Ann Arbor, Mich.-based University of Michigan Health has developed an AI model capable of scanning brain MRIs in seconds.
The model, dubbed Prima, detected neurological conditions with up to 97.5% accuracy and determined how urgently patients needed treatment, according to a Feb. 6 system news release. It was tested on more than 30,000 MRI studies and evaluated across more than 50 radiologic diagnoses. Researchers said Prima outperformed other AI models in diagnostic performance and triage capabilities.
Designed as a vision language model, it was trained on more than 200,000 studies and 5.6 million sequences dating back to the start of digital radiology at the health system. It incorporates clinical histories and ordering context to improve performance and recommend appropriate subspecialist referrals in real time.
Leaders said the technology could reduce diagnostic delays and support providers in prioritizing critical cases. Researchers also noted the tool remains in early evaluation, with plans to expand its capabilities by integrating more EHR data for broader applications beyond neuroimaging.
The study was published Feb. 6 in Nature Biomedical Engineering.
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.