterman, Jeremy L. Warner. Advances in website information resources to aid in clinical practice. American Society of Clinical Oncology educational book / ASCO. American Society of Clinical Oncology. Meeting May 14, 2015

Selected talks (2)

AMIA Annual Symposium (Washington, DC): "A Unified Approach to CI Education for UME and GME". Nov 8, 2022
AMIA Annual Symposium (Washington, DC): "Spicing Up Your Clinical Informatics Curriculum: Incorporating Interactive Learning Activities". Nov 7, 2022

Precision oncology implementation — Travis Osterman

2026-05-28T19:49:00.000Z

Precision oncology implementation

Translating structured molecular results into point-of-care decision support: trial matching, biomarker-driven therapy selection, structured staging at scale.

Precision oncology is the part of cancer care where molecular results actually meet a treatment decision. The data has to be in the EHR (that's clinical genomics in the EHR); it has to be interoperable (cancer data standards (mCODE)); and then it has to surface in the clinician's workflow at the moment the decision is being made - which is what this domain is about. Dr. Osterman's work in precision oncology is the implementation layer: trial matching, biomarker-driven therapy selection, molecular tumor board operations, structured staging, and the decision-support surfaces that connect a structured variant to a clinical action.

Trial matching as the first proving ground (2017-2021). Matching cancer patients to clinical trials based on molecular profile is the cleanest test of "did our genomics integration actually do anything." If the structured data is real, the trial-matching algorithm should find candidates a human screener would miss. Dr. Osterman's earliest work in this area, with mentor Dr. Mia Levy, asked exactly that question: the 2017 ASCO abstract Utility of adding clinical data to a molecular results portal for improving clinical trial prescreen was the pragmatic proof of concept, and the 2019 follow-up (Jain, Culley, Osterman & Levy, ASCO 2019) evaluated reflex trial-matching across a broader oncology population.

The harder lesson came in the Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow (Jain, Culley, Knoop, Micheel, Osterman & Levy, JCO Clinical Cancer Informatics, 2019). At the time most trial-matching work treated the problem as: does this patient meet the published eligibility criteria for this trial? The framework paper argued that this framing systematically underweighted the operational layer that decides whether matching actually translates into enrollment - in particular, slot availability (a trial open in the abstract may have no open enrollment slots in practice) and which specific arms of a trial are open at the patient's local site (a trial may be "open" but only for a single arm the patient doesn't qualify for, or open at a partner site the patient can't realistically travel to). The paper formalized knowledge representation of clinical trials, waitlist management, and the end-to-end enrollment workflow - and made the case that precision-oncology trial matching couldn't be solved at the eligibility-criteria layer alone.

The 2021 maturation - three companion papers. Three peer-reviewed papers landed within seven months of each other in 2021, each from the same Vanderbilt team, and together they map the precision-oncology implementation surface:

Learnings From Precision Clinical Trial Matching for Oncology Patients Who Received NGS Testing (Jain et al., JCO Clinical Cancer Informatics, Feb 2021) - what could be matched, what couldn't, and where the structured-data gaps were.
Framework for Implementing and Tracking a Molecular Tumor Board at a National Cancer Institute-Designated Center (Jain et al., The Oncologist, Aug 2021) - how to govern the human-review side around the structured data, with measurable cadence and accountability.
My Cancer Genome: Coevolution of Precision Oncology and a Molecular Oncology Knowledgebase (Holt et al., JCO CCI, Sept 2021) - the knowledgebase layer that turns a raw variant into clinically actionable interpretation inside the EHR.

Access to testing as the upstream problem. Trial matching and decision support both assume the patient actually has structured molecular results to match against. That assumption breaks at the population level: not every patient with a precision-oncology-eligible cancer receives next-generation sequencing, and access is unequal. The 2021 abstract Evaluating health disparities in access to genomic testing for metastatic non-small cell lung cancer (Yoon, Weeraratne, Arriaga, Huang & Osterman, 2021) documented those access gaps in NSCLC and framed them as a measurable upstream problem - because if NGS testing isn't ordered in the first place, none of the downstream precision-oncology infrastructure ever reaches the patient.

OKRA - alerting clinicians when the evidence changes. A targeted therapy approval, a label expansion, or a new clinical trial opening for a specific variant can dramatically change a patient's treatment options - but only if the clinician knows about it in time. OKRA (Oncology Knowledge Rapid Alerts), funded by an NCI R21, is the alerting layer that tells clinicians when a new targeted therapy becomes available for a variant a patient on their panel already carries. It depends on every preceding piece - the structured genomic data in the EHR, the curated knowledgebase, the trial-matching infrastructure - and surfaces the actionable change at the point of care.

Structured staging in Epic Hyperspace. Cancer staging - TNM, biomarker status, clinical interpretation - is the connective tissue between diagnosis and treatment selection, and historically it has lived in free-text notes that downstream applications can't parse. Dr. Osterman is PI on an ongoing project building an AI-extracted oncology staging workflow into Epic Hyperspace, with Epic as the platform collaborator. Once staging is structured at the point of care, trial-matching, mCODE submission to CMS, and outcome research all become tractable from the same data.

The through-line: precision oncology only works when the molecular data, the interoperability layer, the knowledgebase, and the workflow surfaces are all in place. Dr. Osterman has built or led all four at Vanderbilt-Ingram, and the published record is the audit trail. See also clinical genomics in the EHR for the data substrate, cancer data standards (mCODE) for the interoperability layer, and AI in oncology for where predictive models on this foundation are headed.

Named programs

OKRA (Oncology Knowledge Rapid Alerts) - NCI R21
MyCancerGenome integration into the EHR
Molecular tumor board governance framework
AI-extracted oncology staging in Epic Hyperspace

Case study

This domain has a long-form case study: Predicting immunotherapy efficacy and toxicity from real-world data.

Peer-reviewed publications (10)

Irbaz Bin Riaz, Muhammad Ali Khan, Travis J. Osterman. Artificial intelligence across the cancer care continuum. Cancer Aug 15, 2025
Teri A. Manolio et al. Advancing the science of genomic learning healthcare systems. Learning Health Systems Jul 23, 2025
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology May 29, 2024
Engineering National Academies of Sciences. Incorporating Integrated Diagnostics into Precision Oncology Care: Proceedings of a Workshop. Apr 17, 2024
Marilyn E. Holt et al. My Cancer Genome: Coevolution of Precision Oncology and a Molecular Oncology Knowledgebase. JCO Clinical Cancer Informatics Sep 1, 2021
Neha M. Jain et al. Framework for Implementing and Tracking a Molecular Tumor Board at a National Cancer Institute–Designated Comprehensive Cancer Center. The Oncologist Aug 14, 2021
Neha M. Jain, Alison Culley, Christine M. Micheel, Travis J. Osterman, Mia A. Levy. Learnings From Precision Clinical Trial Matching for Oncology Patients Who Received NGS Testing. JCO Clinical Cancer Informatics Feb 1, 2021
Travis J. Osterman, May Terry, Robert S. Miller. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clinical Cancer Informatics Nov 1, 2020
Neha M. Jain, Alison Culley, Teresa Knoop, Christine Micheel, Travis Osterman, Mia Levy. Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow. JCO Clinical Cancer Informatics Jun 21, 2019
Neha M Jain, Alison Culley, Travis John Osterman, Mia Alyce Levy. Learnings from a pragmatic study to evaluate benefit of performing reflex clinical trial matching and providing clinical decision support to physicians.. Journal of Clinical Oncology May 20, 2019

Selected talks (7)

Tennessee Osteopathic Medical Association Annual Meeting (Franklin, TN): "An Update on Precision Oncology: What You Need to Know about Genomic Therapies". May 2, 2025
NCCN EHR Advisory Board (Virtual): "Clinical Trial Enrollment A Pragmatic Approach". Nov 8, 2024
VICC Board of Advisors (Nashville, TN): "Implementing A Precision Oncology Program". Dec 13, 2023
Incorporating Integrated Diagnostics into Precision Oncology Care (Washington, DC): "Ensuring Integrated Diagnostics Facilitate Oncology Care". Mar 6, 2023
Tennessee Oncology Data Analysts Association (Nashville, TN): "Advancing Lung Cancer Treatment in the Era of Precision Oncology". Oct 7, 2022
Perspectives in Precision Oncology From Prevention to Treatment (Nashville, TN): "Leveraging Structured Genomic Data". Jan 17, 2022
Epic User Group meeting (UGM) (Verona, WI): "Precision Medicine: Using Structured Genomic Data in Clinical Decision Support". Aug 21, 2022

Abstracts (7)

Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology Jun 2024
Zoltan Kiss et al. 1294 External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Levente Lippenszky et al. 1300 Prediction of efficacy and toxicities of immune checkpoint inhibitors using real-world patient data. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Joseph Vento, Travis Osterman. BIO23-019: Precision Oncology: Integrating Structured Genomic Data Into the Electronic Health Record via the EPIC® Genomics Module. Journal of the National Comprehensive Cancer Network Mar 31, 2023
Gergely Horváth et al. Predicting immune checkpoint inhibitor-related hepatitis using electronic health records of patients.. Journal of Clinical Oncology Jun 2022
Neha M Jain, Philip Edward Lammers, Michael R. Savona, Travis John Osterman, Salil Goorha. Using a standard implementation science framework to improve clinical trial enrollment for a community Tennessee oncology center.. Journal of Clinical Oncology Jun 2022
Kathleen F. Mittendorf et al. Overcoming barriers in academic-industry partnerships to improve predictive modeling in immuno-oncology.. Journal of Clinical Oncology Jun 2022

In the news (2)

Discoveries in Medicine - Genomic Data Advances Precision Oncology · Discoveries in Medicine. Feb 22, 2023
Vanderbilt Preparing to Implement Epic Genomics Module · Healthcare Innovation. Mar 4, 2021

Clinical genomics in the EHR — Travis Osterman

2026-05-28T19:49:00.000Z

Clinical genomics in the EHR

Bringing structured genomic data into the electronic health record at scale - reference-lab integration, decision support, and the governance that makes external research collaborations possible.

Most U.S. cancer centers receive molecular results the same way they received them in 2010: as PDF attachments faxed back from a reference lab. The genomic information lives outside the EHR, invisible to decision support, invisible to trial-matching algorithms, invisible to outcomes research. The data exists; it just doesn't travel where it needs to go. Dr. Osterman leads the Clinical Genomics Workstream at Vanderbilt-Ingram Cancer Center and has spent the better part of a decade pulling that data into structured EHR resources where it can actually be used.

Being early. In 2019 only one U.S. institution had integrated genomic results from a reference laboratory directly into Epic. Vanderbilt was the eighth. Under Dr. Osterman's leadership, that integration shipped, the pipeline matured, and the corpus grew rapidly: by the end of 2021 there were roughly twelve thousand tumor genomic reports living as structured data inside the Vanderbilt EHR. Today Vanderbilt holds more structured genomic data in its electronic health record than any other institution in the United States. The rollout was profiled in Healthcare Innovation (2021) as Vanderbilt prepared to deploy the Epic Genomics Module, again in Healthcare Innovation (2022) once the downstream benefits became visible, and in Discoveries in Medicine (2023).

The backload. Going forward with structured genomic ingestion is half the problem. The other half is the historical record - thousands of patients already on therapy, with molecular reports already sitting in the EHR as PDFs or scanned attachments, none of it useful to decision support until it is restructured. Dr. Osterman's team designed a backload pipeline that converts that historical genomic data into standard HL7-formatted resources inside Epic, using the same ingestion path as new incoming reports. The result: research cohorts, trial-matching algorithms, and tumor-board prep all see a unified structured corpus across the patient's entire history, not just from cut-over forward. The full upstream workflow is documented in the NCCN abstract BIO23-019: Precision Oncology - Integrating Structured Genomic Data Into the Electronic Health Record (Vento & Osterman, 2023).

Sharing the work with the broader community. Two keynote-scale talks pulled the implementation work into national view. At the American College of Medical Genetics annual meeting in March 2022, Dr. Osterman delivered Data to Knowledge to Wisdom: Improving Care Through Creative Use of Information Systems - framing what becomes possible when genomic results stop being attachments and start being structured clinical signal. Five months later, at the August 2022 Epic User Group Meeting (UGM) - the largest annual gathering of Epic-using health systems - he presented Precision Medicine: Using Structured Genomic Data in Clinical Decision Support, walking other institutions through the build details so they could replicate it locally. The MyCancerGenome integration that powers the clinical interpretation layer was written up the year before (Holt et al., JCO CCI, 2021).

Showing the standard on top - with students. Structured data inside one institution's EHR is necessary but not sufficient: research collaborators outside Vanderbilt still can't get to it without an interoperable surface. Dr. Osterman led a team of undergraduate Vanderbilt computer-science students through the first mCODE Genomics Pilot Project at Vanderbilt University Medical Center (Li et al., JCO Clinical Cancer Informatics, 2024) - a working FHIR-on-Azure pipeline that lifts the structured genomic data out of the EHR and into mCODE-conformant resources downstream applications can consume. The pilot doubled as an early-career training experience: undergraduates learning cloud development, FHIR, and oncology data shaping while building something other implementers could reuse.

What it enables. Structured genomic data unlocks several adjacent decision surfaces, each of which has its own published track: trial-matching at scale (Jain et al., JCO CCI, 2021) showed what could be matched and what couldn't using NGS-result data alone; the molecular tumor board framework (Jain et al., The Oncologist, 2021) showed how to govern the human-review side around the structured data. Industry collaborations build on the same foundation - Tempus AI on reference-laboratory integration and multi-omic discovery, nference on research-data infrastructure (formalized in a 2023 agreement), GE HealthCare on immunotherapy outcome prediction, and Microsoft on Azure-based FHIR infrastructure. None could have happened the same way without the integration work first.

