New paper on Nature Communications, where we investigate knowledge-enhanced multimodal representation learning with chest X-rays and radiology reports. nature.com/articles/s4146…

We present you our new efforts on building medical generalist foundation models for radiology: Arxiv: arxiv.org/abs/2308.02463 Website: chaoyi-wu.github.io/RadFM/ Hope this can promote the development of medical foundation models ！ (1/5)

Our final model outperforms ChatGPT and LLaMA-2 on multiple Medical QA benchmarks ! Open-source materials: Code: github.com/chaoyi-wu/PMC-… Model: huggingface.co/axiong/PMC_LLa… DATA: huggingface.co/datasets/axion… Hope this can promote the development of open-source LLM for healthcare.

🚀 Can’t wait to try it on medical images！#MedicalAI

Excited to see U 😊

Hoping this dataset will make a meaningful contribution to the community's progress in radiological diagnostics🤗.

🚀Ready to make an impact? Email me to submit your model on ReXrank! To keep up to date with major changes to the leaderboard and dataset, please subscribe 🙌

⭐️ Announcing ReXRank, a competition for radiology report generation from Chest X-Rays. Featuring 16 existing competitive approaches, a large private test leaderboard of 10k cases indicative of more real-world performance, and a live arena where radiology experts can do a

💫Excited to share our new study: “Uncovering Knowledge Gaps in Radiology Report Generation Models through Knowledge Graphs”. We’ve developed a system ReXKG to extract structured information from radiology reports, building a comprehensive knowledge graph for in-depth model

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

Automating AI research is exciting! But can LLMs actually produce novel, expert-level research ideas? After a year-long study, we obtained the first statistically significant conclusion: LLM-generated ideas are more novel than ideas written by expert human researchers.

Are you a radiologist interested in shaping the future of AI in healthcare? At Rajpurkar Lab, we are looking for collaborators to join us in an exciting upcoming project. If interested, please reach out - DMs are open! Pranav Rajpurkar

✨ In radiology, the long-term accumulation of specialized knowledge (ontologies for anatomy / observations) , represents a valuable resource that we must leverage for AI model training & evaluation!

🎉 Excited to share our latest research work from Harvard Medical School: ReXplain. Rexplain is about how we make radiology reports understandable for patients. Our key idea? Transform hard-to-understand radiology reports into video explanations.

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Chemistry with one half to David Baker “for computational protein design” and the other half jointly to Demis Hassabis and John M. Jumper “for protein structure prediction.”

Big launch today from our lab at Harvard: 𝐑𝐞𝐗𝐕𝐐𝐀 – a large-scale, high quality question answering benchmark for chest X-rays. This represents a challenge for existing large frontier models, but one we believe will serve as a good progress measure for next gen models.

Our human reader study with 3 radiology residents on 200 cases revealed remarkable results: MedGemma achieved 83.84% accuracy, outperforming the best radiology resident at 77.27% accuracy. A milestone demonstrating AI's potential to augment clinical decision-making in radiology!

Excited to share our work to be presented at CVPR this week – 𝐹𝑎𝑐𝑡𝐶ℎ𝑒𝑋𝑐𝑘𝑒𝑟, an agentic approach to fixing measurement hallucinations in multimodal generative AI models. aliceheiman Xiaoman Zhang

Exciting #AI news! @SidDograMD Xiaoman Zhang Zeke Silva Pranav Rajpurkar explore how Generalist Radiology AI combines generalist medical AI & radiology-specific features to overcome financial, operational, & clinical limitations of narrow AI solutions. pubs.rsna.org/doi/10.1148/ra…