🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

🚨 New Multimodal Generative Medical AI 🚨 Our team just announced MedVersa, a groundbreaking multimodal medical image interpretation model. 🩺🖥️ MedVersa is the first generalist medical AI capable of learning from both visual & linguistic supervision to tackle a wide range of

In 2020, James Diao and I began studying race adjustment in lung-function equations. With amazing coauthors, today we published the past years of work in NEJM, estimating the many clinical, financial, and occupational implications. #ATS2024 Full paper: nejm.org/doi/full/10.10…

Excited to share our review article on incorporating human values into medical AI systems. nejm.org/doi/full/10.10… Arjun (Raj) Manrai ElizabethHealey Tze-YunLeong Isaac Kohane @harvardMED DBMI at Harvard Med NEJM

Please join us for the 2024 PQG conference on AI for genomics and Health on Oct 17-18. A fantastic speaker lineup

LLMs aren’t going to replace scientists in the near future; but scientists who are good at using LLMs for the tasks where they’re most capable will be much more competitive than those who ignore them.

Excited to embark on a journey to develop integrative pathology AI methods for predicting hematologic comorbidities! NIH NHLBI NIH DBMI at Harvard Med Harvard Medical School

It was great catching up with Bo Wang! Thanks for organizing #MLHC2024 Machine Learning for Healthcare University of Toronto

When should we trust a radiology report generation model? When the metrics say to. But when should we trust the metrics? New paper ✨ ReXamine-Global, a framework for stress-testing metrics used to automatically score AI-generated radiology reports. Oishi Banerjee

Our lab’s new work that shows deeper integration of vision and robotic learning! 🤩🦾

We should use soft-max to mean "log of sum of exponentials." What is often called soft-max should really be called soft-argmax. Even John Bridle, who coined the word soft-max, agrees.

The SplitsTree App: interactive analysis and visualization using phylogenetic trees and networks | Nature Methods nature.com/articles/s4159…

Excited to share our new paper in Nature Communications! Learning co-plane attention across MRI sequences for diagnosing twelve types of knee abnormalities. The diagnosis accuracy of all radiologists was improved significantly with model assistance. nature.com/articles/s4146…

🎉 Excited to share that my NIH R35 grant on causal inference methods for treatment heterogeneity has been funded! Grateful for the support from my mentors and colleagues! 🙌 @NIGMS Penn State News Penn State Health #CausalInference

🚀 Beyond excited to announce our release of the #Cell2Sentence (C2S) API and new foundation models! 🎉 Our C2S API makes it incredibly easy to convert #singlecell data into cell sentences, perform inference with LLM-based C2S models, fine-tune them, and convert cell sentences

Nugget in an otherwise AI-optimistic perspective "...medicine has lost some of its most fundamental human aspects, such as trust between patients and doctors and a holistic understanding of patient suffering and personal context..." NEJM AI ai.nejm.org/doi/full/10.10…

After benchmarking more than 80 LLMs, Markus Zimmermann & team found that the best model isn't always a great match for your programming language. Google's Gemini Pro 1.5 worked well for Go, but not so much for Java & Ruby, for example. The best overall LLM was Anthropic’s Sonnet 3.5.