The latest artificial intelligence system developed at the University of Michigan can analyze brain MRI scans and deliver a diagnosis in just a few seconds. According to recent studies, this model identifies neurological diseases with up to 97.5% accuracy and can assess the urgency of medical intervention required for a patient.

Impact on Medical Diagnostics

Researchers note that this unique technology has the potential to transform how brain imaging is processed in U.S. healthcare facilities. With the growing global demand for MRIs and increasing workloads for doctors and healthcare systems, implementing such an AI model could significantly reduce strain by providing rapid and accurate diagnostics, as well as timely notifications to specialists.

Prima Technology: Features and Advantages

The new system, named Prima, was tested over the course of a year on more than 30,000 MRI studies covering over 50 different radiological diagnoses related to major neurological disorders. Prima demonstrated superior diagnostic performance compared to other modern AI models. In addition to detecting diseases, the system can determine which cases require higher priority.

Certain neurological conditions, such as strokes and brain hemorrhages, demand immediate intervention. In these situations, Prima automatically notifies the relevant specialists—such as a stroke neurologist or neurosurgeon—so that action can be taken quickly. Feedback becomes available immediately after the patient’s examination is completed.

Speed and Accuracy: Key Benefits

Accuracy is crucial when interpreting brain MRIs, but the speed of obtaining results is also critical for timely diagnosis and improved treatment outcomes. At key stages of the process, Prima can optimize workflows and enhance clinical efficiency without sacrificing precision.

How the Model Works and Is Trained

Prima is classified as a visual-language model (VLM)—a type of artificial intelligence capable of processing images, video, and text in real time. Unlike previous models that were trained on limited and carefully curated datasets for narrow tasks, Prima was trained on a much broader array of information.

The team utilized all available MRIs collected since the digitization of radiology records at the University of Michigan—over 200,000 studies and 5.6 million image sequences. The model also took into account patients’ clinical histories and the reasons for each examination.

Integration with Clinical Practice

Prima functions like a radiologist, integrating patient medical history and imaging results to form a comprehensive understanding of health status. This approach ensures high accuracy across a wide range of diagnostic tasks.

Addressing Current Healthcare Challenges

Millions of MRIs are performed worldwide each year, many related to neurological diseases. The demand for such studies is growing faster than the availability of neuroradiology services, leading to staff shortages, diagnostic delays, and errors. Depending on the location, results may take several days or even longer to become available.

Innovative technologies are essential to improve access to radiology services, both in large, high-volume medical centers and in rural hospitals with limited resources. The development of Prima represents a cutting-edge solution with significant potential for scaling.

Development Prospects

Despite its impressive results, researchers emphasize that the work is still in the early stages of evaluation. In the future, there are plans to incorporate more detailed patient information and electronic medical record data to further enhance diagnostic accuracy. This approach mirrors real-world clinical practice, where doctors interpret MRIs and other studies based on complex data.

Potential for the Future of Medicine

Prima has already been dubbed the “ChatGPT for medical imaging,” with experts noting that similar technologies could be adapted for other types of studies, including mammography, chest X-rays, and ultrasound. Just as AI tools help draft letters or provide recommendations, Prima is designed to become a “copilot” for interpreting medical images. This system demonstrates the transformative potential of integrating healthcare and AI to advance medical care through innovation.