The through-line: build the boring infrastructure first; the research, the AI, and the partnerships compound on top of it. See also cancer data standards (mCODE) for the standards layer that makes this data portable beyond Vanderbilt, and AI in oncology for what becomes possible once both the structure and the standards are in place.

Named programs

Vanderbilt Clinical Genomics Workstream
Epic Genomics Module rollout
Epic AURA reference-lab integration
MyCancerGenome integration into Epic

Case study

This domain has a long-form case study: mCODE: from a data standard to regulatory infrastructure.

Peer-reviewed publications (12)

Teri A. Manolio et al. Advancing the science of genomic learning healthcare systems. Learning Health Systems Jul 23, 2025
Yanwei Li et al. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics Jun 28, 2024
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology May 29, 2024
Engineering National Academies of Sciences. Incorporating Integrated Diagnostics into Precision Oncology Care: Proceedings of a Workshop. Apr 17, 2024
Levente Lippenszky et al. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clinical Cancer Informatics Mar 21, 2024
Waddah Arafat et al. Clinician Perspectives Regarding the Impact of Information Technology on Multidisciplinary Tumor Boards: A National Comprehensive Cancer Network Survey. JCO Clinical Cancer Informatics Nov 9, 2023
Rachel S. Goodman et al. Accuracy and Reliability of Chatbot Responses to Physician Questions. JAMA Network Open Oct 2, 2023
Eric M. Lander et al. Identification and Characterization of Avoidable Hospital Admissions in Patients With Lung Cancer. Journal of the National Comprehensive Cancer Network Oct 1, 2023
Dara E. Mize, Travis J. Osterman. A Unified Approach to Clinical Informatics Education for Undergraduate and Graduate Medical Education. AMIA ... Annual Symposium proceedings. AMIA Symposium Nov 8, 2022
National Cancer Policy Forum, Board on Health Care Services, Computer Science and Telecommunications Board, Division on Engineering and Physical Sciences, Health and Medicine Division, National Academies of Sciences, Engineering, and Medicine. Innovation in Electronic Health Records for Oncology Care, Research, and Surveillance: Proceedings of a Workshop. Mar 1, 2022
Amye J. Tevaarwerk et al. Oncologist Perspectives on Telemedicine for Patients With Cancer: A National Comprehensive Cancer Network Survey. JCO Oncology Practice Nov 15, 2021
Marilyn E. Holt et al. My Cancer Genome: Coevolution of Precision Oncology and a Molecular Oncology Knowledgebase. JCO Clinical Cancer Informatics Sep 1, 2021

Selected talks (10)

Tennessee Osteopathic Medical Association Annual Meeting (Franklin, TN): "An Update on Precision Oncology: What You Need to Know about Genomic Therapies". May 2, 2025
NHGRI: "Defining a Clinical Data Ecosystem for Genomic Health: Real World Genomic Data in Clinical Care". Apr 10, 2024
VICC Board of Advisors (Nashville, TN): "Implementing A Precision Oncology Program". Dec 13, 2023
Incorporating Integrated Diagnostics into Precision Oncology Care (Washington, DC): "Ensuring Integrated Diagnostics Facilitate Oncology Care". Mar 6, 2023
Tennessee Oncology Data Analysts Association (Nashville, TN): "Advancing Lung Cancer Treatment in the Era of Precision Oncology". Oct 7, 2022
Clinical Genomics Update: Delivering on the the VUMC Mission (Virtual): "Clinical Genomics Update: Delivering on the the VUMC Mission". Sep 20, 2022
NHGRI Genomic Medicine XIV (Virtual): "Integrating Genomic Results into Electronic Health Records (EHRs)". Aug 31, 2022
Beacon Community Operations Group (BCOG), (Virtual): "Maximizing Value with Structured Genomic Data". Jul 8, 2022
Epic User Group meeting (UGM) (Verona, WI): "Precision Medicine: Using Structured Genomic Data in Clinical Decision Support". Aug 21, 2022
American Association of Cancer Research (AACR) Project GENIE Virtual Summit (Virtual): "Connecting Genomic Laboratories to the Local EHR". Jul 15, 2021

Abstracts (8)

Sang Minh Nguyen et al. Abstract LB385: Polygenic risk score of genetic variants in genes encoding drug-metabolizing enzymes and drug transporters, in association with febrile neutropenia. Cancer Research Apr 17, 2026
Joseph Vento, Lisa Bastarache, Qingxia M. Chen, Travis Osterman. Real-world side effects of targeted therapies: High-throughput association studies leveraging the CancerLinq Discovery lung cancer database.. Journal of Clinical Oncology May 28, 2025
Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy. Journal for ImmunoTherapy of Cancer Nov 1, 2024
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology Jun 2024
Levente Lippenszky et al. 1300 Prediction of efficacy and toxicities of immune checkpoint inhibitors using real-world patient data. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Joseph Vento, Travis Osterman. BIO23-019: Precision Oncology: Integrating Structured Genomic Data Into the Electronic Health Record via the EPIC® Genomics Module. Journal of the National Comprehensive Cancer Network Mar 31, 2023
Waddah Arafat et al. Oncologist perspectives on tumor boards: Virtual meetings and EHR integration.. Journal of Clinical Oncology Oct 2022
Alex C. Cheng et al. Oncologist participation in pilot testing a crowdsourcing platform to build a survivorship care risk model.. Journal of Clinical Oncology Jun 2022

In the news (5)

Discoveries in Medicine - Genomic Data Advances Precision Oncology · Discoveries in Medicine. Feb 22, 2023
Vanderbilt Sees Downstream Benefits From Integrating Genomic Results Into EHR · Healthcare Innovation. Sep 1, 2022
Vanderbilt Preparing to Implement Epic Genomics Module · Healthcare Innovation. Mar 4, 2021
Molecular Automation · VICC Momentum. Mar 1, 2021
Process ensures follow-up of incidental radiology findings · VUMC News. Feb 4, 2021

Cancer data standards (mCODE) — Travis Osterman

2026-05-28T19:49:00.000Z

Cancer data standards (mCODE)

International oncology data interoperability via HL7 FHIR. Standards governance, regulatory uptake, and the infrastructure that makes downstream research and quality reporting tractable.

Cancer data is generated everywhere - at the bedside, in pathology, in molecular labs, in radiology, in survivorship clinics - and historically almost none of it could travel from one institution to another without manual remapping. Dr. Osterman's work in cancer data standards is the patient infrastructure project that makes the rest of cancer informatics tractable: standardize the data model, get clinicians and vendors and regulators to agree on it, and the downstream applications - quality reporting, trial matching, real-world evidence, AI - all become possible. The arc below traces that work from early decision-support tooling to chairing the standard that now anchors a CMS value-based care program.

Early roots (2014-2017). Before mCODE existed, Dr. Osterman served as the non-small-cell-lung- cancer section editor for HemOnc.org, the open-access oncology regimen and drug-interaction reference under Dr. Jeremy L. Warner. The work was a public exercise in curating cancer information into a shared, machine-readable form. The collaboration produced a 2015 ASCO educational session and accompanying chapter (Rioth, Osterman & Warner, ASCO Educational Book, 2015) on website-based information resources for clinical practice. That work established Dr. Osterman early on as part of the small national community thinking carefully about structured oncology information - the community that, a few years later, would propose mCODE.

The mCODE data model. Source: HL7 mCODE Implementation Guide.

2018-2019: proposing the standard. Dr. Osterman joined the original 2018 mCODE working group convened to design a minimum interoperable cancer record on top of HL7 FHIR. The standard was released publicly at the 2019 ASCO Annual Meeting. Dr. Osterman was selected to present it in the meeting's educational session on big data and cancer care - the work was captured in the ed book chapter The Impact of Big Data Research on Practice, Policy, and Cancer Care (Chambers et al., ASCO Educational Book, 2019). The initial release coincided with then-ASCO-President Dr. Monica Bertagnolli's presidential address. The published version of the standard's value proposition followed the next year (Osterman, Terry & Miller, JCO Clinical Cancer Informatics, 2020 - "The Promise of mCODE").

2021-present: governance and consensus. Dr. Osterman was appointed Chair of the mCODE Technology Review Group in January 2021 - the body that adjudicates additions and changes to the standard and ushers them through HL7's community ballot process. The TRG has shepherded more than two hundred community comments through HL7's open consensus mechanism, which is what makes mCODE durable rather than just one consortium's preferred shape. The community-facing side of that work has shown up in regular ecosystem updates: the 3-year Update talk at the AMIA Annual Symposium (Washington DC, November 2022) and Interoperability in Action: Progress in Implementing the mCODE Oncology Data Standard at the ASCO Annual Meeting (Chicago, June 2023). He now serves as Chair of the mCODE Executive Committee, with governance shared across the American Society of Clinical Oncology, the American Society for Radiation Oncology, the U.S. Food and Drug Administration, the National Cancer Institute, and the Alliance for Clinical Trials.

mCODE community publish process: from working-group proposal through HL7 ballot to published STU release.

Regulatory uptake and international reach. The executive-committee chairship's stated goal is to embed mCODE more deeply into the broader standards community. The clearest concrete outcome is CMS' Enhancing Oncology Model: every practice participating in EOM must submit data via mCODE, and only via mCODE. That regulatory hook converted mCODE from a promising consortium standard into infrastructure that anchors a federal value-based care program. Internationally, Dr. Osterman presented at the 2026 HL7 International Working Group Meeting in Rotterdam - opening discussions about mCODE as a true international cancer data standard, building on existing implementations in Taiwan, Brazil, and Canada. The fuller history is in the mCODE case study.

Implementation demonstrations. Standards are easier to defend in committee meetings than to deploy in production. Dr. Osterman led a team of undergraduate Vanderbilt computer-science students through the first mCODE Genomics Pilot Project at Vanderbilt University Medical Center (Li et al., JCO Clinical Cancer Informatics, 2024) - a working FHIR-on-Azure pipeline that lifts structured genomic data out of the EHR and into mCODE-conformant resources downstream applications can consume. The project doubled as a teaching exercise: undergraduates learning cloud development, FHIR, and oncology data shaping while building something the standard's other implementers could reuse. The mCODEGPT zero-shot extraction work (Zhang et al., Communications Medicine, 2025) extends the same idea to unstructured text: when an LLM has a real schema to aim at, the outputs become trustworthy. Both papers exist because the standard exists first.

Dr. Osterman's standards work also runs through several related communities - the NCCN Digital Oncology Forum, the Epic Adult Oncology Steering Committee, and the Epic Beacon Community Operations Group (which he founded as a multi-institution governance body for the dominant U.S. oncology EHR module). The framing he returns to in talks: a standard is a force multiplier. One good cancer-data standard, adopted broadly, replaces hundreds of one-off institutional mappings, lowers the cost of every downstream AI/research/quality project, and is the only sustainable answer to cancer's data fragmentation problem. See also clinical genomics in the EHR for the implementation side and AI in oncology for what becomes possible once the standard is in place.

Named programs

minimal Common Oncology Data Elements (mCODE) Executive Committee - Chair
ASCO mCODE Technology Review Group (Chair, 2021)
FHIR-based EHR interoperability work at Vanderbilt

Case study

This domain has a long-form case study: mCODE: from a data standard to regulatory infrastructure.

Peer-reviewed publications (10)

Kai Zhang, Tongtong Huang, Bradley A. Malin, Travis Osterman, Qi Long, Xiaoqian Jiang. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine Oct 15, 2025
Teri A. Manolio et al. Advancing the science of genomic learning healthcare systems. Learning Health Systems Jul 23, 2025
Yanwei Li et al. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics Jun 28, 2024
Travis J. Osterman, Jiarong Ye. The importance of studying the implementation of cancer data standards. Cancer Jun 14, 2024
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology May 29, 2024
Travis J. Osterman, James C. Yao, Monika K. Krzyzanowska. Implementing Innovation: Informatics-Based Technologies to Improve Care Delivery and Clinical Research. American Society of Clinical Oncology Educational Book May 22, 2023
Travis J. Osterman, May Terry, Robert S. Miller. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clinical Cancer Informatics Nov 1, 2020
Heidi D. Klepin et al. Hypertension and use of bevacizumab among patients treated in community settings.. Journal of Clinical Oncology May 20, 2019
Mia Alyce Levy, Travis John Osterman, Neha Jain, Kathleen F Mittendorf, Christine Micheel. Utility of adding clinical data to a molecular results portal for improving clinical trial prescreening efficiency.. Journal of Clinical Oncology May 20, 2017
Junyu Li et al. A software utility for creating interactive maps for 2D gel-based proteomics. Analytical Biochemistry Sep 1, 2004

Selected talks (13)

HL7 International Working Group Meeting (Europe): "mCODE Overview and Current State". May 18, 2026
University of Hawaii, Artificial Intelligence, Precision Health Institute (Virtual): "Predictive AI Models - Data Standards in Action". May 10, 2024
Alzheimer’s Association, Diversity and Disparities PIA: "mCODE Update®". Nov 1, 2023
(Nashville, TN) "From Data Standards to Discovery: Making Clinical Data Available for Research". Sep 19, 2023
NCCN EHR Advisory Board (Virtual): "mCODE® Update". Jul 19, 2023
Indiana University / Regenstrief (Indianapolis, IN): "Standards in Action: Improving Interoperability and Data Access". May 18, 2023
Tennessee Oncology Data Analysts Association (Nashville, TN): "Advancing Lung Cancer Treatment in the Era of Precision Oncology". Oct 7, 2022
mCODE® Community of Practice (Virtual): "Categorization of mCODE® via the FHIR Maturity Model". May 27, 2022
ASPIRE (Durhman, NC): "Growing an International Data Standard". Feb 28, 2025
ASCO Annual Meeting (Chicago, IL): "Interoperability in Action: Progress in Implementing the mCODE™ Oncology Data Standard". Jun 3, 2023
AMIA Annual Symposium (Washington, DC): "mCODE® (minimal Common Oncology Data Elements): a 3-year Update". Nov 8, 2022
NCCN EHR Workgroup (Plymouth, PA): "ASCO’s mCODE™ Initiative: Potential Impact from Member Institution Perspective". Jul 18, 2019
ASCO Annual Meeting (Chicago, IL): "mCODE™ (Minimal Common Oncology Data Elements) in Cancer Practice". Jun 3, 2019

Abstracts (3)

Joseph Vento, Lisa Bastarache, Qingxia M. Chen, Travis Osterman. Real-world side effects of targeted therapies: High-throughput association studies leveraging the CancerLinq Discovery lung cancer database.. Journal of Clinical Oncology May 28, 2025
Karen M. Huelsman et al. Integrating electronic health records (EHRs) to facilitate cancer biomarker testing: Real-world implementation barriers and solutions.. Journal of Clinical Oncology Jun 2024
Zoltan Kiss et al. 1294 External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. Journal for ImmunoTherapy of Cancer Nov 1, 2023

In the news (3)

How mCODE is Driving EHR Interoperability for Cancer Research | TechTarget · Health IT and EHR. Jan 17, 2024
Vanderbilt-Ingram Cancer Center Participating in ASCO’s CancerLinQ · Healthcare Innovation. Feb 21, 2023
CancerLinQ®: Current Achievements and Future Opportunities · ASCO Daily News. Apr 24, 2019

AI in oncology — Travis Osterman

2026-05-28T19:49:00.000Z

AI in oncology

Machine learning and large language models for clinical decision support, efficacy and toxicity prediction, and structured data extraction at scale.

