Bao, T orcid.org/0000-0002-1103-2660, Zaidi, SAR orcid.org/0000-0003-1969-3727, Xie, S orcid.org/0000-0002-8082-9112 et al. (2 more authors) (2021) Inter-Subject Domain Adaptation for CNN-Based Wrist Kinematics Estimation Using sEMG. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 29. pp. 1068-1078. ISSN 1534-4320
Abstract
Recently, convolutional neural network (CNN) has been widely investigated to decode human intentions using surface Electromyography (sEMG) signals. However, a pre-trained CNN model usually suffers from severe degradation when testing on a new individual, and this is mainly due to domain shift where characteristics of training and testing sEMG data differ substantially. To enhance inter-subject performances of CNN in the wrist kinematics estimation, we propose a novel regression scheme for supervised domain adaptation (SDA), based on which domain shift effects can be effectively reduced. Specifically, a two-stream CNN with shared weights is established to exploit source and target sEMG data simultaneously, such that domain-invariant features can be extracted. To tune CNN weights, both regression losses and a domain discrepancy loss are employed, where the former enable supervised learning and the latter minimizes distribution divergences between two domains. In this study, eight healthy subjects were recruited to perform wrist flexion-extension movements. Experiment results illustrated that the proposed regression SDA outperformed fine-tuning, a state-of-the-art transfer learning method, in both single-single and multiple-single scenarios of kinematics estimation. Unlike fine-tuning which suffers from catastrophic forgetting, regression SDA can maintain much better performances in original domains, which boosts the model reusability among multiple subjects.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | This is protected by copyright. All rights reserved. This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ |
Keywords: | Estimation , Wrist , Kinematics , Training , Feature extraction , Adaptation models , Transfer learning |
Dates: |
|
Institution: | The University of Leeds |
Academic Units: | The University of Leeds > Faculty of Engineering & Physical Sciences (Leeds) > School of Electronic & Electrical Engineering (Leeds) > Robotics, Autonomous Systems & Sensing (Leeds) |
Funding Information: | Funder Grant number Royal Society IE161218 EPSRC (Engineering and Physical Sciences Research Council) EP/S019219/1 |
Depositing User: | Symplectic Publications |
Date Deposited: | 02 Jun 2021 13:41 |
Last Modified: | 07 Mar 2023 19:43 |
Status: | Published |
Publisher: | IEEE |
Identification Number: | 10.1109/TNSRE.2021.3086401 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:174690 |
Download
Filename: Inter-Subject_Domain_Adaptation_for_CNN-Based_Wrist_Kinematics_Estimation_Using_sEMG.pdf
Licence: CC-BY 4.0