AI-Powered MRI Technology Improves Parkinson’s Diagnoses

By MedImaging International staff writers
Posted on 20 Mar 2025

Image: The automated MRI processing and machine learning software features a noninvasive biomarker technique (Photo courtesy of University of Florida)

Current research shows that the accuracy of diagnosing Parkinson’s disease typically ranges from 55% to 78% within the first five years of assessment. This is partly due to the similarities shared by Parkinson’s and other movement disorders, which can make it challenging to make a definitive diagnosis early on. While Parkinson’s disease is widely recognized, it actually refers to several conditions, from idiopathic Parkinson’s, the most common type, to other related disorders such as multiple system atrophy Parkinsonian variant and progressive supranuclear palsy. These disorders share motor and nonmotor symptoms like changes in gait but have distinct pathologies and prognoses. Misdiagnosis occurs in roughly one in four or even one in two patients. Now, a new software tool can assist clinicians in differentiating between Parkinson’s disease and similar conditions, reducing the time needed for diagnosis and increasing accuracy to over 96%.

The software, named Automated Imaging Differentiation for Parkinsonism (AIDP), was developed by researchers at the University of Florida (Gainesville, FL, USA) and the UF Health Norman Fixel Institute for Neurological Diseases (Gainesville, FL, USA). AIDP is an automated MRI processing and machine learning program that uses a noninvasive biomarker technique. The software leverages diffusion-weighted MRI, a method that measures the diffusion of water molecules in the brain, allowing the identification of areas where neurodegeneration occurs. The machine learning algorithm, tested rigorously against clinical diagnoses, analyzes these brain scans and provides results, indicating the specific type of Parkinson’s disease present.

While no tool can replace the human aspect of diagnosis, even highly experienced physicians who specialize in movement disorders can benefit from this tool, which enhances diagnostic accuracy across different conditions, according to the researchers. The software was tested in a study conducted across 21 sites, 19 in the United States and two in Canada, with the findings published in JAMA Neurology. The next step for the team is to seek approval from the U.S. Food and Drug Administration (FDA).

“In many cases, MRI manufacturers don’t communicate with each other due to marketplace competition,” said David Vaillancourt, Ph.D., chair and a professor in the UF Department of Applied Physiology and Kinesiology. “They all have their own software and their own sequences. Here, we’ve developed novel software that works across all of them.”

“This is an instance where the innovation between technology and artificial intelligence has been proven to enhance diagnostic precision, allowing us the opportunity to further improve treatment for patients with Parkinson’s disease,” added Michael Okun, M.D., medical adviser to the Parkinson’s Foundation and director of the Norman Fixel Institute for Neurological Diseases at UF Health.