A new artificial intelligence tool has been developed by researchers at the University of Rochester to help people with Parkinson’s disease assess the severity of their symptoms remotely. This tool, described in a study in npj Digital Medicine, allows users to tap their fingers in front of a webcam and receive a motor performance assessment on a scale of 0-4.
The Significance of the AI Tool
Doctors typically have patients perform simple motor tasks to assess movement disorders and determine their severity using guidelines like the Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS). The AI model in this study offers a rapid assessment based on the MDS-UPDRS guidelines. It automatically generates computational metrics, such as speed, amplitude, frequency, and period, that are consistent with medical guidebooks. These attributes are used to classify the severity of tremors.
Comparing AI Ratings with Human Experts
The finger-tapping task was performed by 250 participants with Parkinson’s disease worldwide. The AI system’s ratings were then compared with those given by three neurologists and three primary care physicians. The results show that while expert neurologists performed slightly better than the AI model, the AI model outperformed the primary care physicians with UPDRS certification.
Benefits of the AI-Based Severity Test
The AI-based Parkinson’s disease severity test not only generates computational metrics for assessing tremor severity but also provides an alternative option for patients who struggle to access neurologists for appointments and hospital visits. Ehsan Hoque, an associate professor in Rochester’s Department of Computer Science, highlights the potential impact of this technology on improving health equity and access outside of the clinic.
Md. Saiful Islam, a Google PhD fellow and a graduate student in computer science, along with a team of computer scientists, led the study in collaboration with members of the Medical Center’s Department of Neurology. They believe their approach can be applied to other motor tasks, expanding its potential in evaluating movement disorders such as ataxia and Huntington’s disease.
While the new Parkinson’s disease assessment tool is available online, the researchers caution that it should not be considered a definitive measure of the disease’s presence or severity without input from a physician. They acknowledge that this technology is still in its early stages and should be used in conjunction with a healthcare professional’s expertise.