An artificial intelligence system developed at the University of Michigan can now analyze brain MRI scans and identify neurological conditions with up to 97.5% accuracy, delivering a diagnosis in a matter of seconds, researchers announced Tuesday.
The novel technology, named Prima by its creator Todd Hollon, M.D., a neurosurgeon at University of Michigan Health and assistant professor of neurosurgery at U-M Medical School, was evaluated over a one-year period using more than 30,000 MRI studies. Prima demonstrated stronger diagnostic performance than other advanced AI models across more than 50 different radiologic diagnoses involving major neurological disorders.
“As the global demand for MRI rises and places significant strain our physicians and health systems, our AI model has potential to reduce burden by improving diagnosis and treatment with fast, accurate information,” Hollon said. The findings were published in Nature Biomedical Engineering.
The AI model was trained on hundreds of thousands of real-world scans, coupled with corresponding patient histories, to achieve its high level of accuracy. Researchers at the University of Michigan’s Laboratory of Integrated Brain Imaging, which aims to accelerate neuroscience and artificial intelligence through the development of novel tools, believe the system could reshape how brain imaging is handled across health systems in the United States.
The system is also capable of assessing the urgency of patient care needs, flagging life-threatening conditions for immediate attention. The development comes as demand for MRI scans continues to increase globally, placing a strain on medical resources.
