Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2406.19581 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909027333570560 |
|---|---|
| author | Clarke, Alexander Kenneth Halatsis, Dimitrios Grison, Agnese Guerra, Irene Mendez Ezaz-Nikpay, Noura Mamidanna, Pranav Ma, Shihan Muceli, Silvia Farina, Dario |
| author_facet | Clarke, Alexander Kenneth Halatsis, Dimitrios Grison, Agnese Guerra, Irene Mendez Ezaz-Nikpay, Noura Mamidanna, Pranav Ma, Shihan Muceli, Silvia Farina, Dario |
| contents | Decomposing surface electromyography (EMG) into the spike trains of individual motor neurons is a long-standing inverse problem and a key step toward motor-neuron-driven neural interfaces such as prosthetics and exoskeletons. The standard approach, independent component analysis (ICA) of the multichannel signal, assumes that the mixing from neurons to electrodes is stationary in time. This assumption fails during movement, when volume-conductor deformation makes the mixing time-varying, and current decomposition algorithms are correspondingly restricted to isometric contractions. We introduce a quasi-linear ICA formulation in which a static linear separator is preceded by a learned, low-rank, time-varying invertible transformation. The separator is trained with an independence loss on the uncompensated projection, and the transformation with a stationarity loss on the recovered source. Gradients are not shared between the two, so the source-extraction step reduces to classical linear ICA and inherits its identifiability guarantee, while non-stationary distortion is absorbed by the transformation. The closed-form inverse of the transformation enables per-spike subtraction with a time-varying template during sequential peel-off. On a public benchmark of dynamic high-density EMG with ground-truth spike trains, the method outperforms four adaptive ICA baselines at every recall threshold, recovering more units at a higher accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_19581 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Quasi-Linear ICA for Motor Unit Decomposition during Dynamic Contractions Clarke, Alexander Kenneth Halatsis, Dimitrios Grison, Agnese Guerra, Irene Mendez Ezaz-Nikpay, Noura Mamidanna, Pranav Ma, Shihan Muceli, Silvia Farina, Dario Human-Computer Interaction Machine Learning Decomposing surface electromyography (EMG) into the spike trains of individual motor neurons is a long-standing inverse problem and a key step toward motor-neuron-driven neural interfaces such as prosthetics and exoskeletons. The standard approach, independent component analysis (ICA) of the multichannel signal, assumes that the mixing from neurons to electrodes is stationary in time. This assumption fails during movement, when volume-conductor deformation makes the mixing time-varying, and current decomposition algorithms are correspondingly restricted to isometric contractions. We introduce a quasi-linear ICA formulation in which a static linear separator is preceded by a learned, low-rank, time-varying invertible transformation. The separator is trained with an independence loss on the uncompensated projection, and the transformation with a stationarity loss on the recovered source. Gradients are not shared between the two, so the source-extraction step reduces to classical linear ICA and inherits its identifiability guarantee, while non-stationary distortion is absorbed by the transformation. The closed-form inverse of the transformation enables per-spike subtraction with a time-varying template during sequential peel-off. On a public benchmark of dynamic high-density EMG with ground-truth spike trains, the method outperforms four adaptive ICA baselines at every recall threshold, recovering more units at a higher accuracy. |
| title | Quasi-Linear ICA for Motor Unit Decomposition during Dynamic Contractions |
| topic | Human-Computer Interaction Machine Learning |
| url | https://arxiv.org/abs/2406.19581 |