Excited to be at #NeurIPS2024 in Vancouver this week and #ML4H the next! Lots of exciting stuff to talk about this year, be it AUROC/AUPRC, Reproducibility, the MEDS ecosystem effort, and my upcoming lab at Columbia. Reach out if you're in Vancouver and would like to chat!

Really enjoying Fei-Fei Li 's keynote right now at #NeurIPS2024 on the progress and opportunities of visual understanding in AI. She highlights a cool work highlighting major discrepancies between machine and human performance in understanding 3D videos.

I promise if you see my talk this picture will make sense. One key thing I learned while preparing it: ChatGPT really likes generating tiny handles on snow equipment. You'll hear not about that, but about reproducibility, standardization, and the science of health AI

I realized today that Whova, the NeurIPS conference app, linked my profile to some LinkedIn account that was not mine and thinks I'm an insurance agent 😂 It also thought I was a bartender and worked for a Massachusetts school district (though thankfully not at the same time)

Delighted to launch 2 new short courses on AI for Medicine, now open for registration at Harvard Medical School online learning platform! I've curated lessons using real-world research examples, where I go through classic problem formulations and how they have been approached by

Interesting perspective from Matthew McDermott on “The (lack of?) Science of Health AI 🤔

I really like this connection between stability and adversarial subset selection and fairness/bias in Adarsh Subbaswamy 's keynote now at #ML4H2024

I really like this bidirectional flow between AI and the clinic highlighting the major operational barriers in navigating the last mile of AI deployment in health in Dr Shannon McWeeney 🇺🇦 is on BlueSky & Mastodon's keynote at #ML4H2024

Critical area of needed work in MLM usage in Medicine highlighted by Isaac Kohane today at #ML4H2024! How can we understand & align LLM/stakeholder values across medical decisions. My question--are there new opportunities here too for transparency in medical incentives?

Incredibly honored to win Best Paper at the ML for Health Symposium, co-located w NeurIPS!! 🤩 Congrats to co-authors Emily Robitschek , Jin Ge , and many others 👏. Check out our work using LLMs to understand liver transplants, esp understudied social/ econ factors 🏥💰🏠 #ml4h2024

Matthew McDermott chilconference Irene Chen CHIL = - Coffee Hype Is Life - Caffeine Hides Insomnia Levels - Craving Hot, Iced Lattes - Coffee Helps Improve Longevity (The last one might actually be true? There was a UKB study on this: academic.oup.com/eurjpc/article…)

Super excited to join Columbia DBMI this coming July! If you're looking for postdoctoral or PhD opportunities in Health AI (in particular in building foundation models for EHR and multimodal health data), message me!

A new study finds ants best humans at tests of collective intelligence. Learn more: scim.ag/41ZVpVc News from Science

I'm going to remember this every time from now on that I read a new "AI can beat humans at X" study...

For #CHIL2025, we're gathering top literature from 2024 in AI & healthcare. "Top" can mean an exciting topic, a method you found extremely useful, or a game-changing dataset. Drop your recommendations (and why you chose them) in the coments below! 👇

🚨 Calling all health AI founders & builders! Join us at Health AI Builders: A CHIL Unconference — June 25th @ UC Berkeley. 💡 Real talk on AI, regulation, GTM, & fundraising 👥 Small-group convos, big impact 🎯 Apply to attend: lu.ma/2arsxv64 #CHIL2025 #HealthAI #ML4H

To anyone at #KDD2025 & interested in health AI, check out the MEDS tutorial tomorrow, 8/3, from 8-11am! Or see the notebooks offline here: medical-event-data-standard.github.io/docs/tutorials… And thanks to Suhana Bedi Kamilė Stankevičiūtė Justin Xu and the whole MEDS community for putting this together!