当前位置： SCI文献检索 > Journal of Neural Engineering期刊下所有文献 > Evaluation of windowing techniques for intramuscular EMG-based diagnostic, rehabilitative and assistive devices.

Evaluation of windowing techniques for intramuscular EMG-based diagnostic, rehabilitative and assistive devices.

Abstract：

OBJECTIVE:Intramuscular electromyography (iEMG) signals, non-invasively recorded, directly from the muscles are used to diagnose various neuromuscular disorders/diseases, also to control rehabilitative and assistive robotic devices. iEMG signals are being potentially used in neurology, kinesiology, rehabilitation, and ergonomics, to detect/diagnose various diseases/ disorders. Electromyography (EMG) based classification systems are being designed and tested for classification of various neuromuscular disorders and to control rehabilitative and assistive robotic devices. Many studies have explored parameters, such as pre-processing, feature extraction and selection of classifier that can affect the performance and efficacy of iEMG-based classification systems. Pre-processing stage includes removal of any unwanted noise from original signal and windowing of the signal. APPROACH:This study investigated and presented optimum windowing configurations for robust control and better classification results of iEMG-based classification system. Both, disjoint and overlap, windowing techniques with varying window and overlap sizes have been investigated using a machine learning (ML) based classification algorithm called linear discriminant analysis (LDA). MAIN RESULTS:The optimum window size ranges are from 200ms to 300ms for disjoint and 225ms to 300ms for overlap windowing technique, respectively. The inferred results show that for overlap windowing technique the optimum range of overlap size is from 10% to 40% of the length of a window size. Mean classification accuracy (MCA) was found to be lower in disjoint windowing technique as compared to overlap windowing at all investigated overlap sizes. Statistical analysis (two-way analysis of variance test) showed that MCA of overlap windowing technique was significantly different at overlap sizes of 10% to 40% (p-values < 0.05). SIGNIFICANCE:These results can be used to achieve best possible classification performance for any iEMG based real-time diagnosis, detection, and control system.

journal_name

J Neural Eng

journal_title

Journal of neural engineering

authors

Ashraf H,Waris A,Gilani SO,Kashif AS,Jamil M,Jochumsen M,Niazi IK

doi

10.1088/1741-2552/abcc7f

subject

Has Abstract

pub_date

2020-11-20 00:00:00

eissn

1741-2560

issn

1741-2552

pub_type

杂志文章

在线工具

Evaluation of windowing techniques for intramuscular EMG-based diagnostic, rehabilitative and assistive devices.