Yao, P.; Wang, K.; Xia, W.; Guo, Y.; Liu, T.; Han, M.; Gou, G.; Liu, C.; Xue, N. Effects of Training and Calibration Data on Surface Electromyogram-Based Recognition for Upper Limb Amputees. Sensors2024, 24, 920.
Yao, P.; Wang, K.; Xia, W.; Guo, Y.; Liu, T.; Han, M.; Gou, G.; Liu, C.; Xue, N. Effects of Training and Calibration Data on Surface Electromyogram-Based Recognition for Upper Limb Amputees. Sensors 2024, 24, 920.
Yao, P.; Wang, K.; Xia, W.; Guo, Y.; Liu, T.; Han, M.; Gou, G.; Liu, C.; Xue, N. Effects of Training and Calibration Data on Surface Electromyogram-Based Recognition for Upper Limb Amputees. Sensors2024, 24, 920.
Yao, P.; Wang, K.; Xia, W.; Guo, Y.; Liu, T.; Han, M.; Gou, G.; Liu, C.; Xue, N. Effects of Training and Calibration Data on Surface Electromyogram-Based Recognition for Upper Limb Amputees. Sensors 2024, 24, 920.
Abstract
The surface electromyogram (sEMG)--based gesture recognition to the development of intelligent prostheses has emerged as a promising avenue for upper limb amputees. However, the temporal variations in sEMG have rendered recognition models less efficient than anticipated. Cross-session calibration and training data indicate the capacity to reduce this variation. The impact of varying calibration and training data amounts on gesture recognition performance for amputees is still unknown. To assess these effects, we present four datasets for the evaluation of calibration data and examine the impact of training data volume on benchmark performance. Two amputees who had undergone amputations years prior were recruited and seven sessions of data were collected for analysis from each of them. The experimental results show that the calibration data improved the average accuracy per subject by 3.03% and 6.16%, respectively, compared to the baseline results. Moreover, it was found that to collect training data for modeling, boosting the number of training sessions proves more efficient than increasing the number of trials within a single session. Two potentially effective strategies are proposed in light of this research to enhance the cross-session models further. We consider these findings to be of the utmost importance for the commercialization of intelligent prostheses, as they demonstrate the criticality of gathering calibration and cross-session training data, as well as offer effective strategies for maximizing the utilization of the entire dataset.
Keywords
surface electromyogram; cross-session gesture recognition; model calibration; amputee; intelligent prostheses
Subject
Engineering, Bioengineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.