A Multidataset Characterization of Window-Based Hyperparameters for Deep CNN-Driven sEMG Pattern Recognition

• Published in: IEEE transactions on human-machine systems Vol. 54; no. 1; pp. 131 - 142
• Main Authors: Kulwa, Frank, Zhang, Haoshi, Samuel, Oluwarotimi Williams, Asogbon, Mojisola Grace, Scheme, Erik, Khushaba, Rami, McEwan, Alistair A., Li, Guanglin
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Artificial neural networks; Convolution neural network (CNN); Convolutional neural networks; Decoding; Electromyography; Feature extraction; Kernels; Myoelectric control; Myoelectricity; Neural networks; Pattern recognition; Prostheses; Prosthetics; surface electromyogram (sEMG) signal; upper limb; window length; window overlap; Windows

The control performance of myoelectric prostheses would not only depend on the feature extraction and classification algorithms but also on interactions of dynamic window-based hyperparameters (WBHP) used to construct input signals. However, the relationship between these hyperparameters and how the...