Abstract
Peripheral nerve interfaces (PNIs) facilitate communication with the peripheral nervous system and have diverse applications in bioelectronic medicine and neuroprosthetics. These interfaces can regulate neural activity through stimulation or monitor physiological conditions by recording peripheral nerve signals. With the rapid advancement of machine learning (ML), various ML techniques have been applied to PNIs, particularly for classification and regression tasks. However, the scope of ML integration in PNIs and the range of applicable techniques remain largely undocumented. To address this gap, we conducted a scoping review to assess the current state of ML in the PNI field. A comprehensive search across five databases identified 63 relevant studies after full-text review. Most studies employed supervised learning approaches for activity classification, with neural networks—such as artificial neural networks, convolutional neural networks, and recurrent neural networks—being the most commonly used models. In contrast, unsupervised, semi-supervised, and reinforcement learning approaches remain underutilized, presenting opportunities for further advancements in this domain.