Assessing Neuromorphic Computing for Fingertip Force Decoding from Electromyography
This work addresses the problem of mapping neural activity to motor intent for assistive and rehabilitation control, but it is incremental as it assesses an existing neuromorphic approach against a standard method.
The study tackled decoding fingertip force from electromyography by comparing a spiking neural network (SNN) to a temporal convolutional network (TCN), finding that the TCN achieved 4.44% MVC RMSE with Pearson r = 0.974, while the SNN achieved 8.25% MVC with r = 0.922.
High-density surface electromyography (HD-sEMG) provides a noninvasive neural interface for assistive and rehabilitation control, but mapping neural activity to user motor intent remains challenging. We assess a spiking neural network (SNN) as a neuromorphic architecture against a temporal convolutional network (TCN) for decoding fingertip force from motor-unit (MU) firing derived from HD-sEMG. Data were collected from a single participant (10 trials) with two forearm electrode arrays; MU activity was obtained via FastICA-based decomposition, and models were trained on overlapping windows with end-to-end causal convolutions. On held-out trials, the TCN achieved 4.44% MVC RMSE (Pearson r = 0.974) while the SNN achieved 8.25% MVC (r = 0.922). While the TCN was more accurate, we view the SNN as a realistic neuromorphic baseline that could close much of this gap with modest architectural and hyperparameter refinements.