Dr. Osterman's AI work in oncology is grounded in clinical practice. He treats patients, sees what data is actually available, and builds models around that constraint rather than around what would be ideal in a research setting. The work breaks into four threads: a multi-year pre-LLM bet on real-world immunotherapy prediction, an early structured evaluation of clinical large language models, a recent invited framework piece for the field, and ongoing national thought leadership at the policy table.

The first thread is the five-year GE HealthCare Digital Precision Oncology collaboration (2019-2024), which predated the current AI mainstream and was built against two prevailing assumptions in the immune checkpoint inhibitor literature. Most prior ICI-prediction work used carefully curated cohorts; the DPO work used routinely collected EHR data so resulting models could be deployed without first reorganizing the underlying data pipeline. Most prior work focused on a single cancer type; the DPO work was deliberately pan-cancer, accepting the sample-size challenge in exchange for models that generalized. The program produced three peer-reviewed manuscripts, multiple abstracts, twelve conference presentations, and a U.S. patent (Wolber, Csernai, Kiss, Lippenszky, Horváth, Osterman et al.). Dr. Osterman led teams across the U.S., Hungary, and Germany, and presented the work internationally, including the ESMO Immuno-Oncology Annual Congress in Geneva. The full arc is covered in the Digital Precision Oncology case study.

The second thread began in early 2023 when ChatGPT entered the clinic. Dr. Osterman saw the need for an actual evaluation rather than the speculative discourse that dominated the conversation. He designed a multidisciplinary prospective trial structure for evaluating LLM responses to physician-authored questions and supervised Dr. Rachel Goodman (then a medical student) and Dr. Douglas Johnson in executing it. The result, Accuracy and Reliability of Chatbot Responses to Physician Questions (Goodman et al., JAMA Network Open, 2023), was one of the earliest peer-reviewed clinical LLM evaluations and has been widely cited in the policy debates that followed.

In August 2025, Cancer invited Dr. Osterman and colleagues to write Artificial intelligence across the cancer care continuum (Riaz, Khan & Osterman, 2025) - a framework piece mapping AI applications across screening, diagnosis, treatment selection, toxicity prediction, survivorship, and quality of life. The framing is intentional: AI is not one thing in oncology; it is many overlapping technologies at different stages of validation. The review now functions as one of the reference frames the field uses to scope its own work.

Cancer care continuum framework (Riaz, Khan & Osterman, Cancer, 2025).

The work has reach well beyond academic medicine. Dr. Osterman delivered the keynote at the 2025 NCCN AI policy meeting (AI and the Cancer Journey: Navigating New Frontiers in Policy and Technology, September 2025) and was the AI keynote speaker at the 2026 NCCN Annual Conference (Harnessing Artificial Intelligence to Improve Oncology Care, March 2026). NCCN's audience extends beyond physicians to oncology nursing, industry leadership, regulators, and payers - the constituencies that have to act on AI-in-oncology conclusions, not just produce them.

The through-line across these four threads is structure. ICI prediction works because the EHR data underneath is curated and consistent. LLM evaluation matters because that's where unstructured text becomes downstream clinical action. The Cancer review and the NCCN keynotes argue the same point at policy scale: AI's clinical value scales with the quality of the data and standards underneath it. See also cancer data standards (mCODE) and clinical genomics in the EHR for the foundation that makes this work tractable.

Named programs

GE HealthCare Digital Precision Oncology
mCODEGPT zero-shot extraction (Communications Medicine, 2025)
ChatGPT clinical Q&A evaluation (JAMA Network Open, 2023)
Cancer care continuum review (Cancer, 2025)

Case study

This domain has a long-form case study: AI in oncology.

Peer-reviewed publications (12)

Kai Zhang, Tongtong Huang, Bradley A. Malin, Travis Osterman, Qi Long, Xiaoqian Jiang. Introducing mCODEGPT as a zero-shot information extraction from clinical free text data tool for cancer research. Communications Medicine Oct 15, 2025
Irbaz Bin Riaz, Muhammad Ali Khan, Travis J. Osterman. Artificial intelligence across the cancer care continuum. Cancer Aug 15, 2025
David S. Smith et al. Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography. JCO Clinical Cancer Informatics Feb 20, 2025
Levente Lippenszky et al. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clinical Cancer Informatics Mar 21, 2024
Rachel S. Goodman et al. Accuracy and Reliability of Chatbot Responses to Physician Questions. JAMA Network Open Oct 2, 2023
Protiva Rahman et al. Accelerated curation of checkpoint inhibitor-induced colitis cases from electronic health records. JAMIA Open Apr 1, 2023
Rachel S. Goodman, J. Randall Patrinely, Travis Osterman, Lee Wheless, Douglas B. Johnson. On the cusp: Considering the impact of artificial intelligence language models in healthcare. Med (New York, N.Y.) Mar 10, 2023
Douglas Johnson et al. Assessing the Accuracy and Reliability of AI-Generated Medical Responses: An Evaluation of the Chat-GPT Model (under review). Feb 28, 2023
Kim L Sandler et al. Women screened for breast cancer are dying from lung cancer: An opportunity to improve lung cancer screening in a mammography population. Journal of Medical Screening May 4, 2021
Neha M. Jain, Alison Culley, Christine M. Micheel, Travis J. Osterman, Mia A. Levy. Learnings From Precision Clinical Trial Matching for Oncology Patients Who Received NGS Testing. JCO Clinical Cancer Informatics Feb 1, 2021
Neha M. Jain, Alison Culley, Teresa Knoop, Christine Micheel, Travis Osterman, Mia Levy. Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow. JCO Clinical Cancer Informatics Jun 21, 2019
Neha M Jain, Alison Culley, Travis John Osterman, Mia Alyce Levy. Learnings from a pragmatic study to evaluate benefit of performing reflex clinical trial matching and providing clinical decision support to physicians.. Journal of Clinical Oncology May 20, 2019

Selected talks (12)

Vanderbilt Lecture Series CME (Nashville, Tennessee): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". May 7, 2026
2026 NCCN Annual Conference (Orlando, Florida): "Harnessing Artificial Intelligence to Improve Oncology Care". Mar 28, 2026
NASEM Workshop: Policy Issues for Integrating Artificial Intelligence in Cancer Research and Care (Washington, DC): "AI in Cancer Care: 2 Wins, 2 Current Challenges". Mar 9, 2026
Jackson-Madison County General Hospital (Jackson, Tennessee): "An Update on the Use of AI in Clinical Practice". Mar 6, 2026
Blanchfield Army Community Hospital (Fort Campbell, Kentucky (virtual)): "Using AI in Clinical Practice: Current Trends and Emerging Federal Regulations". Feb 10, 2026
NCCN EHR Advisory Board (Virtual): "Clinical Trial Enrollment A Pragmatic Approach". Nov 8, 2024
University of Hawaii, Artificial Intelligence, Precision Health Institute (Virtual): "Predictive AI Models - Data Standards in Action". May 10, 2024
The Evolving Artificial Intelligence Landscape in Cancer Care: "AI and the Cancer Journey: Navigating New Frontiers in Policy and Technology". Sep 9, 2025
NICT (Durhman, NC): "Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data". Mar 1, 2025
ESMO Immuno-Oncology Annual Congress (Geneva, Switzerland): "Prediction of immune checkpoint inhibitor outcome and side effects by use of machine learning and routine real-world data". Dec 11, 2024
Society for Immunotherapy of Cancer Annual Meeting (San Diego, CA): "EHR-based Models for Predicting Efficacy and Toxicities Prior to ICI Treatment". Nov 4, 2023
IO360 (Brooklyn, NY): "Predicting Efficacy and Toxicities Prior to Immune Checkpoint Inhibitor Treatment". Feb 9, 2023

Abstracts (8)

Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy | Journal for ImmunoTherapy of Cancer. Nov 5, 2024
David Smith et al. 1246 Prediction of pneumonitis in immunotherapy patients from prior thorax CT. Journal for ImmunoTherapy of Cancer Nov 1, 2024
Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy. Journal for ImmunoTherapy of Cancer Nov 1, 2024
Zoltan Kiss et al. 1294 External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Levente Lippenszky et al. 1300 Prediction of efficacy and toxicities of immune checkpoint inhibitors using real-world patient data. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Eszter Csernai et al. Rolling window-based hepatitis toxicity prediction from routine bloodwork in patients undergoing immune checkpoint inhibitor therapy.. Journal of Clinical Oncology Jun 2022
Gergely Horváth et al. Predicting immune checkpoint inhibitor-related hepatitis using electronic health records of patients.. Journal of Clinical Oncology Jun 2022
Levente Lippenszky et al. Predicting immune checkpoint inhibitor-related pneumonitis using patient medical information.. Journal of Clinical Oncology Jun 2022

In the news (7)

AI could predict whether cancer treatments will work, experts say | Fox News . Apr 23, 2024
GE Healthcare, Vanderbilt Explore Use of AI to Predict Immunotherapy Toxicity, Efficacy · Precision Medicine Online. Nov 27, 2023
The immunotherapy hurdle – and why doctors could soon predict how each patient will respond . Apr 9, 2019
GE and VUMC partner to make cancer immunotherapy safer and more precise · dotmed.com. Jan 8, 2019
Navigating AI Use for Improving Oncology Care—Challenges, Regulation, and Opportunities · ASCO AI in Oncology.
Machine Learning Model Could Help Predict Risk-Benefit of Immune Checkpoint Inhibitors · ASCO Daily News.
Podcast: New Machine Learning Framework Uses EHR Data to Assess ICI Effectiveness, Toxicity · ASCO Daily News.

NCCN Collaboration — Travis Osterman

2026-05-28T19:49:00.000Z

NCCN Digital Oncology Forum and AI thought leadership, ongoing

Member of the NCCN Digital Oncology Forum and an active voice across NCCN policy and clinical channels. Featured AI keynote speaker at the 2026 NCCN Annual Conference and the NCCN AI policy summit.

Dr. Osterman is a member of the NCCN Digital Oncology Forum and an active voice across NCCN policy and clinical channels. The most visible recent work has been NCCN's AI-in-oncology agenda. In September 2025, Dr. Osterman delivered the keynote at the NCCN AI policy meeting (AI and the Cancer Journey: Navigating New Frontiers in Policy and Technology). In March 2026 he was the AI keynote speaker at the NCCN Annual Conference (Harnessing Artificial Intelligence to Improve Oncology Care) - reaching an audience that includes not only physicians but oncology nursing, industry leadership, regulators, and payers. The framing of regulation, integration, and clinical impact was published in JNCCN and featured in Oncodaily. The full arc lives in the AI in oncology case study.

Named programs

NCCN Digital Oncology Forum - member
AI keynote speaker, 2026 NCCN Annual Conference
AI policy keynote, 2025 NCCN AI summit

Case study

Long-form narrative: AI in oncology: from ChatGPT in the clinic to mapping the care continuum.

