.png){kind=icon}
.png){kind=icon}
Return
From muscle potentials to computer interfaces: A multiscale review of electromyography for clinical and computational applications
.png){kind=icon}
OA
AI
V
S
K
A
R
M
M
H
DOI：10.1016/j.sbsr.2026.100969.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
EMG has been developed as a result of the early physiological experimentation into a primary modality of neuromuscular function characterization, clinical diagno-sis, rehabilitation planning, and human-machine interface. This Review summarizes the history, methodological basis, and computational advances, as well as biomedical applications of EMG, along the entire acquisition-to-interpretation spectrum. The discussion will follow the historical process of increasing the sophistication of early biophysical measurements through surface and intramuscular recording paradigmsthe relevant engineering optimizations, such as optimizing electrode to skin inter-face, signal conditioning, artifact reduction, sampling fidelity, and normalization processes, that cumulatively realize the modern high-quality EMG recording. The review also summarizes the wide range of clinical and biomedical areas influenced by EMG such as the electrophysiological diagnostics, motor control mea-sures, movement-disorder measurements, neuroprosthetic and exoskeletal controls, and the neurorehabilitation based on biofeedback. The advances in analytics are discussed extensively, including handcrafted temporal, spectral and time-frequency representations; hybrid feature-fusion approaches; classical machine-learning classi-fication paradigms and more recent developments in deep learning, transfer learning, through this synthesis, the Review will provide a rigorous and thorough platform of advances that can further advance EMG to highly intelligent, adaptive, and clinically embedded applications in modern biomedicine.
Keywords:
Electromyography (EMG)
EMG signal processing
High-density EMG
Motor unit physiology
Computational modeling
Machine learning for biosignals
Pattern recognition
Human–machine interfaces
Neural and biomedical interfaces
Medical informatics
Rehabilitation engineering
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.9
Papers:
736
Citations:
2.6K
View more organizations