AI accurately reads brain MRIs in seconds
A new MRI-specific artificial intelligence tool could significantly improve the diagnosis of neurological conditions in busy settings.
Developed by researchers at the University of Michigan, the tool can read brain scans in just seconds. The tool, named Prima, is a video language model that can simultaneously process video, images and text in real time. Experts involved in its development are hopeful it can help address the rising imaging volumes.
“As the global demand for MRI rises and places significant strain on our physicians and health systems, our AI model has potential to reduce burden by improving diagnosis and treatment with fast, accurate information,” said senior study author Todd Hollon, MD, a neurosurgeon at U-M Health.
Researchers trained Prima using more than 200,000 MRI exams collected at the university over several decades. Imaging data were included alongside patients’ medical histories and clinical indications for the scans. The team tested the model on more than 30,000 brain studies over a one-year period. Unlike earlier AI tools that focus on just one disease, Prima was designed to analyze all available imaging and clinical information at once, similar to how a radiologist reviews a case, giving it broad applicability.
“Prima works like a radiologist by integrating information regarding the patient’s medical history and imaging data to produce a comprehensive understanding of their health,” explained co-first author Samir Harake, an MD candidate and data scientist in Michigan's Machine Learning in Neurosurgery Lab. “This enables better performance across a broad range of prediction tasks.”
Testing revealed that Prima could identify more than 50 neurological conditions, including strokes, brain tumors and hemorrhages. And it did so with accuracy as high as 97.5%. The system also determined how urgently a patient needed treatment and was configured to automatically alert the appropriate specialist when a critical finding was detected.
Hollon described the tool as “ChatGPT for medical imaging,” praising its potential to ease workflows in resource-strapped settings, especially in rural areas where there might not be a neurology specialist immediately available to review imaging.
“We believe that Prima exemplifies the transformative potential of integrating health systems and AI-driven models to improve healthcare through innovation,” Hollon said.
Read more here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.