Peer-reviewed publications (5)

Waddah Arafat et al. Clinician Perspectives Regarding the Impact of Information Technology on Multidisciplinary Tumor Boards: A National Comprehensive Cancer Network Survey. JCO Clinical Cancer Informatics Nov 9, 2023
Eric M. Lander et al. Identification and Characterization of Avoidable Hospital Admissions in Patients With Lung Cancer. Journal of the National Comprehensive Cancer Network Oct 1, 2023
Lucy R. Langer, Amye Tevaarwerk, Robin Zon, Travis Osterman. The Future of Telemedicine in Oncology. Journal of the National Comprehensive Cancer Network Jun 2, 2022
Peter D. Stetson et al. Adoption of Patient-Generated Health Data in Oncology: A Report From the NCCN EHR Oncology Advisory Group. Journal of the National Comprehensive Cancer Network Jan 18, 2022
Amye J. Tevaarwerk et al. Oncologist Perspectives on Telemedicine for Patients With Cancer: A National Comprehensive Cancer Network Survey. JCO Oncology Practice Nov 15, 2021

Selected talks (9)

2026 NCCN Annual Conference (Orlando, Florida): "Harnessing Artificial Intelligence to Improve Oncology Care". Mar 28, 2026
NCCN EHR Advisory Board (Virtual): "Clinical Trial Enrollment A Pragmatic Approach". Nov 8, 2024
NCCN EHR Oncology Advisory Group (Pymouth, PA): "Digital Transformation". Jul 27, 2023
NCCN EHR Advisory Board (Virtual): "mCODE® Update". Jul 19, 2023
NCCN EHR Advisory Group (Virtual): "Leveraging the EHR to facilitate actionable research data". Jan 20, 2023
NCCN Annual Congress (Virtual): "Improving Clinical Trial Accrual: Doing More with Less". Jun 15, 2025
NCCN Annual Conference: "The Future of Telemedicine in Oncology". Mar 31, 2022
NCCN Annual Conference (Virtual): "The Future of Telemedicine in Oncology". Mar 21, 2022
NCCN EHR Workgroup (Plymouth, PA): "ASCO’s mCODE™ Initiative: Potential Impact from Member Institution Perspective". Jul 18, 2019

Abstracts (5)

Joseph Vento, Travis Osterman. BIO23-019: Precision Oncology: Integrating Structured Genomic Data Into the Electronic Health Record via the EPIC® Genomics Module. Journal of the National Comprehensive Cancer Network Mar 31, 2023
Waddah Arafat et al. Oncologist perspectives on tumor boards: Virtual meetings and EHR integration.. Journal of Clinical Oncology Oct 2022
Alex C. Cheng et al. Oncologist participation in pilot testing a crowdsourcing platform to build a survivorship care risk model.. Journal of Clinical Oncology Jun 2022
Neha M Jain, Emma Schremp, Lucy Spalluto, Travis John Osterman, Debra L. Friedman. Incorporating mediation-based interventions at an academic cancer center: A six-step process.. Journal of Clinical Oncology Jun 2022
Amye Tevaarwerk et al. BIO21-011: Oncology Provider Perspectives on Telemedicine for Patients With Cancer: A National Comprehensive Cancer Network (NCCN®) Survey. Journal of the National Comprehensive Cancer Network Mar 17, 2021

Disclosures

This collaboration is publicly disclosed through two independent records: ASCO COI and CMS Open Payments.

Related: all collaborations · GE HealthCare · Epic · Microsoft · Tempus AI · nference · ai in oncology.

nference Collaboration — Travis Osterman

2026-05-28T19:49:00.000Z

nference Research-data infrastructure partnership, 2023-present

Vanderbilt Health's partnership with nference reduces the friction of accessing clinical data for research while preserving privacy and regulatory compliance. Formalized in a 2023 agreement.

Dr. Osterman leads Vanderbilt Health's partnership with nference, formalized in a 2023 agreement (VUMC News, July 2023) to advance real-world evidence generation in complex disease populations. The collaboration reduces the friction of accessing clinical data for research while preserving privacy and regulatory compliance - extending the broader structured-data agenda described in clinical genomics in the EHR.

Named programs

Vanderbilt Health-nference research-data partnership (2023 agreement)
Real-world evidence generation in complex disease populations

Peer-reviewed publications (1)

Elise Russo et al. Vanderbilt Clinical Informatics Center Education Strategy: To Infinity and Beyond!. Applied Clinical Informatics Feb 26, 2025

Selected talks (4)

2026 NCCN Annual Conference (Orlando, Florida): "Harnessing Artificial Intelligence to Improve Oncology Care". Mar 28, 2026
NCCN Annual Conference: "The Future of Telemedicine in Oncology". Mar 31, 2022
NCCN Annual Conference (Virtual): "The Future of Telemedicine in Oncology". Mar 21, 2022
28. NLM Informatics Training Conference (Columbus, OH): "EHR-Wide GxE Study using Smoking Information Extracted from Clinical Notes,”". Jun 29, 2016

Abstracts (2)

Sang Minh Nguyen et al. Abstract LB385: Polygenic risk score of genetic variants in genes encoding drug-metabolizing enzymes and drug transporters, in association with febrile neutropenia. Cancer Research Apr 17, 2026
Travis J. Osterman. EHR-Wide GxE Study using Smoking Information Extracted from Clinical Notes. Jun 29, 2016

Disclosures

This collaboration is publicly disclosed through two independent records: ASCO COI and CMS Open Payments.

Related: all collaborations · GE HealthCare · Epic · Microsoft · Tempus AI · NCCN · clinical genomics ehr.

Tempus AI Collaboration — Travis Osterman

2026-05-28T19:49:00.000Z

Tempus AI Multi-year precision-oncology platform partnership, ongoing

Multi-year partnership with Tempus AI spanning three workstreams: structured electronic integration of genomic results into clinical workflows, digital pathology, and multi-omic discovery infrastructure.

Vanderbilt's research collaboration with Tempus AI spans three workstreams: structured electronic integration of genomic results into Vanderbilt's clinical workflows, digital pathology, and multi-omic discovery infrastructure. The reference-laboratory integration was profiled by VUMC News in November 2024 as part of a broader effort to make reference-laboratory data first-class within the electronic health record - the same body of work that put Vanderbilt's structured-genomic-data EHR ahead of any other U.S. institution (see clinical genomics in the EHR).

Named programs

Structured electronic integration of genomic reference-lab results into Vanderbilt's EHR
Digital pathology collaboration
Multi-omic discovery infrastructure

Selected talks (1)

25. Tempus Webinar (Virtual): "Integrating Structured Genomic Data in Clinic". May 12, 2022

Disclosures

This collaboration is publicly disclosed through two independent records: ASCO COI and CMS Open Payments.

Microsoft Collaboration — Travis Osterman

2026-05-28T19:49:00.000Z

Microsoft Investigator Fellowship and Azure research collaboration, 2020-present

Microsoft Investigator Fellow (2020) - a two-year, $200,000 award for scaling clinical informatics education across UME and GME via Azure-based virtualization. Continuing service on the Microsoft Azure Research Community Advisory Board on structured genomic data, FHIR research infrastructure, and the mCODE data standard.

Dr. Osterman was selected as a Microsoft Investigator Fellow in 2020 - a two-year, $200,000 award - for work leveraging Azure-based virtualization to scale clinical informatics education across undergraduate and graduate medical training. The fellowship funded the cloud infrastructure that lets learners build and test against realistic clinical data and systems without the licensing and security overhead that typically makes hands-on informatics education impractical. The infrastructure is now part of the Vanderbilt CI Integrated Science Curriculum and the GME Clinical Informatics Elective Rotation (see CI education).

Beyond the fellowship, Dr. Osterman serves on the Microsoft Azure Research Community Advisory Board (AMC CAB), contributing on structured genomic data, FHIR-based research infrastructure, and the minimal Common Oncology Data Elements (mCODE™) standard. The mCODE Genomics Pilot at Vanderbilt (Li et al., JCO CCI, 2024) was built on Azure, demonstrating an end-to-end FHIR-on-Azure pipeline lifting structured genomic data out of Epic into mCODE-conformant resources.

Named programs

Microsoft Investigator Fellowship (2020) - $200K, 2 years, PI
Microsoft Azure Research Community Advisory Board (AMC CAB) - member
Azure-based virtualization for CI education (UME + GME)
Azure-based FHIR research infrastructure for mCODE workflows

Case study

Long-form narrative: mCODE: from a data standard to regulatory infrastructure.

Peer-reviewed publications (1)

Yanwei Li et al. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics Jun 28, 2024

In the news (2)

Microsoft Investigator Fellow Dr. Travis Osterman uses Azure to support lung cancer treatment protocols · Microsoft Customers Stories. Jul 23, 2021
VUMC’s Osterman awarded Microsoft Investigator Fellowship · VUMC News. Nov 7, 2019

Disclosures

This collaboration is publicly disclosed through two independent records: ASCO COI and CMS Open Payments.

Epic Collaboration — Travis Osterman

2026-05-28T19:49:00.000Z

Epic Multi-year, multi-level EHR collaboration, ongoing

Epic is the EHR platform anchoring Dr. Osterman's clinical informatics work. Multiple collaboration surfaces: the Adult Oncology Steering Committee, the Beacon Community Operations Group (founder), Epic Genomics Module, Epic AURA, and the AI-extracted oncology staging workflow in Epic Hyperspace.

Epic is the EHR platform that anchors most of Dr. Osterman's clinical informatics work, and the collaboration with the company takes several distinct shapes: governance, build, training, and research.

Governance. Dr. Osterman serves on Epic's Adult Oncology Steering Committee, the body that influences the direction of Epic's oncology-specific features. He also founded the Epic Beacon Community Operations Group (BCOG), a multi-institution governance body for the dominant U.S. oncology EHR module - the group of academic cancer centers that coordinate on shared challenges and shared builds across the Beacon platform.

Build and integration. Three implementation pieces have shipped at Vanderbilt under Dr. Osterman's leadership: the Epic Genomics Module as the structured-variant home inside the EHR; Epic AURA as the bridge that pulls reference-laboratory genomic reports in over standardized interfaces; and the AI-extracted oncology staging workflow being built into Epic Hyperspace (Dr. Osterman is PI). The implementation work is documented in Li et al., JCO CCI, 2024 (mCODE Genomics Pilot) and the Vento & Osterman NCCN abstract (2023) on the upstream workflow.

Speaking and community. Featured speaker at Epic's annual User Group Meeting (UGM) and Expert Group Meeting (XGM) on community governance, structured genomic data backloading, in-basket workflow, and other topics where the BCOG cross-institution conversations have generated reusable patterns. The 2022 UGM talk Precision Medicine: Using Structured Genomic Data in Clinical Decision Support is the canonical public version of the Vanderbilt build story.

Named programs

Epic Adult Oncology Steering Committee
Epic Beacon Community Operations Group (BCOG) - Founder
Epic Genomics Module rollout at Vanderbilt
Epic AURA reference-laboratory integration
AI-extracted oncology staging in Epic Hyperspace (PI: Osterman)

Case study

Long-form narrative: mCODE: from a data standard to regulatory infrastructure.

Peer-reviewed publications (1)

Yanwei Li et al. Minimal Common Oncology Data Elements Genomics Pilot Project: Enhancing Oncology Research Through Electronic Health Record Interoperability at Vanderbilt University Medical Center. JCO Clinical Cancer Informatics Jun 28, 2024

Selected talks (4)

Beacon Community Operations Group (BCOG), (Virtual): "Maximizing Value with Structured Genomic Data". Jul 8, 2022
Epic: Beacon Community Oncology Group (BCOG) (Virtual): "Auto-Populate Care Team with Encounter Provider". May 6, 2020
(Virtual) "Document Link in eStar". Apr 10, 2020
Epic User Group Meeting (UGM): "Improving In Basket User Experience: Leveraging Governance, Build, Development, and Collaboration". Aug 28, 2019

Abstracts (1)

Joseph Vento, Travis Osterman. BIO23-019: Precision Oncology: Integrating Structured Genomic Data Into the Electronic Health Record via the EPIC® Genomics Module. Journal of the National Comprehensive Cancer Network Mar 31, 2023

In the news (2)

Vanderbilt Preparing to Implement Epic Genomics Module · Healthcare Innovation. Mar 4, 2021
One year after Go Live, focus remains on advancing eStar · VUMC Reporter. Nov 15, 2018

Disclosures

This collaboration is publicly disclosed through two independent records: ASCO COI and CMS Open Payments.

GE HealthCare Collaboration — Travis Osterman

2026-05-28T19:49:00.000Z

GE HealthCare Strategic research collaboration, 2019-2024

Five-year strategic research collaboration on predicting immune checkpoint inhibitor efficacy and toxicity from real-world EHR data. Dr. Osterman served as principal investigator on the flagship Digital Precision Oncology study.

Vanderbilt and GE HealthCare announced a five-year strategic research collaboration in January 2019 (dotmed.com) centered on applying machine learning to predict immune checkpoint inhibitor (ICI) effectiveness and toxicity from real-world EHR data. Dr. Osterman served as principal investigator from Vanderbilt; Dr. Jan Wolber led on the GE HealthCare side. Multi-disciplinary teams on both sides spanned medical oncology, radiochemistry, biostatistics, bioinformatics, and machine learning. The collaboration ran through 2024 and produced three peer-reviewed manuscripts, twelve conference presentations, multiple abstracts, and a U.S. patent on the underlying model-generation framework.

The flagship paper - Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data (Lippenszky et al., JCO Clinical Cancer Informatics, 2024) - demonstrated that routinely-collected EHR data is enough to build clinically useful prediction models across both efficacy and toxicity, without bespoke biomarker panels or unstructured-data extraction pipelines. The imaging companion paper extended the framework to pre-treatment thorax CT (Smith et al., JCO CCI, 2025) using radiomic features and deep learning to predict ICI-induced pneumonitis from 2,700 CT volumes.

Beyond Vanderbilt the work was externally validated on multi-center German pan-cancer cohorts in collaboration with the Essen-Bochum team (Kiss, Lippenszky et al., SITC 2023). The full narrative arc - what the team bet against the prevailing assumptions of the field (real-world data over curated cohorts, pan-cancer over single-cancer), what shipped, and what's next - lives in the Digital Precision Oncology case study.

Named programs

Digital Precision Oncology study (PI: Osterman) - GE HealthCare flagship
ICI hepatitis, colitis, and pneumonitis prediction models
Pre-treatment CT radiomics for pneumonitis prediction
External validation across multi-center pan-cancer cohorts (Germany)

Case study

Long-form narrative: Digital Precision Oncology with GE HealthCare.

