Muscle Activation Patterns Differentiate Post-stroke and Healthy Population

Upper limb impairments can give rise to alterations or functional limitations that may result from strokes. These limitations have conducted to the development of novel tools to aid in various stages of the rehabilitation process. One such tool is the assessment of the patient's level of impairment. Traditionally, the assessment necessitates the continuous involvement of a specialized therapist, which leads to substantial consumption of time and medical resources. To address this, some works have proposed the automatic evaluation of the patient’s affectation level by using electromyography, by extracting parameters like the muscle activation level or the co-contraction index, others use the muscle activity patterns to look at movement compensation. Nonetheless, none of those works attempt to use muscle activation patterns and the corresponding muscle compensation to automatically assess the impairments. Hence, this work proposes a method for automated detection of muscle activations in electromyography and utilizes them in statistical analysis to compare healthy and post-stroke subjects. It highlights similarities for both groups in muscles with higher activation during tasks, particularly in muscles such as the deltoid and extensor carpi, as well as differences in overall muscle activation percentages. Additionally, significant differences between activation patterns were observed.