A computational framework for EMG simulation and coherence-based biomarker analysis on neural crosstalk in bimanual
Osmar Pinto Neto1, Tatiana O R Pinho2, Yiyu Wang3
1Department of Kinesiology, California State University, San Marcos, CA, USA; Department of Biomedical Engineering, Anhembi Morumbi University, SP, Brazil; Center of Innovation, Technology and Education-CITÉ, São José dos Campos, SP, Brazil.
Background And Objective:
This study presents a computationally and physiologically grounded framework for electromyography (EMG) simulation to examine how assumed cortical and subcortical inputs shape bilateral EMG and intermuscular coherence during bimanual coordination.
Methods:
Simulated first dorsal interosseous (FDI) EMG was generated for a 1:2 force coordination task (approximately 40% maximum voluntary contraction) using motor unit models incorporating (i) variable cortical and subcortical drive weighting, (ii) narrowband versus broadband cortical drive (13-30 vs. 13-60 Hz), and (iii) asymmetric bilateral coupling. Intermuscular EMG-EMG wavelet coherence was computed across α, β, low-γ, and high-γ bands. Model outputs were validated against experimental human EMG at the signal level, comparing simulated and recorded bipolar first dorsal interosseous traces across eight amplitude and frequency-domain signal features.
Results:
Coherence magnitude increased with greater modeled neural crosstalk and cortical weighting. The frequency band most sensitive to drive weighting depended on cortical bandwidth: β-band coherence (13-30 Hz) best differentiated cortical and subcortical influences under narrowband drive, whereas low-γ (30-60 Hz) was most informative under broadband conditions. α-band coherence showed minimal sensitivity to subcortical modulation in this context.
Conclusions:
Within this simulation framework, intermuscular coherence was sensitive to modeled bilateral coupling and cortical-subcortical weighting, with diagnostic sensitivity depending on the spectral bandwidth of cortical drive. Thus, the model should be interpreted as a hypothesis generating, signal level framework rather than a direct neural measurement. This open-source framework bridges computational modeling and biomarker signal analysis, supporting reproducible, hypothesis driven biomarker development for neuromotor disorders such as Parkinson's disease, spinal cord injury, and stroke.
More Related Videos
08:09Multifunctional Setup for Studying Human Motor Control Using Transcranial Magnetic Stimulation, Electromyography, Motion Capture, and Virtual Reality
Published on: September 3, 2015
08:45Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
Published on: October 24, 2012