Researchers at The University of Texas Health Science Center at San Antonio have developed a groundbreaking artificial intelligence (AI) tool capable of accurately counting brain lesions on MRI scans within seconds. This innovative tool is expected to play a vital role in helping neuroradiologists assess patients’ brain diseases at earlier stages.

Led by researcher Mohamad Habes, PhD, from the Glenn Biggs Institute for Alzheimer’s and Neurodegenerative Diseases, the team demonstrated the AI tool’s effectiveness in identifying and counting enlarged perivascular spaces (ePVS’s). These spaces, filled with cerebrospinal fluid, surround arteries and veins and serve as a marker for cerebral small-vessel disease, which can lead to stroke and dementia. The study involved a follow-up analysis of 1,026 individuals who participated in the Multi-Ethnic Study of Atherosclerosis (MESA).

Previously, quantifying ePVS’s on MRI scans was a challenging task, leading to their general neglect. A middle-aged person might have around 500 to 600 of these small spaces on an MRI, making manual counting impractical and time-consuming for neuroradiologists in busy clinics.

However, the AI tool’s deep-learning capabilities have changed the game. The researchers trained the algorithm with expert knowledge, enabling it to automatically quantify ePVS’s with precision, mapping the patient’s small-vessel disease. The AI tool identifies and counts the spaces, providing more detailed information than a human could achieve.

The study highlighted the significance of enlarged perivascular spaces in two brain regions, the basal ganglia and the thalamus, which showed associations with stroke and small-vessel disease markers. The basal ganglia, related to neurodegenerative disorders and motor functions, and the thalamus, associated with sensory functions, are crucial in understanding brain health.

The team aims to further study the AI tool at Alzheimer’s Disease Research Centers (ADRCs), designated U.S. National Institute on Aging Centers of Excellence. The South Texas ADRC, in collaboration with UT Health Science Center San Antonio’s Biggs Institute and The University of Texas Rio Grande Valley, has shown keen interest in the AI methodology, which could significantly benefit large-scale studies conducted across the nation’s ADRCs.

The AI tool harnesses the computational power of UT Health Science Center San Antonio’s Genie supercomputer, offering new possibilities for early detection and intervention in brain diseases. As AI continues to advance, this tool holds great promise for revolutionizing the way brain lesions are identified and managed in clinical practice.

By Impact Lab