Correlation Analysis Based on Muscle sEMG Time-frequency Characteristics and Ultrasound-measured Muscle Mass Parameters.

Monitoring and quantification of skeletal muscle quality can assess human health status and physical function, and simple and portable detection tools are urgently needed in clinical practice. Surface electromyography (sEMG) is used to estimate muscle activation, atrophy, and function, but few studies have explored the relationship between different muscle sEMG features and muscle quality evaluation indicators. In this study, 16 subjects were recruited and sEMG data were collected during maximum voluntary contraction (MVC) of the rectus femoris (RF), vastus medialis (VM), vastus lateralis (VL), vastus intermedius (VI), biceps femoris (BF), and semitendinosus (ST). Then 14 representative features such as root mean square (RMS) and probability density function (PDF) were extracted. Ultrasound (US) examination was used to measure muscle thickness (MT) and cross-sectional area (CSA), and then Pearson correlation was used to analyze the association between sEMG features and US parameters (Params). The LASSO regularized model was used to predict MT and CSA. The result showed that most sEMG features demonstrated correlations ranging from moderate to strong (0.3 ≤ |r| ≤ 0.7) with muscular Params, including MVC, MT, and CSA. At the same time, the prediction results for the CSA of the quadriceps femoris (QF) and hamstrings (HS) were promising (R2_adj = 0.75, 0.87), and the regression result of HS MT was also good (R2_adj = 0.61). The results reveal the potential of sEMG technology in muscle health monitoring and functional assessment, and it is expected to become a new tool for wearable, low-cost muscle health monitoring and disease diagnosis in the future.