Microsoft Research Releases PRISM2 Pathology AI Model
Microsoft has described PRISM2, a pathology foundation model developed by researchers at Microsoft Research and Paige, now part of Tempus. The study behind it was recently published in Nature Medicine, and the full model weights are publicly available for research use on Hugging Face.
Transcript
Today Microsoft Research and Paige released a pathology AI model trained on tissue images and language.
Most pathology AI systems are built for one task. They must be rebuilt for each new application. A study in Nature Medicine describes a different approach.
Researchers paired pathology images with information from pathology reports. That created millions of question and answer examples. The model works with images alone or with images and text.
In testing it matched or exceeded specialized cancer detection systems on several benchmark tasks. No separate model was needed for each task. The weights are public for research use.
Inboxsmith helps small businesses handle calls and messages so nothing gets missed. Please like and subscribe for more news.
Sources
Every claim in this video comes from the top ranking coverage of this topic. The claims and where each one came from:
- Researchers from Microsoft Research and Paige, now part of Tempus, developed PRISM2, a pathology foundation model trained on both tissue images and language based on real pathology reports.(Microsoft's official announcement)
- But many of today's pathology AI systems are built for a single task and must be rebuilt for each new application.(Microsoft's official announcement)
- A study recently published in Nature Medicine describes a different approach aimed at advancing research in this field.(Microsoft's official announcement)
- To train the model, researchers paired pathology images with information derived from pathology reports, creating millions of question-and-answer examples that helped connect visual findings with diagnostic language.(Microsoft's official announcement)
- The resulting model can work with either images only or images and text, enabling users to interact with it through prompts rather than relying solely on task-specific software.(Microsoft's official announcement)
- In testing, the model matched or exceeded the performance of specialized cancer-detection systems on several benchmark tasks, including prostate cancer, breast cancer and breast lymph node metastasis detection.(Microsoft's official announcement)
- Importantly, researchers reported these results without creating a separate model for each task.(Microsoft's official announcement)
- The full PRISM2 model weights are publicly available for research use on Hugging Face.(Microsoft's official announcement)