Researchers from the University of California, Berkeley and University of California, San Francisco have released an open-source AI model, Pillar-0, trained to interpret and analyze medical images.
The model’s research, code and documentation are available for free online to “accelerate research and clinical adoption,” according to a Nov. 20 news release from UC Berkeley.
Pillar-0 was validated on chest, abdomen and brain CT scans, and breast MRI scans, achieving an AUC — area under the curve — of .87, compared to AUCs of .76 from Google’s MedGemma, .75 from Microsoft’s MI2, and .70 from Alibaba’s Lingshu models, according to the release.
Access the model here.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
