Measuring signal fluctuations in gait rhythm time series of patients with Parkinson's disease using entropy parameters. (January 2017)

• Record Type: Journal Article
• Authors: Wu, Yunfeng; Chen, Pinnan; Luo, Xin; Wu, Meihong; Liao, Lifang; Yang, Shanshan; Rangayyan, Rangaraj M.
• Is Part Of: Biomedical signal processing and control. Volume 31(2017)
• Page Start: 265
• Page End: 271
• Publication Date: 2017-01
• Subjects: Approximate entropy -- Gait analysis -- Generalized linear regression analysis -- Parkinson's disease -- Signal turns count -- Stride time -- Symbolic entropy; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2016.08.022
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract : Highlights: Gait irregularity is measured by approximate entropy, symbolic entropy, and signal turns count. Generalized linear regression analysis and SVM are employed for gait pattern classifications. Cross-validation results are evaluated with Matthews correlation coefficient and ROC curve. Abstract: Gait rhythm disturbances due to abnormal strides indicate the degenerative mobility regulation of motor neurons affected by Parkinson's disease (PD). The aim of this work is to compute the approximate entropy (ApEn), normalized symbolic entropy (NSE), and signal turns count (STC) parameters for the measurements of stride fluctuations in PD. Generalized linear regression analysis (GLRA) and support vector machine (SVM) techniques were employed to implement nonlinear gait pattern classifications. The classification performance was evaluated in terms of overall accuracy, sensitivity, specificity, precision, Matthews correlation coefficient (MCC), and area under the receiver operating characteristic (ROC) curve. Our experimental results indicated that the ApEn, NSE, and STC parameters computed from the stride series of PD patients were all significantly larger (Wilcoxon rank-sum test: p < 0.01) than those of healthy control subjects. Based on the distinct features of ApEn, NSE, and STC, the SVM provided an accuracy rate of 84.48% and MCC of 0.7107, which are better than those of the GLRA (accuracy: 82.76%, MCC: 0.6552). The SVM and GLRA methods were able to distinguish PD gait patterns from healthy control cases with area of 0.9049 (SVM sensitivity: 0.7241, specificity: 0.9655) and 0.9037 (GLRA sensitivity: 0.8276, specificity: 0.8276) under the ROC curve, respectively, which are better or comparable with the classification results achieved by the other popular pattern classification methods.