Pattern analysis of computer keystroke time series in healthy control and early-stage Parkinson's disease subjects using fuzzy recurrence and scalable recurrence network features

• Published in: Journal of neuroscience methods Vol. 307; pp. 194 - 202
• Main Author: Pham, Tuan D.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.09.2018
• Subjects: Early Parkinson's disease; Fuzzy recurrence plots; Keystroke time series; Noncommutative catenoid; Noncommutative curvature; Noncommutative minimal surface; Noncommutative Riemannian geometry; Pattern classification; Scalable recurrence networks; Texture analysis; Keystroke time series; Texture analysis; Early Parkinson's disease; Fuzzy recurrence plots; Pattern classification; Scalable recurrence networks
• ISSN: 0165-0270; 1872-678X; 1872-678X
• DOI: 10.1016/j.jneumeth.2018.05.019

•The proposed methods provide the best classification results.•For the first time, 2D analysis of computer keystroke time series is carried out.•For the first time, scalable recurrence networks are applied to studying real-life time series. Identifying patients with early stages of Parkinson's disease (PD) in a home environment is an important area of neurological disorder research, because it is of therapeutic and economic benefits to optimal intervention and management of the disease. This paper presents a nonlinear dynamics approach, including recurrence plots, recurrence quantification analysis, fuzzy recurrence plots, and scalable recurrence networks for visualization, classification, and characterization of keystroke time series obtained from healthy control (HC) and early-stage PD subjects. Several differentiative properties for characterizing early PD and HC subjects can be obtained from fuzzy recurrence plots (FRPs) and scalable recurrence networks. Comparison with existing methods: cross-validation results obtained from FRP-based texture are highest among other methods. The method of fuzzy recurrence plots outperforms other existing methods for classification of HC and PD subjects. Features extracted from the nonlinear dynamics analysis of the keystroke time series are found to be very effective for machine learning and the properties of the scalable recurrence networks have the potential to be utilized as physiologic markers of the disease.