A low-cost and portable wrist exoskeleton using EEG-sEMG combined strategy for prolonged active rehabilitation

• DOI: 10.3389/fnbot.2023.1161187
• Publication Date: 2023-05-24T06:04:02Z
• Authors: Shiqi Yang; Min Li; Jiale Wang; Zhilei Shi; Bo He; Jun Xie; Guanghua Xu
• Keywords: Biofeedback

Hemiparesis is a common consequence of stroke that severely impacts the life quality patients. Active training key factor in achieving optimal neural recovery, but current systems for wrist rehabilitation present challenges terms portability, cost, and potential muscle fatigue during prolonged use. To address these challenges, this paper proposes low-cost, portable system with control strategy combines surface electromyogram (sEMG) electroencephalogram (EEG) signals to encourage patients engage consecutive, spontaneous sessions. In addition, detection method based on Boruta algorithm post-processing layer are proposed, allowing switch between sEMG EEG modes when occurs. This significantly improves accuracy from 4.90 10.49% four distinct motions, while selects most essential features stabilizes effects post-processing. The also presents an alternative mode employs maintain active control, approximately 80% detecting motion intention. For occurrence long term training, proposed promising approach addressing limitations existing systems.