Random Channel Masks for Regularization of Least Squares-Based Finger EMG-Force Modeling to Improve Cross-Day Performance
Open Access
- 27 July 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Systems and Rehabilitation Engineering
- Vol. 30 (15344320) , 2157-2167
Abstract
Estimating the finger forces from surface electromyography (sEMG) is essential for diverse applications (e.g., human-machine interfacing). The performance of pre-trained sEMG-force models degenerates significantly when applied on a second day, due to the large cross-day variation of sEMG characteristics. Previous studies mainly employed transfer learning algorithms to tackle this problem. However, transfer learning algorithms normally require data collected on the second day for model calibration, increasing the inconvenience in practical use. In this work, we investigated the effect of model regularization on this issue. Specifically, 256-channel high-density sEMG (HDsEMG) signals with varying finger forces were collected on different days (3–25 days apart). We applied randomly generated channel perturbations (“masks”) to feature maps of randomly selected channels in training dataset. The channel masks of the training set were generated randomly and independently in each narrow time window (~20 ms). We assumed that by learning from randomly masked feature maps (randomness is the central aspect), the model would not be biased by a small number of features but would be based on learning from a global perspective, therefore avoiding overfitting to the within-day EMG patterns. Moore-Penrose inverse model regularization was also employed as a baseline method, with results showing that cross-day EMG-force models require a higher tolerance parameter compared with within-day applications. In combination with the Moore-Penrose inverse model regularization, further applying random channel masks to the training set significantly improved model performance in cross-day validation.Keywords
Funding Information
- National Natural Science Foundation of China (62001122)
- Shanghai Municipal Science and Technology Project (20510710500)
- Natural Science Foundation of Shanghai Municipality (20ZR1403400)
This publication has 27 references indexed in Scilit:
- Transfer Learning for Rapid Re-calibration of a Myoelectric Prosthesis After Electrode Shift2016
- Improving the Robustness of Myoelectric Pattern Recognition for Upper Limb Prostheses by Covariate Shift AdaptationIEEE Transactions on Neural Systems and Rehabilitation Engineering, 2015
- Hand Grip-EMG Muscle ResponseActa Physica Polonica A, 2014
- EMG feature evaluation for improving myoelectric pattern recognition robustnessExpert Systems with Applications, 2013
- The Moore–Penrose Pseudoinverse: A Tutorial Review of the TheoryBrazilian Journal of Physics, 2011
- Domain Adaptation via Transfer Component AnalysisIEEE Transactions on Neural Networks, 2010
- The linear electrode array: a useful tool with many applicationsJournal of Electromyography and Kinesiology, 2003
- A clinically applicable EMG–force model to quantify active stabilization of the knee after a lesion of the anterior cruciate ligamentClinical Biomechanics, 2003
- The generalized Moore-Penrose inverseLinear Algebra and its Applications, 1992
- Principal component analysisChemometrics and Intelligent Laboratory Systems, 1987