Ahmed Alaa
@_ahmedmalaa
Followers
2K
Following
2K
Media
69
Statuses
2K
Assistant Professor @UCBerkeley + @UCSF, @berkeley_ai Prev @broadinstitute, @MIT, @UCLA, @Cambridge_Uni, @UniofOxford. Machine Learning & AI for Healthcare.
Berkeley, CA
Joined February 2011
📢 Please retweet: We're recruiting **PhD students and Postdocs** at UC Berkeley & UCSF! Please apply if you are interests in AI for healthcare, statistics/causal inference, or medical vision-language models. For more details, check out this link: https://t.co/3ycodEjdrM
10
148
571
Website live! My UC Berkeley's undergrad ML for healthcare class (w @YalaTweets) was oversubscribed 3X more than room capacity. Putting our course materials online to broaden reach
2
8
81
Our new paper with Sonali Sharma and @RoxanaDaneshjou is out in @npjDigitalMed! We examine how medical safety and disclaimer messages in public LLMs have changed over time when answering patient questions.
Generative AI models are giving fewer medical disclaimers over time. 📉 In 2022, ~26% of AI health answers had a disclaimer. By 2025? <1%. As models get smarter, they’re getting less safe. Patients may take outputs as medical advice. https://t.co/2OYQvKdezT
3
8
17
Our postdoc application is now open! Interested in pursuing #machinelearning, #appliedmathematics, #statistics, or #computationalresearch to work on biomedical problems at the @broadinstitute? Apply to become a @Schmidt_Center postdoctoral associate: https://t.co/VXqP62c19w
1
13
21
📢Please retweet: We are hiring a **Postdoc** at UVA to work on Continually Monitoring and Updating Multi-modal Medical AI Models! Great opportunity to design impactful methods alongside great collaborators @_ahmedmalaa and @RoxanaDaneshjou More info:
0
13
37
University of California faculty and alumni won five Nobel Prizes this week, setting a new record for the most faculty of one institution to achieve this great honor in a single year 🏅💙 💛 These remarkable achievement highlight the ongoing contributions of America’s #1
universityofcalifornia.edu
These discoveries span decades and disciplines, but they all have one important thing in common: They’ve all relied on competitive funding from the federal government.
17
253
1K
Excited to launch our project on clinician-in-the-loop post-training methods for medical multimodal AI with @RoxanaDaneshjou and @tom_hartvigsen. Grateful to @NIH for supporting this work!
Good news! CPH's @_ahmedmalaa and colleagues @RoxanaDaneshjou and @tom_hartvigsen have received a grant from the NIH Director's office for work on participatory mutimodal AI and editable foundation models to improve diagnosis and tx. @BerkeleyCDSS @StanfordDBDS @UVA
1
2
15
The CPH PhD program is now accepting applications for Fall 2026 🎓 ✨ We're looking for passionate, driven students ready to shape the future of health with the power of computation. 🔗 Learn more and apply: https://t.co/X8hGQoKywo
#AIHealth @UCSFGradDiv @Graddivision
0
5
11
Our paper "Generative AI enables medical image segmentation in ultra low-data regimes" is selected as Nature Communications Editors’ Highlights ("the 50 best papers recently published")! https://t.co/VDYDXOcTV1
@james_y_zou @_ahmedmalaa @segal_eran
nature.com
On this page we provide a snapshot of some of the most exciting works recently published in Nature Communications in cancer research. We cover all aspects of ...
1
6
25
One-liner summaries are central to emergency care, but generating them is cognitively demanding/time-pressured. We evaluated LLMs for automated one-liner generation & compared them head-to-head w/ physician summaries. New work w/ @MelMolinaMD, led by @nbgolchini @nikita_mehandru!
We show LLMs perform better than physicians in summarizing patient clinical information from the EHR as one-liners. One-liners are essential in emergency medicine, where providers make high-stakes decisions with limited information. @_ahmedmalaa @nikita_mehandru @nbgolchini
3
8
32
One-liners summarizing a patient's clinical state are essential in the emergency department. Not surprising that LLM's/GenAI do better than human doctors at producing these one liners
medrxiv.org
High-quality one-liner summaries are essential in the emergency department (ED) to support rapid decision-making, but generating them is cognitively demanding and adds to documentation burden; large...
1
13
53
Medical disclaimers in AI outputs are quietly fading, but they matter for safety and trust! Our recent study on this topic is featured in a new MIT @techreview piece.
AI companies have stopped warning you that their chatbots aren’t doctors
0
0
12
If you are at #ICML2025, don't miss our talks/posters! - Medical LLM Benchmarking (Position Paper): Oral - July 16, 3:30 PM + Poster - July 16, 4:30 PM by @tom_hartvigsen - Venn-Abers Calibration + Conformal Prediction: Poster on July 16, 4:30 PM by @LarsvanderLaan3
0
7
25
How has medical disclaimer messaging changed in generative AI models over the past few years? New work with Sonali Sharma and @RoxanaDaneshjou!
Brilliant student Sonali Sharma came to me with a question. If patients are using AI to answer their medical questions, are they being adequately warned by AI systems that it cannot provide medical advice? What we found surprised us!
0
3
7
New paper out in @NatureComms led by @cmuptx 's team! We study how generative AI models can enable accurate medical image segmentation even with limited labeled data.
🚀 Excited to share that our work GenSeg has been published in Nature Communications! GenSeg is an end-to-end, downstream-task-guided framework for generating synthetic training data for medical image segmentation. It significantly reduces the need for manual annotations — by a
0
5
13
Looking for an emergency reviewer for NeurIPS paper for a fairness paper, due in a week. Email or DM me! Ideally you’ve first-authored a paper in a major ML conference Reward is a thoughtful compliment about you and your work
1
2
10
individual reporting for post-deployment evals — a little manifesto (& new preprints!) tldr: end users have unique insights about how deployed systems are failing; we should figure out how to translate their experiences into formal evaluations of those systems.
7
31
142
What more could we understand about the fractal, “jagged” edges of AI system deployments if we had better ways to listen to the people who interact with them? What a joy to work w @jessicadai_ using individual experiences to inform AI evaluation (blog/ICML/arXiv links 👇)
4
19
66
Honored to receive a Google Research Scholar Award. Thanks @GoogleResearch for supporting our work!
Congratulations to CPH's @_ahmedmalaa, receiving @googleresearch Scholar award for groundbreaking work on use of AI and high-throughput proteomics to predict treatment impact for populations often excluded from clinical trials. @BerkeleyDataSci @UCSF_BCHSI
8
1
78
Just did a major revision to our paper on Lifelong Knowledge Editing!🔍 Key takeaway (+ our new title) - "Lifelong Knowledge Editing requires Better Regularization" Fixing this leads to consistent downstream performance! @tom_hartvigsen @_ahmedmalaa @GopalaSpeech @berkeley_ai
1
7
27