New AI tool being tested in UF Health study to improve Parkison's diagnosis

University of Florida researchers will test an artificial intelligence tool to improve diagnosis for early cases of Parkinson's disease.

Research has shown the accuracy of diagnosis in the early stages of Parkinson's disease is as low as 58 percent, according to a March 18 news release from UF Health. There are three neurodegenerative disorders that share overlapping motor dysfunction features, which contribute to misdiagnosis. Researchers from Gainesville-based UF are testing an AI tool to change that.  

Physicians at 21 sites will upload MRI images from 315 participating patients. The tool will calculate a diagnosis, which will be compared with the official diagnosis. The study is expected to begin this summer and take two years to complete.

The research will be funded using a $5 million grant from the National Institutes of Health.

"This isn't going to replace the physician's decision-making; it's just meant to be another tool in their toolkit," David Vaillancourt, PhD, chair of UF's department of applied physiology and kinesiology and a principal investigator on the study, told UF Health. "The goal is that clinical trials will be better because they will focus on specific variants. Patients will be able to know their diagnosis earlier."

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.