Surface EMG Wristbands for Gesture Recognition A Critical Review of Cost and Reproducibility
Year: 2026
Type: preprint
Abstract: Surface electromyography (sEMG) offers a wearable route to gesture-based interaction by recording electrical activity associated with muscle activation. Translating laboratory recognition results into affordable, reproducible interfaces requires more than selecting an electrode material or a classifier with high reported accuracy. This focused narrative review examines selected acquisition systems, dry electrodes, embedded signal chains, and decoding studies through the linked questions of cost, reproducibility, and evaluation validity. Published examples illustrate how channel topology, sampling bandwidth, electrode contact, processing location, and calibration shape design choices. Two comparison tables distinguish acquisition specifications and accessible research artifacts from task-specific recognition outcomes. The analysis identifies recurring barriers to comparison: inconsistent cost definitions, incomplete descriptions of data partitions, differences between window classification and continuous interaction, and the substitution of component specifications for system-level evidence. A requirements-based decision framework is proposed for individual developers and small laboratories. The review is nonexhaustive and does not provide a meta-analysis, new measurements, or predicted classification accuracies. Its contribution is a practical synthesis of what selected published evidence supports and what must remain an explicit design assumption. This manuscript is a preprint and has not been peer reviewed. OpenAI Codex was used for manuscript revision, organization of the comparative synthesis, and checks against cited primary sources. No new experimental data were generated.
Author: Koshmambetov, Aimer
Language: English
Subfield: Biomedical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7221762885
Cites:
Cited by:
Related to: