Determination of Optimum Segmentation Schemes for Pattern Recognition-Based Myoelectric Control: A Multi-Dataset Investigation

• Published in: IEEE access 2020, Vol.8, p.90862-90877
• Main Authors: Ashraf, Hassan, Waris, Asim, Jamil, Mohsin, Gilani, Syed Omer, Niazi, Imran Khan, Kamavuako, Ernest Nlandu, Gilani, Syed Hammad Nazeer
• Format: Article
• Language: English
• Subjects: Algorithms; Classification; Computer Science (all); Control systems; Datasets; Discriminant analysis; Electrical & electronics engineering; Electromyography; Engineering (all); Engineering, computing & technology; General Computer Science; General Engineering; General Materials Science; Ingénierie électrique & électronique; Ingénierie, informatique & technologie; Linear discriminant analysis; Machine learning; Materials Science (all); Mean classification; Motion artifacts; Motion segmentation; Multiple data sets; Myoelectric control; Myoelectricity; Pattern recognition; Performance measure; Post-processing techniques; Recording devices; Robustness; Segmentation; Segmentation scheme; Segments; Statistical analysis; Support vector machines; Windowing
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2020.2994829

Pattern recognition (PR) algorithms have shown promising results for upper limb myoelectric control (MEC). Several studies have explored the efficacy of different pre and post processing techniques in implementing PR-based MECs. This paper explores the effect of segmentation type (disjoint and overlap) and segment size on the performance of PR-based MEC, for multiple datasets recorded with different recording devices. Two PR-based methods; linear discriminant analysis (LDA) and support vector machine (SVM) are used to classify hand gestures. Optimum values of segment size, step size and segmentation type were considered as performance measure for a robust MEC. Statistical analysis showed that optimum values of segment size for disjoint segmentation are between 250ms and 300ms for both LDA and SVM. For overlap segmentation, best results have been observed in the range of 250ms-300ms for LDA and 275ms-300ms for SVM. For both classifiers the step size of 20% achieved highest mean classification accuracy (MCA) on all datasets for overlap segmentation. Overall, there is no significant difference in MCA of disjoint and overlap segmentation for LDA (P-value = 0.15) but differ significantly in the case of SVM (P-value < 0.05). For disjoint segmentation, MCA of LDA is 88.68% and for SVM, it is 77.83%. Statistical analysis showed that LDA outperformed SVM for disjoint segmentation (P-value