Peer-reviewed publications (3)

David S. Smith et al. Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography. JCO Clinical Cancer Informatics Feb 20, 2025
Levente Lippenszky et al. Prediction of Effectiveness and Toxicities of Immune Checkpoint Inhibitors Using Real-World Patient Data. JCO Clinical Cancer Informatics Mar 21, 2024
Protiva Rahman et al. Accelerated curation of checkpoint inhibitor-induced colitis cases from electronic health records. JAMIA Open Apr 1, 2023

Selected talks (1)

Society for Immunotherapy of Cancer Annual Meeting (San Diego, CA): "EHR-based Models for Predicting Efficacy and Toxicities Prior to ICI Treatment". Nov 4, 2023

Abstracts (10)

Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy | Journal for ImmunoTherapy of Cancer. Nov 5, 2024
David Smith et al. 1246 Prediction of pneumonitis in immunotherapy patients from prior thorax CT. Journal for ImmunoTherapy of Cancer Nov 1, 2024
Pablo Napan Molina et al. 1228 Machine learning models can predict efficacy and toxicities using short medical history prior to ICI therapy. Journal for ImmunoTherapy of Cancer Nov 1, 2024
Zoltan Kiss et al. 1294 External validation of machine learning models to predict efficacy and toxicity of immune checkpoint inhibitors using real-world pan cancer cohorts. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Levente Lippenszky et al. 1300 Prediction of efficacy and toxicities of immune checkpoint inhibitors using real-world patient data. Journal for ImmunoTherapy of Cancer Nov 1, 2023
Eszter Csernai et al. Rolling window-based hepatitis toxicity prediction from routine bloodwork in patients undergoing immune checkpoint inhibitor therapy.. Journal of Clinical Oncology Jun 2022
Gergely Horváth et al. Predicting immune checkpoint inhibitor-related hepatitis using electronic health records of patients.. Journal of Clinical Oncology Jun 2022
Levente Lippenszky et al. Predicting immune checkpoint inhibitor-related pneumonitis using patient medical information.. Journal of Clinical Oncology Jun 2022
Neha M Jain, Philip Edward Lammers, Michael R. Savona, Travis John Osterman, Salil Goorha. Using a standard implementation science framework to improve clinical trial enrollment for a community Tennessee oncology center.. Journal of Clinical Oncology Jun 2022
Kathleen F. Mittendorf et al. Overcoming barriers in academic-industry partnerships to improve predictive modeling in immuno-oncology.. Journal of Clinical Oncology Jun 2022

Patents (1)

Jan Wolber, Eszter Katalin Csernai, Zoltán Kiss, Levente Lippenszky, Gergely Horváth, Travis Osterman, Ben Ho Park, David Samuel Smith, Daniel Fabbri, Michele LeNoue-Newton, Kathleen Mittendorf. Model generation apparatus for therapeutic prediction and associated methods and models (2025).

In the news (1)

GE Healthcare, Vanderbilt Explore Use of AI to Predict Immunotherapy Toxicity, Efficacy · Precision Medicine Online. Nov 27, 2023

Disclosures

This collaboration is publicly disclosed through two independent records: ASCO COI and CMS Open Payments.

Related: all collaborations · Epic · Microsoft · Tempus AI · nference · NCCN · ai in oncology · lung cancer.

Christmas archive

2026-05-26T20:48:05.000Z

Christmas archive

Annual Osterman family Christmas pages and newsletters. The current year is open; older years are gated.

2025-11-25

Christmas 2025
2025-11-25

Christmas 2025 Newsletter
2024-12-08 · private

Christmas 2024 Newsletter
2024-12-07 · private

Christmas 2024
2023-12-05 · private

Christmas 2023
2023-12-05 · private

Christmas 2023 Newsletter
2022-12-10 · private

Christmas 2022
2022-12-10 · private

Christmas 2022 Newsletter
2021-12-04 · private

Christmas 2021
2021-12-04 · private

Christmas 2021 Newsletter
2020-11-15 · private

Christmas 2020
2020-11-15 · private

Christmas 2020 Newsletter
2019-12-08 · private

Christmas 2019
2019-12-08 · private

Christmas 2019 Newsletter
2018-12-18 · private

Christmas 2018
2018-12-18 · private

Christmas 2018 Newsletter
2018-07-22 · private

owen-2018
2017-12-09 · private

Christmas 2017 Newsletter
2017-11-24 · private

Christmas 2017
2016-12-21 · private

Christmas 2016
2016-12-20 · private

Christmas 2016 Newsletter
2015-12-16 · private

Christmas 2015 Newsletter
2015-12-15 · private

Christmas 2015
2014-12-17 · private

Christmas 2014
2013-12-13 · private

Christmas 2013

Christmas 2025

2025-11-25T00:00:00.000Z

Welcome to our Christmas 2025 update. It has been another exciting year. Be sure to check out our 2025 newsletter for details.

Christmas 2025 Newsletter

2025-11-25T00:00:00.000Z

Willow (7)

Willow is a busy second-grader. In addition to school, she has enjoyed dance this year with lots of new friends. She continues to be the artist and crafter of the family and can often be found surrounded by markers, glue, tape, glitter, pipe cleaners, and piles of scrap paper.

Gavin (9)

Gavin is now in third grade. His highlights include becoming an expert modded Minecraft player. He loves exploring new worlds with his family, learning all the winding questlines, and gathering chests filled with the best loot. He continues to be active in scouts and swimming.

Owen (13)

Owen is loving seventh grade with highlights of joining the Brentwood Middle School Student Government Association and making a Southeastern cut time in swimming. He has enjoyed bringing his unique brand of junior high humor to the dinner table.

Laura

Laura has enjoyed immersing herself in the new world of 3d printing. She also coordinated getting all the McEwens to a Blue Jays game during their historic 2025 season. She continues to be an active volunteer at the schools and in extracurricular activities. Finally, she led the campaign to add a new member to the family (pictured above).

Travis

This has been a hard and reflective year for Travis. Brenda Osterman, his mother, died. She was an amazing, kind, and giving mother, sister, wife, and grandmother who fought Parkinson's Disease for twenty-five years. She was a testament to seizing each day and an inspiration to those around her, as reflected in the outpouring of family, friends, and community at her memorial services. Life continues but differently.

Christmas Present

Christmas 2025 photos

Christmas Past

Christmas 2024 photos newsletter
Christmas 2023 photos newsletter
Christmas 2022 photos newsletter
Christmas 2021 photos newsletter
Christmas 2020 photos newsletter
Christmas 2019 photos newsletter
Christmas 2018 photos newsletter
Christmas 2017 photos newsletter
Christmas 2016 photos newsletter
Christmas 2015 photos newsletter
Christmas 2014
Christmas 2013

Favorite Christmas Story this Year

2017-12-15T00:00:00.000Z

Dr. William Weaver is the Chief of Surgery at the Fayetteville VA Medical Center in North Carolina. He has had two recent stories covered by NPR. In the first, he recounts being one of 14 black students integrating the all-white West High School in Knoxville, TN. I heard the second story today during my commute and was moved by his Story Corps recording recounting the theft of his brother's bike during Christmas when he was a child. This story of family, giving, and forgiveness comes at a great time. Thank you NPR and thank you Dr. Weaver for sharing.

Good Enough Dad

2013-03-25T00:00:00.000Z

A 'Good Enough' Dad And His Special Son

Tim Harris (right) and his father, Keith, visited StoryCorps in their hometown of Albuquerque, N.M.

National Public Radio (NPR) runs a series called Story Corps which is simply recordings of people interviewing people. The link at the top is probably my favourite interview I've heard all year, in any context. Tim is a college graduate with Down's Syndrome and is owner of Tim's Place in Albuquerque, New Mexico. The interview is mostly a father interviewing his son. If you're a father or son or both, the interview is three minutes and thirteen seconds long and you'll probably get less than sixty seconds in before tearing up.

Hydroponics - Cost To Get Started with Ebb and Flow Hydroponics

2012-10-29T00:00:00.000Z

To get started with hydroponics at home, you should take an inventory of your goals and budget to begin. For a 48x24 inch ebb and flow (flood and drain) system, Your total cost will be around $600. If you grow potatoes, it will take quire a while to recoup this cost. Basil, however, at $3 for a young, fresh plant for pesto, will probably return your investment in as few as 2 years (6 inch diameter per plant = 8x4 plants in table = 32 plants x $3/plant = $96/harvest / 3 months maturation = $32/month - electricity - more nutrient - seeds = ~$25/month). If you choose to use all organic and can find a buyer, expect a much more lucrative return.

Ebb and Flow Hydroponics Overview

Here a a small diagram of the setup that will be described below.

If you don't have $600 laying around, you can step into hydroponics more slowly by starting with a deep water culture system or smaller ebb and flow setup. I'll argue though that planning a setup this size and buying it incrementally will make the most sense. We'll breakdown the various components here.

Hydroponics Light Source

The Amazon product widgets originally embedded here no longer render. Product links: T5 grow light fixtures (1, 2).

Starting with the light source makes the most sense. Begin by purchasing the most lumens per square foot that you can afford. Note that will not necessary be the largest fixture. A 2x2 ft fixture that packs in many bulbs is just as good. Direct sunlight is around 9000-10,000 lumens per square foot (or 100,000 lumens/sq meter) so that is our target.

The fixture show above contains 8 T5 bulbs that will produce 40k lumens each or 320K lumens total over an area of approximately 4x2 feet (8 sqft). That makes 40,000 lumens per square foot. After accounting for inefficiency in reflection for light that is initially directed up, light that is lost horizontally, and decreased efficacy of the bulb as it ages, we should still be around the 10k/sqft sunlight equivalent. We'll ignore wavelengths for now and say that T5 is a good bulb choice because they are cool enough to be very close to the plant canopy.

You can germinate plants with much less light. This amount of light will allow your plants to thrive, bloom, and fruit however. Once you have this set up, try it for a few months on a traditional potted plant and you'll be sold.

Total: $200

Hydroponic Grow Media and Flood Table

The Amazon product widgets originally embedded here no longer render. Product links: flood table and grow media (1, 2, 3).

The next large investment in your ebb and flow hydroponics setup will be a flood table, reservoir and grow medium. From the diagram above, hopefully you can get the idea that your reservoir can be just about anything that holds water. I scoped out craigslist for a few months as was able to find a fifty-five gallon aquarium for $50 that holds water just fine.

The flood table will either need to be supported by the reservoir or by some sort of stand. Since the flood table will contain grow media which will displace some of the water, the volume to flood the the flood table will be less (maybe 50% less) that calculated volume. Assuming that the reservoir is covered to minimize evaporation loss, You can begin to estimate the maximal amount of time between filling the reservoir. This is also known as how long you can go on vacation.

A 48 x 24 x 6 flood table is 6912 cubic inches or just under 30 gallons. If we assume 50% displacement but grow media, we'll need 15 gallons to flood the table. Let us add in 2 gallons of water to keep the pump submerged and the minimum water requirement would be 17 gallons. Since our reservoir is 55 gallons, if we fill it completely, will will have 38 gallons of "extra" water that we will either loose to our plants or evaporation.

Different plants have different needs for water. Lettuce is notoriously good for large scale hydroponics as it sucks a large amount of water. This isn't to say that you can't raise hydroponic cactus though. Hydroponics Made Easy claims a young tomato plant will use 300ml-1L of water per day. We would have 96 plants in a full bed or 28.8-96L/day (7.6-25gal/day). So this system probably has between a 1-7 refill period depending on plant type, maturity, and density.

The hydroponic grow media simply serves a means to keep roots moist but not wet (like a sponge. Hydroton is a commercially available clay pellet that is commonly used.

Total: $200

Ebb and Flow Water Pump System

The Amazon product widgets originally embedded here no longer render. Product links: water pump and timer (1, 2).

The water pump is the key component of the ebb and flow (flood and drain) hydroponic system. The goal is to have a pump capable of raising an appropriate volume per minute to flood your table relatively quickly but not so fast as to cause erosion of your grow media. The higher your flood table is from the base of your reservoir and the larger your flood table, the larger the pump your need. The larger the pump, the (generally) louder it will be and more electricity it will consume; although nowhere near the consumption of your grow lights.

The water pump will be on a timer to flood your plants several times daily while your grow lights are on.

Total: $50

Hydroponic Nutrients

The Amazon product widgets originally embedded here no longer render. Product link: hydroponic nutrients (a second product is no longer listed).

Plants cannot live on water alone. The nutrients, associated pH solutions, water, and electricity will be the main recurring expenses.

Total: $50

Hydroponics Aeration

The Amazon product widgets originally embedded here no longer render. Product links: air pump, air stones, and check valve (1, 2, 3, 4).

You can put off aeration to the end but you will not want to put this off too long. like your plants, algae will have all the ingredients to grow. Changing your water at regular intervals and keeping your water aerated will help combat this thread.

Remember that air stones will help get more of the gas dissolved into the water so plan on picking up a few air stones. Also, a one-way check valve is a good idea if your pump is not submersible as there is always a chance of backflow of water when the system is turned off.

Aeration, unlike lighting and the water pump can continue 24 hours a day. Make sure that your air outlet are not is the same area as your water pump intake as this will decrease efficiency and life of your water pump.

Total: $30

Hydroponics Conclusion

There are the rough estimates for getting started with a moderate-sized hydroponics system at home. This hydroponics setup will grow, blood, and fruit anything from tropical trees to cactus if you tweak the solutions and times of flooding. Happy harvesting.

Ender's Game - Audiobook for the Family

2012-10-22T00:00:00.000Z

September and October have now become the months that one interviews for the internal medicine sub-specialties. For years I have been a fan of using my daily commute and extended travel times to "read" audiobooks or listen to medical lectures. I have been on the road over the past two months and decided with my last trip out of state to break up my MKSAP study with a leisure novel - Ender's Game. This will not be a book review. I am neither well-read enough to make insightful comparisons nor well written enough to write a critique to far outside my comfort zone of technical writing. One idea did come to me as I sat in the parking lot after arriving at work, listening to the final ten minutes of Orson Scott Card's commentary. He emphasizes that this (the audiobook) is the medium that he would most like his stories to be shared.

The Amazon product widget originally embedded here no longer renders. The product: Ender’s Game audiobook on Amazon.

Family Audiobook Trip

What do your family road trips look like? I realized instead of drop-down dvd screens or mp3 players loaded with the latest pop superstars, maybe I will, at least try, listen to audiobooks with my family. How great to have the opportunity to pause the story so that your children can share their insights or questions. What better way to introduce literary analysis and fundamental concepts of simile, metaphor, comparison, and antagonism. Why not make that road trip as productive for my family as I make my daily commute or interview trips? One faculty member asked recently whether I did anything that wasn't purposeful after reviewing my hobbies. Interesting that is my most recent attempt at a purely leisure reading experience, I may have altered one vision I have for my family. I love surprises like that.

