Objective: To assist non-specialists in evaluating patients with advanced Parkinson’s disease who have difficulty visiting specialists’ hospitals, we developed an algorithm to automatically rate the severity of motor symptoms of Parkinson’s disease. This study aims to assess the accuracy of this algorithm.

Background: Treatment of Parkinson’s disease requires a high level of expertise, and it is difficult for general physicians to individualize medication regimens for patients. It has been reported that Parkinson’s disease treated by specialists has a better prognosis than by general physicians. There is an unmet need for a methodology to receive specialized treatment for patients with advanced Parkinson’s disease who have difficulty visiting specialists’ hospitals.

Method: Twenty-three patients with Parkinson’s disease were recruited from the Department of Neurology outpatient clinic at Juntendo University Hospital. A neurologist assessed the study participants in person with MDS-UPDRS (MDS-Unified Parkinson’s Disease Rating Scale) part III with videotaping. Each of these videos was analyzed using video analysis software, and an algorithm for determining MDS-UPDRS part III scores based on the frequency of the movements in the videos was produced, which determined whether the score was one or below or above one. The algorithm’s accuracy was evaluated using the neurologist’s evaluation as the gold standard.

Results: The average age of the study participants was 63.3 ± 8.6 years, the male-to-female ratio was 16:7, and the average disease duration was 12.7 ± 6.0 years. The accuracy of right toe-tapping was 0.828, left toe-tapping was 0.690, right lower limb agility was 0.862, and left lower limb agility was 0.724. The results show that this algorithm can estimate whether patients with Parkinson’s disease have mild symptoms with high accuracy.

Conclusion: We developed an algorithm to automatically evaluate motor symptoms of Parkinson’s disease by means of video analysis. Further evaluation of the other items of the MDS-UPDRS part III and accumulation of data for severe cases will be required for future practical use.

« Back to 2023 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/development-and-validation-of-an-algorithm-to-automatically-assess-motor-symptoms-of-parkinsons-disease/