Arjun (Raj) Manrai
@arjunmanrai
Followers
3K
Following
3K
Media
215
Statuses
2K
Assistant Professor of Biomedical Informatics, Harvard Medical School @HarvardDBMI | Founding Deputy Editor, @NEJM_AI | Co-host, NEJM AI Grand Rounds
Boston, MA
Joined July 2014
I’ve been obsessed with the @NEJM CPCs since I was in grad school. Now, with @tabuckley_, it’s surreal to see @NEJM publish the first AI differential diagnosis in the 100+ year history of the series, generated by our AI system Dr. CaBot, alongside the human expert’s.
7
74
285
Original Article by M. Afshar (@Majeans2011) et al.: A Pragmatic Randomized Controlled Trial of Ambient Artificial Intelligence to Improve Health Practitioner Well-Being https://t.co/nIr0ArkzNF
1
4
13
We recently organized #Agents4Science, the 1st conference where LLMs are both authors and reviewers🤖 It was an open experiment to assess how well AI can lead research and review papers. Today we report what we learned in @NatureBiotech Highlights in 🧵
5
72
310
Large language models and the degradation of the medical record. All it takes is one new note relying on this summary, and now it is part of your "ground truth" going forward. The issues discussed in our NEJM editorial in full swing @arjunmanrai @AdamRodmanMD
My doctor at Allina showed me a new AI tool that they're using to summarize medical histories. It can review your whole chart and provide a human-readable summary. He copy-pasted the summary into my notes for me to look at later. The errors were astounding.
0
8
16
In the latest episode of AI Grand Rounds, Dr. Zak Kohane (@zakkohane), editor-in-chief of @NEJM_AI, explains why this year’s advances in AI were personal, why it’s spreading fast, and why its most profound theme is values. Full episode: https://t.co/fhK1HJbHEO
2
9
28
We published one of the first randomized trials of ambient AI scribes, the fastest-growing generative AI application in healthcare. 238 docs. 72K encounters. 2 tools tested over 2 months. Modest time savings. Reduced burnout. Docs reported better patient engagement. No major
2
19
66
Compte rendu de l’utilisation de l’IA en science et du processus accéléré d’évaluation des publications scientifiques assisté par l’IA que nous avons récemment mis à l’essai @NEJM_AI
https://t.co/AKjSRoOlAz HT @arjunmanrai
lemonde.fr
Des éditeurs de revues scientifiques testent des plateformes d’intelligence artificielle pour analyser leurs avantages et inconvénients lors de l’examen d’un article soumis à un journal.
0
2
4
AI scribes may not save a lot of time on clinical notes, but they make clinicians happy @statnews
https://t.co/RgrQA0D8Gm
@nejm_ai HT @arjunmanrai
statnews.com
As hospitals study the impact of AI scribes, they are finding that clinicians love the tools, even if they don’t save that much time.
1
6
28
Medicine evolves by the minute. On @NEJM_AI Grand Rounds, Dr. Laura Zwaan (@LauraZwaan81) explains why diagnostic errors aren’t always what they seem. Listen to the full episode: https://t.co/BmSwwN2v4U
0
4
9
Darn, @AndrewLBeam beat us to Fox News Sunday @zakkohane. Swing state favoritism. Looking sharp Prof Beam!
1
0
3
@NeurIPSConf week Sat, Dec 6 🔹 CUREBench - International Competition on AI agents and reasoning models for therapeutics at scale https://t.co/q40pyydv9C
@GaoShanghua @ScientistTools @RichardYXZhu @sui67713 @ZKong50693 @xiaorui_su with a keynote by @AziziShekoofeh Sun, Dec 7 🔹
1
11
36
New work from our lab @HarvardDBMI led by superstar PhD students @tabuckley_ & @kianweih (who are also good friends!). More on GIANT soon
Manrai x 2 (nepo much?) + boat load of AIM graduate students from @harvarddbmi AIM program for GIANT pathology slide images https://t.co/e6YG6RtKCU Scanning images like a pathologist?
0
4
11
Navigating Gigapixel Pathology Images with Large Multimodal Models GPT-5 with the right agentic scaffold for navigating whole slide images outperforms slide-level pathology models on the novel MultiPathQA benchmark.
5
14
58
Manrai x 2 (nepo much?) + boat load of AIM graduate students from @harvarddbmi AIM program for GIANT pathology slide images https://t.co/e6YG6RtKCU Scanning images like a pathologist?
arxiv.org
Despite being widely used to support clinical care, general-purpose large multimodal models (LMMs) have generally shown poor or inconclusive performance in medical image interpretation,...
2
6
14
A randomized trial of ambient A.I. for reducing data clerk burden (->synthetic notes) for clinicians: significant decrease in burnout @NEJM_AI
https://t.co/vthxEVnh1k
2
34
112
In the latest episode of the @NEJM_AI Grand Rounds podcast, cognitive psychologist Dr. Laura Zwaan (@LauraZwaan81) discusses how machines inherit our biases and why both need transparency and reflection. Listen to the full episode now: https://t.co/BmSwwN2v4U
1
9
23
The Missing Value of Medical AI. It’s the right hill to take for a safe 21st century medicine @NatureMedicine link 👇
2
12
30
Don’t miss JCHI’s Health AI Grand Rounds this Fri, Nov 14! @arjunmanrai (@HarvardDBMI @NEJM_AI ) presents “Dr. CaBot,” exploring how AI can show expert-level clinical reasoning. Title: Advancing Medical Artificial Intelligence Using a Century of Cases When: Friday, Nov. 14,
0
4
8
I'm excited to show off the work of @tabuckley_ and an amazing team of machine learning scientists and physicians at @HarvardDBMI and @BIDMChealth !
Don’t miss JCHI’s Health AI Grand Rounds this Fri, Nov 14! @arjunmanrai (@HarvardDBMI @NEJM_AI ) presents “Dr. CaBot,” exploring how AI can show expert-level clinical reasoning. Title: Advancing Medical Artificial Intelligence Using a Century of Cases When: Friday, Nov. 14,
0
0
9