Background: Surface stimulation systems facilitate dexterous manipulation by achieving targeted and isolated activation of muscle groups through motor-point-based stimulation. Existing catalogs on motor points lack generalization and reproducibility, as they are mostly based on anatomical charts and were obtained from heterogeneous studies.
Objective: By systematically identifying and characterizing the motor points, the aim of this study is to address these limitations and improve the utilization of motor point catalogs toward the design and control for surface stimulation systems, which are targeted to restore complete hand function.
Methods: Sites that allowed motor-point-based stimulation were identified among nine healthy participants. Using bipolar stimulation, a tracing electrode was used to locate these sites along the forearm surface, and the muscle response to motor-point-based stimulation was also graded using isokinetic dynamometry. Ultimately, using machine-learning-based clustering algorithms, the motor point locations were grouped into clusters, and their centroids and confidence regions were derived.
Results: Such experimentally derived clusters had physiological correlations, and further cross validation was also in agreement with two test subjects.
Conclusion: By clustering motor point locations, the potential for deriving a generalized catalog has been demonstrated. With current literature lacking such data, the novelty of this study lies in the representation of baseline information on location, shape, and the recruitment of stimulation zones for various muscle groups using bipolar stimulation.
Significance: This information can improve the design of electrode arrays and existing stimulation mapping algorithms, and aid clinicians toward electrode placement for patient-specific treatments.