Ender's Game

As promised, this is not a book review. I do owe Card a brief analysis on whether the audio novel would be appropriate for a preteen family road trip. Given that the main characters are all preteen for the majority of the story, the language is pristine, and the dialogue is clear, I think that a first grade through junior high audience will be able to follow along without difficulty. For the adults, the characters are well-developed and plot line is engaging enough that you'll reach for the next cd instead of hitting scan to find the local fm mix station that no one like but everyone can tolerate.

Stand Up To Cancer Using Oncology Informatics

2012-10-15T00:00:00.000Z

Stand Up 2 Cancer has several dream teams organized to help oncology research. Today, I came across an article from UCSC discussing how one of their associate professors in biomolecular engineering, Joshua Stuart (right), is joining as a member of the prostate cancer dream team. It looks like Stuart is already active in the breast cancer dream team.

From the article:

A second component of the UCSC team's work will be to develop a novel application called MedBook to connect patients, doctors, and researchers. Inspired by social media, MedBook is the brainchild of UCSC graduate student, Foundation trustee, and former Apple executive Ted Goldstein, who has described it as a "Facebook for cancer." The idea is to bring together the findings from patient tumor samples and connect researchers in a way that will help them work collaboratively to discover new models of disease and match patients with targeted treatments.

"The precision prostate oncology MedBook will act like an iPad app, linking patients, doctors, and researchers in a participatory information platform that will speed the matching of new medicines targeted at new druggable pathways to the patients who will benefit most," said Jonathan Simons, M.D., president and CEO of the Prostate Cancer Foundation.

A few other informaticists are already participating in the dream team projects:

Arul M. Chinnaiyan, M.D., Ph.D. Director of Pathology Research Informatics Director of Cancer Bioinformatics The Regents of the University of Michigan

David Cherba, Ph.D. Van Andel Research Institute (VARI)

Jason Corneveaux Translational Genomic Research Institute (TGen)

This is probably an under-representation of the importance that informatics plays in oncology, but we are getting there. Kudos to everyone providing support as we create tools to help us ... Stand Up 2 Cancer.

Stanford Oncology Informatics

2012-10-01T00:00:00.000Z

Last week was a busy week for The Cardinal. Two days after Dr. Robert Carlson was announced as the new CEO of the National Comprehensive Cancer Network (NCCN), the Indianapolis Star announced that Dr. George Sledge will be leaving Indiana University to become the division chief of hematology and oncology at Stanford.

Both distinguished clinician researchers have a history of breast cancer focus with academic interests at the interface of medical oncology and informatics (oncology informatics). Dr. Carlson has interests in:

Computer science research efforts focus on computer-based systems for patient care. These include: the development of expert systems to assist in the care of patients with cancer (ONCOCIN) or AIDS (T-Helper); the development of computer-based systems using speech and pen-based input for medical records to access sources of medical knowledge; the design of three-dimensional methods of performing and representing literature searches; and the development of standards for implementing of medical computer-based systems; the development of smart agents and medical ontologies.

Dr. Sledge was the ASCO 2011 president and used the opportunity of his presidential address to discuss The Challenge and Promise of the Genomic Era. I had the privilege of working with Dr. Sledge briefly in his breast cancer clinic and it helped solidify oncology as a career path for me. He loves and is loved by his patients, is an eloquent teacher, and has a fantastic sense of humor. He will be missed at Indiana University. I wish Dr. Sledge the best as he embarks on this next phase in his decorated career and thank both for their work championing oncology informatics.

Can I Break Even Selling Ads On My Website

2012-09-06T00:00:00.000Z

I have been using Google AdSense on one of my websites, Gentoovps.net, for one year now and wanted to answer the question, “Can I Break Even Selling Ads On My Website?” Just to be clear, I did not create the site for monetary purpose. I did not troll for the highest cpc keywords, try to trick users into clicking on ads, or anything else shady. The entire purpose of integrating ads into the site was to offset a small amount of the cost of hosting and domain registration.

AdSense Terminology

Here are some online advertising terms and abbreviations that are needed for the discussion

Page views – displaying one page to one user (regardless of the number of ads on the page
Clicks – user interaction with the advertisement
CTR – click through rate = click / page views (as a percentage)
CPC – cash per click or revenue for each time a user clicks and ad
RPM – revenue per thousand impressions (page views)

Ad Revenue

So here are my data for this past year for a highly focused, linux tutorial website.

Page views: 12,181
Clicks: 11
CPC: $0.58
RPM: $0.53

The remainder of the needed adsense variables can be calculated from these. Back to the main question though “can I Break Even Selling Ads On My Website?” This year, no. I earned $6.40 over the course of the year.

When Will I Break Even Selling Ads?

This was the next best question. I was most interested in how much revenue could be anticipated from each post. That equation is found here.

This assumes though that older articles are still being viewed (whether being found via search, linked to, or navigated to once found by a user). To test this, I looked at a simple correlation between monthly pageviews and time (as a surrogate for total posts). Here is the scatter plot with a horrible correlation.

Community Engagement Pays Off

What was the difference between the first month and the remainder? Community engagement. In order to build backlinks, promote traffic, and solicit feedback, I was very engaged in promoting the site. The downside of course is that this approach is very time consuming. When I remove the first month, here is how the graph looks.

So the more content you produce, the more page views you receive (as expected).

So … When Will I Break Even Selling Ads?

If domain registration plus annual hosting costs $150 and each post generates $0.15/year based on the above formula, then $150/$0.15 = 1000 posts.

There are a few caveats however. Scaling a website from 40 posts to 1000 posts (or 50 page views per day to 1250 page views per day) may require more resources (which cost more money). Remember also that at no point have you begun to compensate yourself for the time of writing the articles. If each post takes 3 hours, you are looking at a 3000 hour time investment to save $150/year. Right now, at the current US minimum wage ($7.25), that is an investment of $21,750 which would require 41 million page views to offset.

Why Bother with AdSense

The short answer is that you probably don’t need to bother. Consider affiliate programs in addition to (or replacement for adsense). For instance, I have made nearly as much revenue from hostigation’s affiliate program based on two reviews I’ve written, as from advertising. Amazon and newegg have affiliate programs which may make more “sense” than adsense as well.

Other Factors that will Influence Your Ad Revenue

how competitive are the ads displayed (ie: home refinancing will increase your CPC)?
how good is your search engine optimization?
are you linking to your best content?
can you decrease your bounce rate?
is the placement of your ads optimized
are you writing (and linking to) follow ups of your best content

Improving these will increase your page views without forcing you to create new content. In the end though, content is always king.

Rethinking Secure Electronic Sign out (Hand off or Changeover)

2012-08-30T00:00:00.000Z

A pubmed search for electronic sign out (or hand off) yields 42 results. I wrote recently about secure patient health information and have been thinking more about what it takes to have a secure, physician-to-physician electronic sign out system. To begin, we'll assume two things: 1) no integration with existing systems. This is not a small assumption but I would like to work on a high level and ideally a hand off could be generated by proprietary EMR X and submitted in a secure way via the following system. 2) The system and interface should be as simple as possible to achieve the minimum required security. The definition of minimum security when it comes to signing out patient information is murky at best. Here is my solution in infographic form.

Secure Electronic Sign Out

Immediately, you probably assume I have broken assumption #2 from above. The above diagram describes a system of accessing (and sharing) encrypted data through 3 layers of encryption. Three layers are not more beneficial from a data security standpoint but the process of changing over patients assumes multiple providers accessing the data, changing passwords, etc. Also, we should address problems like session hijacking, database hacking, and password compromise.

The Need For a Complex Electronic Sign Out System

To the provider, the system could be implementing very simply and the only thing that might be noticeable would be needing to type your password again for any data entry or edit. For those of us practicing in the era of computerized physician order entry (CPOE), this is common place.

Primary data key - this is a random key that is generated for each unique data field to be encrypted. This key is used to encrypt the data itself. If the key is compromised only one unit of data are revealed. An encrypted version is stored in the database.

Secondary data key - this is a random key that is generated for each method of accessing each unit of encrypted data. Since each primary data key will be accessed by two ways per user, the total number of database entries is at least data units x 2. If the key is compromised only one unit of data are revealed. An encrypted version is stored in the database.

Master user key - this is a random key that is generated once on creation of a user account and serves to be the initial key by which a secondary data key is encrypted. If the key is compromised all the user's data are revealed. An encrypted version is stored in the database.

Salted password hash - a salted password hash is used as the key to encrypted the master user key. In the event that a user needs to change his or her password, the only object that requires updating is the encrypted signature of the master key. This eliminates any lag as a result of changing passwords.

Session key - this is not mandatory, but may reduce exposure to session hijacking. Instead of placing the master key in the session to decrypt data, a session-specific key is created (along with a new set of secondary data keys). At conclusion of the session or after an expiratory period, the session key and secondary data keys can be removed.

Temp data key - in order to share information, a secondary data key can be cloned and encrypted using a temporary data key. This key can then be transmitted to the second user. Upon accessing the data, the second user will automatically create a secondary data key encrypted with their own mater key. At this point, two users can access hand off (or sign out) data which is encrypted with a single, unknown, primary data key.

It is probably not ideal yet, but this process is superior to sending changeover via email or simply using whole disk encryption.

Why You Can't Be Anything You Want to Be

2012-08-23T00:00:00.000Z

My grandfather worked as a carpenter from the time he was a teenage apprentice until a couple years before his death. I loved making projects with him, and he taught me a lot about woodworking. One of the most striking things about him was that he loved going to work. Whether it was remodeling a kitchen, making a custom bench, or repairing a deck, he loved it. He told me once that he never woke up and told himself that he didn't want to go into work.

Sean Johnson wrote an intriguing article, Why You Can't Be Anything You Want to Be, at his blog this month. It does fly in the face what my generation has been told about pursuing career happiness.

Reading the article, I feel lucky to be able to purse a career that makes me happy, seems to be in demand, and that I'll want to wake up and go do. Sean's five themes are as follows:

the market doesn't care what you love
you must create value
you must be adaptable
you must learn how to sell
you must be entrepreneural

Sean's idea is not new. One scene from Office Space goes like this:

[Peter, Michael, and Samir are chatting as they hang around the printer] Peter Gibbons: Our high school guidance counselor used to ask us what you'd do if you had a million dollars and you didn't have to work. And invariably what you'd say was supposed to be your career. So, if you wanted to fix old cars then you're supposed to be an auto mechanic. Samir: So what did you say? Peter Gibbons: I never had an answer. I guess that's why I'm working at Initech. Michael Bolton: No, you're working at Initech because that question is bullshit to begin with. If everyone listened to her, there'd be no janitors, because no one would clean shit up if they had a million dollars. Samir: You know what I would do if I had a million dollars? I would invest half of it in low risk mutual funds and then take the other half over to my friend Asadulah who works in securities... Michael Bolton: Samir, you're missing the point. The point of the exercise is that you're supposed to figure out what you would want to do if... [printer starts beeping] Michael Bolton: "PC Load Letter"? ...

Interesting my grandfather's generation probably looked more to apprenticeships when joining the workforce as that added a valued skill. My parents generation seemed to value job stability but may have lacked some in adaptability. My generation is trying to add value through education. Unfortunately, the more bachelor, graduate, and doctoral degrees produced, the less each is worth (supply and demand again).

Who knows, maybe Rich Dad Poor Dad will soon be required reading before entering college. Figuring out the best way to find happiness at the bazaar where we all swap time, money, and ideas has never been easy for any generation. I'd be happy to have a career like my grandfather and wake up ready to go to work each morning. Medicine has been great at that so far.

Diuretics and Heart Failure

2012-08-20T00:00:00.000Z

Diuretics and Heart Failure

Test your knowledge of diuretics and heart failure. [slickquiz id=3]

More Quizzes

Find more quizzes while you are here.

Nvidia VDPau Purevideo PCI Video Card for Mythtv

2012-08-16T00:00:00.000Z

Recently, I was in needed of an Nvidia pci video card for a new installation of Mythtv I put together for watching the olympics. Last Christmas my sister and I gave our parents a new computer so I inherited the old one for re-purposing. I didn't want to dump a lot of money into the box so I needed a hardware pci capture card to record over the air (OTA) hdtv and an nvidia pci video card with vdpau (or purevideo) support.

Free Hardware Rundown

One of the barriers to just anyone creating a mythtv system is figuring out whether a given set of hardware will work for a required task. One of the most common questions on the mythtv user mailing list is something along the lines of "hey, will this hardware [record hdtv, play hdtv, stream, support three frontends etc]." Here are the rough specs of what I got given to me.

Pentium 4 2.8 gHz (single core) processor
1 gb ram
160gb ide HD
onboard intel-based graphics
3 pci slots and one agp slot

Will it play it? No. Will it record it? Probably ... but I wasn't too excited about the single ide drive. You should be able to find something like this on ebay for around $100.

Finding a Cheap Fanless VDPau PCI Video Card

After checking wikipedia's purevideo page, I started to look up chipset and see if any manufacturer had put a gpu with complete acceleration for MPEG-1, MPEG-2, MPEG-4 Part 2 (a.k.a. MPEG-4 ASP), VC-1/WMV9 and H.264 onto a pci card, made it fanless, and sold it for cheap. There was really only one card I could find. The Zotac ZT-40605-10L fit the bill. Total cost was around $60 after rebate and decodes full 1080p with minimal use of system resources. It supports 4th generation vdpau and while I may have the option to use pci-express in a future build, I will almost certainly have an open pci slot. There were some concerns that 512mb of video ram might be insufficient, but I haven't run into this problem.

