Jason Alan Fries

I’m recruiting postdocs who are excited to work with real clinical data and partner closely with clinicians.
If you’ll be at ML4H or the first day of NeurIPS, let’s connect!
More about my work: web.stanford.edu/~jfries/

01.12.2025 07:29 — 👍 0    🔁 0    💬 0    📌 0

My lab will focus on multimodal foundation models for healthcare, combining CS and clinical collaboration to understand and treat complex diseases like cancer.

Core interests: representation learning, synthetic data generation, longitudinal benchmarks, and agentic clinical AI.

01.12.2025 07:29 — 👍 0    🔁 0    💬 1    📌 0

I’m excited to share that I’ll be joining Stanford as a tenure-track Assistant Professor of Biomedical Data Science and of Medicine on Dec 1, 2025. 🎉

I’ll hold a joint appointment in DBDS and the Division of Computational Medicine.

01.12.2025 07:29 — 👍 2    🔁 0    💬 1    📌 0

AI in Clinical Science - amazing data being presented today by @jason-fries.bsky.social Sylvia Plevritis @roxanadaneshjou.bsky.social @akshay-chaudhari.bsky.social but still feel like we are just barely cracking the egg in this field. So impatient for the omelette…!

@stanford-cancer.bsky.social

04.11.2025 22:33 — 👍 3    🔁 2    💬 0    📌 0

🎉 Headed to MLHC 2025 this weekend?

Swing by Poster #154 (Session C) on Saturday, Aug 16 to check out FactEHR — our new benchmark for evaluating factuality in clinical notes. As LLMs enter the clinic, we need rigorous, source-grounded tools to measure what they get right (and wrong).

14.08.2025 18:17 — 👍 2    🔁 0    💬 0    📌 0

ICLR Poster Time-to-Event Pretraining for 3D Medical ImagingICLR 2025

🎉 Excited to present our #ICLR2025 work—leveraging future medical outcomes to improve pretraining for prognostic vision models.

🖼️ "Time-to-Event Pretraining for 3D Medical Imaging"
👉 Hall 3+2B #23
📍 Sat 26 Apr, 10 AM–12:30 PM
🔗 iclr.cc/virtual/2025...

23.04.2025 21:00 — 👍 1    🔁 1    💬 0    📌 0

[4/4] This was a massive collaboration involving multiple offices and champions across Stanford University and @stanfordmedicine.bsky.social

Thanks to our research team: Michael Wornow, Ethan Steinberg, Zepeng Frazier Huo, Hejie Cui, Suhana Bedi, Alyssa Unell, Nigam Shah and many others.

13.02.2025 01:38 — 👍 1    🔁 0    💬 0    📌 0

[3/4] 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱𝗶𝘇𝗲𝗱 𝗕𝗲𝗻𝗰𝗵𝗺𝗮𝗿𝗸𝘀
Each dataset includes a set of standardized tasks exploring a technical challenge area in AI.
🎯 Few-shot Learning
🤖 Multimodal Learning & Time-to-Event Modeling
⌛ Long Context Instruction Following & Temporal Reasoning

13.02.2025 01:38 — 👍 0    🔁 0    💬 1    📌 0

[2/4] 𝗗𝗮𝘁𝗮𝘀𝗲𝘁 𝗦𝘂𝗺𝗺𝗮𝗿𝗶𝗲𝘀
📊 3 longitudinal EHR datasets
• Scale: 25,991 patients | 441,680 visits | 295M clinical events (median: 4,882 events/patient)
• Timeframe: 1997–2023 (median: 10 years/patient)
• Multimodal: structured EHR data, 3D medical imaging, and clinical notes

13.02.2025 01:38 — 👍 1    🔁 0    💬 1    📌 0

Advancing Responsible Healthcare AI with Longitudinal EHR Datasets Current evaluations of AI models in healthcare rely on limited datasets like MIMIC, lacking complete patient trajectories. New benchmark datasets offer an alternative.

[1/4] 🎉 We're thrilled to announce the general release of three de-identified, longitudinal EHR datasets from Stanford Medicine—now freely available for non-commercial research use worldwide! 🚀
Learn more on our HAI blog:
hai.stanford.edu/news/advanci...

13.02.2025 01:38 — 👍 7    🔁 3    💬 1    📌 1

Time-to-Event Pretraining for 3D Medical Imaging With the rise of medical foundation models and the growing availability of imaging data, scalable pretraining techniques offer a promising way to identify imaging biomarkers predictive of future disea...

[1/4] Excited to share that our paper 𝘛𝘪𝘮𝘦-𝘵𝘰-𝘌𝘷𝘦𝘯𝘵 𝘗𝘳𝘦𝘵𝘳𝘢𝘪𝘯𝘪𝘯𝘨 𝘧𝘰𝘳 3𝘋 𝘔𝘦𝘥𝘪𝘤𝘢𝘭 𝘐𝘮𝘢𝘨𝘪𝘯𝘨 is accepted at ICLR 2025! 🚀
We introduce 𝗧𝗧𝗘 𝗽𝗿𝗲𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴, using EHR-linked imaging to improve AI-driven prognosis—essential for assessing disease progression.
🔗 Paper: arxiv.org/abs/2411.09361

02.02.2025 06:10 — 👍 4    🔁 1    💬 1    📌 0

Excited to share our paper "Time-to-Event Pretraining for 3D Medical Imaging" is accepted at ICLR 2025! 🚀

We introduce time-to-event pretraining for imaging, leveraging longitudinal EHRs to provide temporal supervision and enhance disease prognosis performance.

🔗 Paper: arxiv.org/abs/2411.09361

02.02.2025 21:31 — 👍 2    🔁 0    💬 0    📌 0