David JH Wu, MD
@davidjhwu
Followers
3K
Following
8K
Media
565
Statuses
3K
@stanfordradonc resident, previously @scvmed intern. research in healthcare + ai intersection
Cupertino, CA
Joined April 2012
Much thanks to my amazing collaborators @drethangoh @vishnuravi @jonc101x @LiamGMcCoy @fatemenateghi.
1
0
4
This small pilot (n=40) opens the door for more rigorous, large-scale real-world evaluation of medical AI. Instead of just testing on clean vignettes, we can assess how AI performs on the actual messy complexity of clinical practice in a scalable way.
medrxiv.org
Specialist consults in primary care and inpatient settings typically address complex clinical questions beyond standard guidelines. eConsults have been developed as a way for specialist physicians to...
1
0
2
Both compared AI answers to actual specialist responses. Key finding: LLM-as-judge matched human physician performance when assessing clinical concordance. This suggests we can scalably benchmark AI for complex medical tasks that is more considerate of physician reviewers.
1
1
4
The challenge? How do you evaluate AI performance on these real-world consults at scale? We can't have human physicians manually review every AI response so we tested two automated evaluation methods:. LLM-as-judge.Decompose-then-verify: Breaking down responses into atomic facts.
1
0
2
Real-world cases are messy, often incomplete. At @Stanford we have collected a unique dataset: thousands of real physician-to-physician eConsult cases. Think actual specialists answering complex questions from primary care docs—not polished, textbook scenarios. (2/6).
1
0
4
Excited to share some of my recent work! . Y'all have probably seen that there's been a lot of hype around LLMs acing the USMLE and medical vignettes. But here's the thing: standardized tests ≠ real-world medicine! How does AI do in the real-world and how can we eval it? Read on
1
4
22
Buying API credits off of Anthropic really has changed my perspective on what $5 is. I’ll never be able to look at a cup of coffee the same any more. I’ll take the $5 API credit over a latte ANY DAY.
2
1
7
Great panel on publishing health in Health AI at #AIMI25 featuring @charlottehaug (NEJM AI), @yhswen (JAMA+AI), and Chris Paton (BMJ) hosted by @drnigam . “What do editors want?” - Prefer something prospective, reproducible, and useful today. “Do science the standard way”
0
1
4
Some interesting quotes: .“These past few months we’ve seen the same amount of progress in model developments as the past few years”.“Fine-tuning on medical tasks usually causes loss of other skills. Training pipeline is delicate. Risk of catastrophic forgetting.”.
1
0
0
Packed house at #AIMI25 for this star-studded panel featuring @Emily_Alsentzer (Stanford CS), @thekaransinghal (OpenAI), @KhaledSaab11 (DeepMind) and @marinkazitnik (Harvard Biomedical Informatics) and @drethangoh . 🎯: foundational model roadmap for health AI
1
7
52
Welcome to #AIMI25! Great opening remarks from @curtlanglotz @RoxanaDaneshjou and @drethangoh and excellent keynote from @j_r_a_m . Stanford HealthAI week- the most wonderful time of the year 😉
0
1
4
Looking forward to being a virtual moderator at #AIMI25! Feels like only yesterday I was watching as a clueless and curious med student at my first AIMI. We have a great lineup this year- researchers, clinicians, health systems leaders and industry all in one event!.
Excited to announce our stellar speaker lineup for #AIMI2025! Join leading experts (virtually or in person) June 3 to explore AI research, clinical applications & real-world implementation. Free for students & trainees. Scholarships available until May 31.
0
1
5
Been helping a lot of folks with their med school PS’s lately. Figured I’d distill my advice and share with the broader internet. Advice applies to ERAS PS’s as well!
2
0
2
Excited to announce that I will be a virtual moderator at Stanford AIMI 2025! . We’re giving out free passes too! Link below. Have been going the past 5 yrs and love it every year— it’s a terrific converging of clinical insight, industry expertise, and cutting-edge research.
2
0
4
Not sure I like the latest wave of DAX copilot ai-generated notes I’m seeing out in the wild…. Adding on a new layer of AI-slop to copy-forwarded note bloat for docs to filter through. the hydra-head conundrum of medico-technical progress continues.
0
0
1