Abstract
Upper limb amputation significantly affects daily functioning and quality of life. Although myoelectric prostheses offer a promising avenue for restoring motor capabilities, high rates of device abandonment underscore challenges in control performance and user integration. Recent advances in high-density electromyography (HD-EMG) and machine learning (ML) algorithms have shown potential to enhance prosthetic dexterity. HD-EMG interfaces capture richer spatial and temporal muscle activation data, while ML algorithms exploit this information to improve intention detection and motion control. This mini-review explores advancements in HD-EMG acquisition systems, including both interface designs and recording technologies, as well as ML algorithms leveraging spatial information. In addition to summarizing the current state of the art, we discuss the challenges and the opportunities of embedding these technologies in prosthetic systems, with the objective of facilitating the translation of laboratory research into clinical applications.