Computed myography: three-dimensional reconstruction of motor functions from surface EMG data

Abstract

We describe a methodology called computed myography (CMG) to qualitatively and quantitatively determine the activation level of individual muscles by voltage measurements from an array of voltage sensors on the skin surface. A finite element model for electrostatics simulation is constructed from morphometric data. For the inverse problem we utilize a generalized Tikhonov regularization. This imposes smoothness on the reconstructed sources inside the muscles and suppresses sources outside the muscles using a penalty term. Results from experiments with simulated and human data are presented for activation reconstructions of three muscles in the upper arm (biceps brachii, bracialis, and triceps). This approach potentially offers a new clinical tool to sensitively assess muscle function in patients suffering from neurological disorders (e.g., spinal cord injury), and could more accurately guide advances in the evaluation of specific rehabilitation training regimens.

Myomatrix arrays for high-definition muscle recording

“…Our approach to spike sorting shares the same ultimate goal as prior work using skin-surface electrode arrays to isolate signals from individual motor units but pursues this goal using different hardware and analysis approaches. A number of groups have developed algorithms for reconstructing the spatial location and spike times of active motor units 18,48 based on skin-surface recordings, in many cases drawing inspiration from earlier efforts to localize cortical activity using EEG recordings from the scalp 49 . Our approach differs substantially.…”

“…1c), the spikes of individual motor units appear on multiple channels and/or overlap with each other in time, requiring a more sophisticated spike sorting approach to identifying the firing times of individual motor units. We therefore adapted Kilosort version 2.5 2,47 and wrote custom MATLAB and Python code to sort waveforms into clusters arising from individual motor units (Supplemental Fig. 2e-h).…”

“…We therefore used Kilosort version 2.5 2,47 and custom MATLAB and Python code to sort waveforms into clusters arising from individual motor units (Supplemental Fig. 2e-h).…”

Myomatrix arrays for high-definition muscle recording

“…Our approach to spike sorting shares the same ultimate goal as prior work using skin-surface electrode arrays to isolate signals from individual motor units but pursues this goal using different hardware and analysis approaches. A number of groups have developed algorithms for reconstructing the spatial location and spike times of active motor units ( Negro et al, 2016 ; van den Doel et al, 2008 ) based on skin-surface recordings, in many cases drawing inspiration from earlier efforts to localize cortical activity using EEG recordings from the scalp ( Michel et al, 2004 ). Our approach differs substantially.…”