Incremental Deep Learning for Single-Finger Myoelectric Control: A Comparative Closed-Loop Study of CNN and LSTM Architectures

Braun H, Okhravi M, Castellini C (2026)

Publication Language: English

Publication Type: Conference contribution

Publication year: 2026

Publisher: IEEE Computer Society

Pages Range: 1001-1008

Conference Proceedings Title: Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics

Event location: Edmonton, AB CA

ISBN: 979-8-3315-4814-8

DOI: 10.1109/BioRob66782.2026.11681092

Abstract

Target achievement control tests provide a standardized way to quantify closed-loop myoelectric control during continuous target acquisition. We compared two electromyography-to-kinematics regression controllers for single-finger control using 32-channel surface electromyography: a pretrained dilated residual convolutional neural network regressor (996k parameters) and a lightweight long shortterm memory regressor without pretraining (65k parameters). Ten non-disabled participants completed a counterbalanced within-subject crossover study. For each controller condition, participant-specific training was performed and target achievement control was evaluated in two blocks: before and after a targeted incremental fine-tuning step using additional data collected from difficult targets. Performance was assessed using completion rate, median time-to-success, overshoot count, and time spent outside the target region after first entry. After participant-specific training, both controllers achieved comparable target achievement control performance across all primary outcomes (| dz| ≤ 0.16, all p>0.6). As the CNN was pretrained on only six participants, this equivalence need not hold at larger pretraining scales. Incremental fine-tuning produced large improvements for both controllers (dz>0.8), increasing completion rate by 12-17 percentage points and reducing movement time by 1.0 - 1.4s and post-entry instability by 3-5 s per target. These results suggest that participantspecific incremental learning can enable a lightweight model without pretraining to match a deeper pretrained architecture in single-finger myoelectric control, while targeted fine-tuning provides a robust mechanism to improve speed and stability during use.

Authors with CRIS profile

Hannah Braun Assistive Intelligent Robotics Lab Claudio Castellini Assistive Intelligent Robotics Lab

How to cite

APA:

Braun, H., Okhravi, M., & Castellini, C. (2026). Incremental Deep Learning for Single-Finger Myoelectric Control: A Comparative Closed-Loop Study of CNN and LSTM Architectures. In Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics (pp. 1001-1008). Edmonton, AB, CA: IEEE Computer Society.

MLA:

Braun, Hannah, Mahdi Okhravi, and Claudio Castellini. "Incremental Deep Learning for Single-Finger Myoelectric Control: A Comparative Closed-Loop Study of CNN and LSTM Architectures." Proceedings of the 11th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob 2026, Edmonton, AB IEEE Computer Society, 2026. 1001-1008.

BibTeX: Download