sEMG: A Window Into Muscle Work, but Not Easy to Teach and Delicate to Practice-A Perspective on the Difficult Path to a Clinical Tool

Abstract

Surface electromyography (sEMG) may not be a simple 1,2,3 (muscle, electrodes, signal)-step operation. Lists of sEMG characteristics and applications have been extensively published. All point out the noise mimicking perniciousness of the sEMG signal. This has resulted in ever more complex manipulations to interpret muscle functioning and sometimes gobbledygook. Hence, as for all delicate but powerful tools, sEMG presents challenges in terms of precision, knowledge, and training. The theory is usually reviewed in courses concerning sensorimotor systems, motor control, biomechanics, ergonomics, etc., but application requires creativity, training, and practice. Software has been developed to navigate the essence extraction (step 4); however, each software requires some parametrization, which returns back to the theory of sEMG and signal processing. Students majoring in Ergonomics or Biomedical Engineering briefly learn about the sEMG method but may not necessarily receive extensive training in the laboratory. Ergonomics applications range from a simple estimation of the muscle load to understanding the sense of effort and sensorimotor asymmetries. In other words, it requires time and the basics of multiple disciplines to acquire the necessary knowledge and skills to perform these studies. As an example, sEMG measurements of left/right limb asymmetries in muscle responses to vibration-induced activity of proprioceptive receptors, which vary with gender, provide insight into the functioning of sensorimotor systems. Beyond its potential clinical benefits, this example also shows that lack of testing time and lack of practitioner's sufficient knowledge are barriers to the utilization of sEMG as a clinical tool.

Keywords: clinical application; education; force control; hand dominance; sensorimotor asymmetries; sensorimotor system gain; surface electromyography.

Figure 1

The EMG forest or forest EMG. After adjusting the color (B), the tree line, and its reflection over the dark water of the lake (unknown photographer) at sunset (A) produce an EMG-like signal (C). Note that counting/identifying trees in the forest profile can be achieved by using filtering and different visual perspectives of the same scene—as done by the processing of photos obtained with multiple cameras in cutting-edge smart phones. Applied to the EMG signal for motor unit decomposition, a similar process is achieved using multiple comparisons of signals recorded by high-density electrode arrays (see text).

Figure 2

Normalized EMG/force. Mean (± SE) for each hand (left and right) for no vibration () and vibration () for females (N = 10, left panel), and males (N = 10, right panel ). *p < 0.05. Each variable (EMG, force) was normalized to its corresponding 100% MVC. The required force exertion was 20% MVC.