Characterization of the Handwriting Skills as a Biomarker for Parkinson Disease

TL;DR

This study evaluates handwriting patterns as potential biomarkers for Parkinson's disease, introducing a new dataset and analyzing various features and classifiers to improve detection accuracy.

Contribution

It presents a new, sizable handwriting database for PD research and assesses multiple features and classifiers for effective PD detection.

Findings

Detection accuracy between 81% and 97%.

Handwriting features can effectively distinguish PD patients from healthy controls.

New insights into feature utility for PD monitoring.

Abstract

In this paper we evaluate the suitability of handwriting patterns as potential biomarkers to model Parkinson disease (PD). Although the study of PD is attracting the interest of many researchers around the world, databases to evaluate handwriting patterns are scarce and knowledge about patterns associated to PD is limited and biased to the existing datasets. This paper introduces a database with a total of 935 handwriting tasks collected from 55 PD patients and 94 healthy controls (45 young and 49 old). Three feature sets are extracted from the signals: neuromotor, kinematic, and nonlinear dynamic. Different classifiers are used to discriminate between PD and healthy subjects: support vector machines, knearest neighbors, and a multilayer perceptron. The proposed features and classifiers enable to detect PD with accuracies between 81% and 97%. Additionally, new insights are presented on the utility of the studied features for monitoring and detecting PD.