Diagnosing Parkinson's disease using features of hand-drawn spirals
Krzysztof Wróbel, Rafał Doroz · Procedia Computer Science · 2022
This paper proposes a novel method for diagnosing Parkinson's disease, based on a hand-drawn spirals and features generated from them. Analyzed spirals were drawn on a drawing tablet by both ill and healthy subjects. During drawing, coordinates of points of the spiral, pressure and angle of the pen at that point, and timestamp were registered. On the basis of the registered data, a set of features has been proposed, by means of which the classification was performed. For classification, several of the most popular machine learning methods were used, for which the accuracy of Parkinson's disease recognition was determined. The study showed that the proposed set of features enables the effective diagnosis of Parkinson's disease. The experiments were conducted on a publicly available database from the UCI archives. This database contains drawings of spirals made by people with Parkinson's disease and healthy people.