Cheap PCI OTA HDTV Capture

Before the nationwide conversion to OTA HD, I loved hauppage capture cards for how easy they were to install so I was hesitant to stray away at first. The PCI requirement was, again, a difficult one. I ended up finding a dvico fusion hdtv5 RT Lite on ebay for less than $20 (including shipping). It works great and I couldn't be happier. I bought a thin, $40, HD antenna and the combination is perfect.

Installation was a bit harder than I remember the pvr 250 being from many years ago but it works and I don't plan on changing to anything else for now. The other benefit of this card is that if you wanted to put two or three of them into a single system, the motherboard wouldn't cost you an arm and a leg.

Finishing Up

Total cost of the upgrade was about $120 after tuner, video card, and antenna. I had some extra ide hard drives laying around so I created a four drive raid0 array to help with disk io (although I don't have any benchmark data for that vs single drive). The whole system is sitting in the original ugly black case in my living room and has had me happily skipping through commercials all Olympics long.

Seizure Risk for Cefepime if not Dose Adjusted

2012-08-13T00:00:00.000Z

Cefepime Seizure Risk

On 6/26/2012, the FDA annouced an a risk for seizure when cefepime is not dose-adjusted for renal dysfunction. The Drug Safety Communication announced the increased seizure risk based on on 59 cases of non-convulsive status epilepticus being reported over the six years that cefepime has been available.

56% of these cases involved patients >65 years of age (range: 7-95 years) and 69% of the 59 cases involved female patients. Renal dysfunction was present in 58/59 patients (renal status was unknown in one patient). In 56/59 patients, cefepime dosing was not appropriately adjusted for renal impairment as recommended in the cefepime label.

FDA Drug Safety Communication

You've probably been dose-adjusting cefepime already, this just adds even more data for your support.

cefepime usage

Pneumonia
Febrile neutropenia
Uncomplicated and complicated urinary tract infections
Uncomplicated skin and skin structure infections
Complicated intra-abdominal infections

Disclaimer

As always, consult your medical provider and/or pharmacist before making any medication changes. This update in no way constitutes personal medical advice.

Coffee and Medicine

2012-08-09T00:00:00.000Z

Coffee and medicine go together like ... well ... coffee and doctors. I have loved coffee at least since medical school - most of us succumb to the peer pressure of drinking coffee while studying medicine. My earliest memory of coffee was sitting at my grandparents home where my grandmother would let me have the last few sips of her coffee as a child. Today, almost randomly, I came across the Travis Coffee Company. You can imagine my excitement.

Coffee and Medicine

I, of course, am not alone. A pubmed search today for studies, systematic reviews, and trials in humans with "coffee" in the title yields 280 results. Here are some of the highlights.

Coffee and All Cause Mortality

This is the coup de gras paper and should be kept in tablet or smart phone of any avid coffee drinker to defend the, "don't you think that's too much coffee?" Freedman, et al showed in the article Association of Coffee Drinking with Total and Cause-Specific Mortality, published New England Journal of Medicine May, 2012, showed the following:

Adjusted hazard ratios for death among men who drank coffee as compared with those who did not were as follows: 0.99 (95% confidence interval [CI], 0.95 to 1.04) for drinking less than 1 cup per day, 0.94 (95% CI, 0.90 to 0.99) for 1 cup, 0.90 (95% CI, 0.86 to 0.93) for 2 or 3 cups, 0.88 (95% CI, 0.84 to 0.93) for 4 or 5 cups, and 0.90 (95% CI, 0.85 to 0.96) for 6 or more cups of coffee per day (P<0.001 for trend); the respective hazard ratios among women were 1.01 (95% CI, 0.96 to 1.07), 0.95 (95% CI, 0.90 to 1.01), 0.87 (95% CI, 0.83 to 0.92), 0.84 (95% CI, 0.79 to 0.90), and 0.85 (95% CI, 0.78 to 0.93) (P<0.001 for trend).

Coffee and Dose Response

The best part (for me) in regards to the above paper is that the more coffee you drank (at least to six cups a day), the larger the benefit. I think it's about time for a refill. Coffee is not a medicine and is not FDA approved for primary prevention of any disease. Almost every other addictive substance has a long list of warnings and adverse effects. It's nice to have data supporting moderate to excessively indulgence. Cheers.

Hostigation Review

2012-08-06T00:00:00.000Z

Hostigation Review

I posted a review of my vps host over at hostigation review on gentoovps.net. Over the past twelve months, I could not be happier with there service. If you are in the market for a vps, I suggest you give them a look.

From the Article

March 31, 2012 the network was down for aprpoximately 5 hours. Quadranet, who provides the physical servers, were sorting through a DDoS attack. Since I could not access their main site, I sent a tweet to @hostigation and got a twitter reply 53 minutes later explainging the situation. I thought that was great response time. The DDoS attack was clearly out of their control and while, I suppose, they could have tried to update dns records to an alternative server to explain the situation, I’m not paying for enterprise-level support.

One month is not a fair amount of time to really judge a hosting providing. Over the past year, everything has just worked. Being able to boot right into the gentoo live cd instead of booting ubuntu into recovery mode and installing gentoo is great. There are other cheaper or faster vps providers out there. Tim appears to be a very competent sysadmin and with 99.9% uptime, speedy service when I need it, and good communication, my money will stay here.
I’ll update if anything changes.

Full Hostigation Review

Read the full review.

End of the Double-Blind Randomized Control Trial?

2012-08-02T00:00:00.000Z

The short answer is probably not - the randomized control trial is safe for now. The longer answer when it comes to exploring molecular targets in oncology we may need to rethink the double-blind randomized control trial.

Since current methods to assess potential cancer treatments are cumbersome, expensive, and often inaccurate, can we develop other methods to rapidly test interventions for cancer treatment or prevention?
Advances in 3-dimensional cell culture suggest that multiple cell types can be assembled in vitro and that engineered tissues often mimic many of the features of human organs. If systems can be developed that mimic the natural environment of tumors, perhaps these models will recapitulate drug action. It also seems possible that complex cell-free systems could be developed that would recapitulate at least some features of drug responses. Since it seems unlikely that any one new system will serve as an accurate model for all tumors, each may need to be tuned to the particular features of a particular tumor type or subtype.

NCI PQ 17*

*NCI provacative questions

Targeted Antibodies and Proteins are all Around Us

One counter to the argument of alternative research methods for these agents is that many drugs that fit this bill already have indications based on randomized control trials. A quick search plus no more than a minute of head scratching brings over thirty to mind.

alemtuzumab
alitretinoin
bevacizumab
bexarotene
bortezomib
brentuximab
cetuximab
crizotinib
dasatinib
denileukin diftitox
erlotinib
everolimus
gefitinib
ibritumomab
imatinib
ipilimumab
lapatinib
nilotinib
ofatumumab
panitumumab
pazopanib
pertuzumab
pralatrexate
rituximab
romidepsin
sorafenib
sunitinib
temsirolimus
tositumomab
trastuzumab
tretinoin
vandetanib
vemurafenib
vorinostat

They Survived the Randomized Control Trial ...

This begs the question why do we need alternative or supplemental methods of investigation? Not every, in fact few, cancers are caused by a single bcr-abl-like mutation. The more common scenario is something like lung cancer whose genome appears is the circus plot below.

Pleasance et al. Nature 463, 184-190 (14 January 2010).

Driver's Seat

Once a driver mutation is identified in animal models, it will be increasingly difficult develop targeted therapies if the rate of mutation in the tumor is small. In the past, it was cost prohibitive to sequence many different tumors looking for these less common mutations. Times are changing and our technology is likely outpacing our investigative model.

Trial Proposals

I would love to see if anyone has an alternative to our standard trials. For now, randomized control trials are safe but the new paradigm may be coming. We just don't know what it looks like yet.

Windows Portable Sync Comparison

2012-07-30T00:00:00.000Z

The Sync Problem

As chief resident, I work on my office computer during the day. Often I work on the same files at night by transporting them via usb stick. At home synchronizing my usb stick to my laptop is easy - I wrote an rsync script for that and can autorun it via a udev rule. I wanted a similar method on my work computer. I did not want a backup or copy of the files, I wanted true synchronization (add, modify, and delete bidirectional support).

I looked at three different portable solutions that do not require installation or administrator privileges.

Dsynchronize

Dsynchronize was the first application I tried. The interface is included here. I was optimistic based on the option for "bidirectional synchronization" and the box at the bottom the allowed you to be prompted prior to file deletion. The synchronization process was fast and modification and new files were propagated easily. When I deleted a file on the "destination" folder, the file was recreated during the next sync. Deal breaker.

Toucan

Toucan by far had the most features and well-polished interface of the three I reviewed. At first I could not understand why it did not have a sync function but after reading the documentation online, "equalise" is essentially the same. Toucan sports some very nice backup options including encryption and file renaming based on variables like date and timestamp. As expected, modification and new file creation were no problem, but the "source" and "destination" model tripped up design (from my standpoint) by recreating deleted files on the destination side. So, while beautiful, a deal breaker.

Synkron

As I'm starting the think the only way to fix this problem is to run a set of Windows batch scripts, I come across synkron. The interface is fine and I can declare multiple "master" folders through the main interface. I stopped exploring, ran a sync, then deleted a file on the network drive. Upon resync, proof, appropriate deletion. We instantly had a winner. Since then, it raised in my book even higher when I clicked the restore tab and have the ability to restore recent versions of the deleted files. Fantastic. I can set it to sync and start and it will synchronize by work computer with my network drive at work and my usb stick. Just what I wanted.

Better Sync Tool?

If you are using a better sync tool, let me know and I'll post it here.

Zofran Causes Prolonged QT at High Dose

2012-07-26T00:00:00.000Z

Ondansetron (Zofran) Prolonged QT

On 6/29/2012, the FDA announced an a risk for prolonged QT in patients given the high dose (32mg) ondansetron injection. The Drug Safety Communication announced that the 32mg dose should be avoid and the drug's label will be revised. The risk appears to be dose dependent.

The most common scenario for using the higher dose is as an anti-emetic administered during chemotherapy. The updated label will state that no single intravenous dose should exceed 16mg.

The new information on QT prolongation does not change any of the recommended oral dosing regimens for ondansetron. It also does not change the recommended lower dose intravenous dosing of ondansetron to prevent post-operative nausea and vomiting.

FDA Recommendations

The use of a single 32 mg intravenous dose of ondansetron should be avoided. New information indicates that QT prolongation occurs in a dose-dependent manner, and specifically at a single intravenous dose of 32 mg.
Patients who may be at particular risk for QT prolongation with ondansetron are those with congenital long QT syndrome, congestive heart failure, bradyarrhythmias, or patients taking concomitant medications that prolong the QT interval

Electrolyte abnormalities (e.g., hypokalemia or hypomagnesemia) should be corrected prior to the infusion of ondansetron.

The lower dose intravenous regimen of 0.15 mg/kg every 4 hours for three doses may be used in adults with chemotherapy-induced nausea and vomiting. However, no single intravenous dose of ondansetron should exceed 16 mg due to the risk of QT prolongation.

The new information does not change any of the recommended oral dosing regimens for ondansetron, including the single oral dose of 24 mg for chemotherapy induced nausea and vomiting.

FDA Drug Safety Communication

Disclaimer

As always, consult your medical provider and/or pharmacist before making any medication changes. This update in no way constitutes personal medical advice.

Occupational Exposure to Metals

2012-07-23T00:00:00.000Z

[slickquiz id=2]

More Quizzes

Find more quizzes while you are here.

Raspberry Pi Video Roundup

2012-07-18T00:00:00.000Z

My first Raspberry Pi post was quite popular so I thought I would do a second. You might have caught that production has ramped up and as more people get their hands on a Raspberry Pi, we are seeing more projects and videos pop up. Here are some of my favorite videos showing the core features of the Rapberry Pi in action.

Unboxing the Raspberry Pi

The most popular video type is probably the unboxing. This demonstrates the boot time and bit of lag that you'll expect when running X windows on the Raspberry Pi.

Peripherals

The Raspberry Pi has headers to support daughterboards and should integrate well with arduino-based projects. Here are a few videos showing off the Pi interacting with other devices.

Touchscreen Anyone? Here is a 3.5 inch LCD touchscreen driven by the Raspberry Pi.

Here is a proof of concept that controls LEDS through a web server on the Raspberry Pi running apache and php.

Android transporter with Raspberry Pi ("beaming content")

Demonstration of consumer electronics control (CEC) on Raspbery Pi via HDMI.

Video Playback

Highly touted, the GPU is said to be capable of 1080p output via hdmi. These video some the Raspberry Pi in action playing HD video.

An example of 1080p playback.

XBMC playing mp4 followed by H.264 with Android remote control

Software

A small computer? Cool, I'll make it into a gaming rig. While this isn't my first idea for a great Raspberry Pi project, you can apparently run Quake 3 and Mame as shown below. I'm much more excited to see Java running on the Raspberry Pi.

Mame on Raspberry Pi. You'll want to jump to 3:50 and you'll see some examples.

Quake 3 on Raspberry pi.

JavaFx running on the RPi

Your Pi

What are you doing with your RPi? Post a comment or use the contact form above. I'll try to update the post periodically.

glycosuria without hyperglycemia

2012-07-16T00:00:00.000Z

[slickquiz id=1]

More Quizzes

Find more quizzes while you are here.

Raspberry Pi, a $35 Computer

2012-07-09T00:00:00.000Z

Raspberry Pi - Cheap Computing for the Masses

Recently I have come across Raspberry Pi, a project to create a $35 computer (for model B). The project started in 2006 and will soon be on sale in the United States. The reason I say "soon" is that these are incredibly difficult to find. Before talking about anything else, we need to look at the specs.

Raspberry Pi Specs

Target price:	US$ 35
SoC:	Broadcom BCM2835 (CPU, GPU, DSP, and SDRAM)
CPU:	700 MHz ARM1176JZF-S core (ARM11 family)
GPU:	Broadcom VideoCore IV,[57] OpenGL ES 2.0, 1080p30 h.264/MPEG-4 AVC high-profile decoder
Memory (SDRAM):	256 MB (shared with GPU)
USB 2.0 ports:	2 (via integrated USB hub)
Video outputs:	Composite RCA (PAL & NTSC), HDMI (rev 1.3 & 1.4),[58] raw LCD Panels via DSI
	14 HDMI resolutions from 640×350 to 1920×1200 plus various PAL and NTSC standards.
Audio outputs:	3.5 mm jack, HDMI
Onboard storage:	SD / MMC / SDIO card slot
Onboard network:	10/100 Ethernet (RJ45)
Low-level peripherals:	8 × GPIO, UART, I²C bus, SPI bus with two chip selects, +3.3 V, +5 V, ground
Power ratings:	700 mA (3.5 W)
Power source:	5 volt via MicroUSB or GPIO header
Size:	85.60 × 53.98 mm (3.370 × 2.125 in)
Weight:	45 g (1.6 oz)

The Creator

Darren Kitchens from Hak5 interviews raspberry pi creator at makerfaire 2012.

The video originally embedded here — Hak5’s Maker Faire 2012 interview with the Raspberry Pi’s creator — is no longer available (the Revision3 player has shut down).

Virtual Pi with VirtualBox

If you can't get your hands on a raspberry pi board but what to see what it is like, you can download the RaspberryPi VirtualBox appliance. Once downloaded, just import as an alliance, reinitialize the mac address, and boot it up.

user: rpi
password: password

Obviously change your password with

passwd

when you first login. Also, you may need to

rm -i /etc/udev/rules.d/70-persistent-net.rules

and reboot to get networking running. Once that's done, open the

README

file on the desktop and follow the directions. The only thing I couldn't get to essentially work out of the box was the networking. Here is a screenshot from my quick installation.

click image for larger view ... note ARM processor

Raspberry Pi and Medicine

There has been a lot of talk about integrating tablet computers into our medical workflows. A very cheap, ultra-mini computer, would have some very interesting uses in medicine and medical education. Here are some brainstorming ideas off the top of my head.

quick presentations anyway in the hospital by packaging with a pocket projector to do
provide wireless access for teams rounding at hospitals without available wifi access (our Veteran's Affairs hospital) by packaging with a usb EVDO and usb wireless adapter to
pod-catcher that be plugged into your car stereo system to listen to lectures and could be sync'ed by plugging into network at work by packaging with a small rechargeable battery pack
educational patient videos could be shown by pluggin the device into the hdmi input on TV's in patient or waiting rooms when coupled with small blue tooth keyboard or even remote control
mobile web application and database server for research assistants
cluster computing for genomics or proteomics by networking multiple devices together. A 23 node cluster, even factoring cost of a 24-port switch, comes to only $900

Computer Science Education

Ben Heck made an old school computer with a cartridge slot to add I/O boards based on raspberry pi.

The video originally embedded here — Ben Heck’s cartridge-based Raspberry Pi computer build — is no longer available (the Revision3 player has shut down).

Conclusion

This is a very exciting product and we'll have to wait and see the many ways it will be utilized.

How Stable Is Your Operating System?

2012-07-05T00:00:00.000Z

Defining Stable

My internet "wasn't working" and after figuring out that I could ping my router but not outside, it was clear that I needed to reset my router so that it would be assigned another IP address via my dsl modem. This scenario is common. The only thing that may be a little different for me is that I don't have one of those iconic blue on gray boxes with two antennas sticking up. I was able to log into my router via ssh to restart my network. My router is a full computer. It is a dell gx1 optiplex 400mHz system running Gentoo Linux that I installed installed initially in 2005.

I've updated my installation a few times over the years, but I've not had to add or remove any of the hardware. More impressively, I've never had to reinstall the operating system. Why is that odd? This is a computer that runs 24 hours a day, seven days a week, every week, without a reboot. The most common reason for turning off or rebooting the computer.

moving - literally changing residence is the most common reason
hurricane - that's happened three times while in Florida due to power outage
power outage unrelated to hurricane - long enough for my UPS to lose charge
random crash - that happened once ( I was scared that a piece of hardware failed - still unsolved)

I've never been able to get even one week out of a Windows installation, no matter how vanilla, without a reboot. To make things more fun, this computer has worse specs than the one I took to college 13 years ago. When people use the association between linux and stability. This is what they mean. My operating system was initially intalled June 27, 2005. Happy belated 12th birthday gentoo router. You're probably middle-aged now.

Oncology Informatics Provides Decision Support Based on Gene Mutations

2012-06-18T00:00:00.000Z

Oncology Informatics Provides Decision Support Based on Gene Mutations

Below is a video introduction to a publicly available project, MyCancerGenome.org, that integrates physician decision support, treatment information, published evidence, and current clinical trials. This is a great example of the power of oncology informatics.

As we continue to move toward personalized and targeted therapies, tools like this will become the norm. The challenge is, of course, integrating these tools into our standard work flow.

The next problem is interpreting the amount of molecular and genomic data to make those data applicable to patients. These, however, are good problems to have.

Secure Personal Health Information in 10 Steps

2012-06-14T00:00:00.000Z

Secure Personal Health Information (PHI)

I get frustrated when someone throws around that they have developed a system that keeps PHI "secure." I have a checklist that I generally go through as attacks come in many forms and there is not a single solution that addresses all attacks.

SSL Access

The first step in security is generally forcing all traffic to and from the website to be sent over a secure connection. This connection can be established by either generating a self signed certificate or purchasing a secure certificate from a signing authority. Neither encryption is stronger per say. The benefit of purchasing the secure certificate from the signing authority is that your identification can be confirmed.

If you purchase a certificate from a signing authority and a third party attempts a man in the middle attack on one of your clients, the client's web browser will warn the client that the secure certificate is not signed by a valid authority.

Often when someone says their system is "secure," this is the extent of their claim.

Prevent Page Caching

If your client forgets to close their browser session at the conclusion of their time on the website, it is possible that a third party with physical access to the machine be simply be able to use the back button to navigate to a page that has been cached. The solution here is relatively easy although you cannot guarantee your clients' web browser implements any of these common measures.

Non-MD5 Passwords

If your clients are using dictionary words as their password or basing passwords on dictionary words, the system may be compromised. While any hashing method is at risk for a brute force attack, if your users' password hashes are compromised, MD5 is probably the least secure and easiest to reverse look up in attempt to retrieve the plain text password.

Sanitizing Database Input

An sql injection attack is an attack on a database where a third party submits malformed input into a form that is submitted to the database. The malformed input then either allowed unauthenticated users to gain access to the system or worse, to allow remote code execution and compromise access to the whole system.

SSL Database Communication

If the database is on a separate machine, there is always the possibility of someone sniffing the connection between the website and database host. This is not an issue of the database and webserver reside on the machine. The solution, like above, is to force encryption of the communication.

Allow Database-level Encryption

You can encrypt the database on the filesystem. This prevents someone from copying the database and accessing it on other system. This isn't a perfect scenario as the keys for the encryption are typically stored on the hard disk so if the keys are compromised as well, the encryption doesn't help a lot.

Allow Filesystem Encryption

Unlike database-level encryption, filesystem-wide encryption typically does not have the keys stored on the device but are instead entered at type of booting the system or mounting the filesystem. This protects against a third party with physical access to the machine from unplugging it and taking it to a lab somewhere. It also prevents a third party from reading the disk by booting off a usb or cd.

Two Factor Authentication

If a third party steals a client's password, two factor authentication prevents the third party from logging in. The basic idea is that you can only log in with something you know and something you have. The former is generally a password. The latter is typically a key fob, pager, or cell phone.

Further Encryption

In the event the only a few users have access to discrete information stored in the database, it might make sense to encrypt the information prior to storage with a key that is entered by the client at time of entering and retrieving the data. The biggest benefit to this system it that even if you had over the data in the form of an sqldump to a third party, that party would still need the key to decrypt the information. Therefore, the only attack would be intercepting the communicating after decryption and before deliver to the client or a brute force attack on the encrypted information. In this scenario, even the system administrator cannot gain access to the information so even a fully compromised system protected personal health information.

There are three major downsides however.

it is difficult to allow new clients access to the information
if the key is a secondary hash of the client's password, great care has to be taken if the client ever changes his or her password
in the event that a client forgets his or her password, the information is lost

Hopefully this helps in a primary security evaluation. Prior to implementation of any system, contact your security officer.

Oncology Informatics

2012-06-07T00:00:00.000Z

Oncology Informatics In the Next Decade

I am writing my personal statement for the upcoming fellowship application season in medical oncology. I plan to integrate training in oncology and informatics (oncology informatics). There was a spark for writing ignited in me so I wanted to make a quick post.

ASCO2011 Presidential Address of George Sledge, MD

If you are interested in oncology and moreso in oncology informatics, you need to have watched this. Indiana University is lucky to have two ASCO past presidents on its faculty.

The video originally embedded here — George Sledge’s ASCO 2011 presidential address (a Brightcove Flash player) — is no longer available.

Oncology Informatics Future and Leaders

Someday I predict the the combination of medical oncology and informatics will be as commonplace as the combination of pulmonary and critical care medicine. The leaders of this field will be a diverse group. There is a unique opportunity in my generation to combine basic science, technology, and clinical oncology to make advances that would have been revolutionary one decade ago routine. The combination of skills and timing are best described in Malcom Gladwell's Outliers: The Story of Success

we pay too much attention to what successful people are like, and too little attention to where they are from: that is, their culture, their family, their generation, and the idiosyncratic experiences of their upbringing.

Oncology Informatics and Me

As an undergraduate, I found myself in a computer science introduction course after a scheduling mix up. As fate would have it, I became hooked on computer science at the same time I was learning how cellular biology and biochemistry made the world make sense. In medical school I continued to pursue technical interests as the molecular biology explained the basis of cancer treatment. In my clerkships, medical oncology was solidified as a career path due to the integration of empathy, treating the whole patient, and application of basic science and pharmacology. Oncology informatics was the final iteration which tied everything together.

I am far from unique but I do have passion for both clinical oncology and information technology. My head spins with possibilities of to integrate these fields in the next decade. Whether I will be able to contribute to this exciting field is yet to be determined. Of one thing I am certain - what a great time to be alive.

Staph Infection?

2012-06-06T00:00:00.000Z

Remember Clindamycin

A good point was brought up yesterday in morning report. If you have an infection (say necrotizing pnuemonia or cellulitis) and it looks worse after broad spectrum antibiotics, you may be seeing the results of the released Staph TSST-1 toxin. These patients benefit from concomitant treatment with clindamycin.

In a brief literature review there were some reports of linezolid having similar effects. There are not good data to support its routine use to bind toxin however.

Unasyn vs Zosyn

2012-06-05T00:00:00.000Z

I've often asked by medical students what additional coverage is gained with using piperacillin-tazobactam (Zosyn) over ampicillin-sulbactam (Unasyn). The question comes up because so many people reflexively reach for "vanc-zosyn" for broad coverage. The answer is Pseudomonas.

The more difficult question is what additional coverage is gained with ampicillin-sulbactam over piperacillin-tazobactam. The answer is acinetobacter.

The additional coverage comes from the the sulbactam portion. This isn't to say that piperacillin-tazobactam should be a first line treatment, but it's interesting nonetheless.

http://phil.cdc.gov/PHIL_Images/10095/10095_lores.jpg

Thunderbird and Google Calendar

2012-06-02T00:00:00.000Z

I have an office for the first time in many years with my new role as chief resident. Unfortunately, I am more familiar with Google and Mozilla's office-like tools than I am with those from Microsoft. One of the first frustrations I found was having a browser and email client window open to be able to read email requests for meetings and then put them in my Google calendar. Two days ago, I stumbled upon the Google calendar add-on for thunderbird. It is perfect. I'll include a screenshot but it does exactly what I would expect. Thanks go to Bryan Clark.

thunderbird google calendar tab

osterman.co

Timeline — Travis Osterman | Career & Publications

2020s

2025

2024

2023

2022

2021

2020

2010s

2019

2018

2017

2016

2015

2012

2000s

2009

2004

2003

By the Numbers — Travis Osterman | Research, Talks & mCODE Scale

Scholarly record

mCODE at scale

Distinctions

Now — Travis Osterman | Current Focus & Recent Activity

Current focus

Recent activity

Latest (last 12 months)

Latest writing

Essays

Find me elsewhere

Collaboration Network — Travis Osterman | Co-Authorship Graph

How to read it

Most frequent co-authors

What it shows, and where it comes from

About Travis Osterman, DO — Oncology & Informatics Executive

Origin

Practice and leadership

Leadership

Boards and credentials

Training

Selected honors

Explore

Profiles

Bringing genomics out of the PDF and into the chart

A PDF is not data

What "structured" actually buys you

Being eighth, and then being first

The unlock, not the trophy

What to do about it

The oncologist's case for clinical informatics as a discipline

Informatics is a clinical specialty, not a technical service line

What changes when a practicing clinician actually leads

The pipeline problem, and why education is the lever

What I'd ask of the people who staff these roles

Home Lab — Travis Osterman | Self-Hosted & Agentic Systems

Archive

Predicting immunotherapy toxicity before it happens

The signal is already collected

Three toxicities, three different data shapes

What it takes to actually act on a prediction

Where this goes

Essays — Travis Osterman | AI, Oncology & Health Data

The academic-industry AI partnership that actually shipped

The default structure is built to fail

What we did instead

The deployability constraint that made it real

Why the IP and the publications both happened

What to actually require if you're funding one of these

Collaborations — Travis Osterman | Epic, Microsoft, GE, NCCN

Disclosures

Inquiries

Predicting immunotherapy efficacy and toxicity from real-world data

The clinical problem

The partnership

The approach

The published results

Total output

What comes next

The lesson