The effects of load, flexion, twisting and window size on the stationarity of trunk muscle EMG signals

The aims of this study were to investigate the stationarity of trunk muscle electromyography (EMG) signals at various loads, flexion, twisting and sampling window sizes and to determine the post-processing conditions that improve the quality of frequency analysis in measuring muscle fatigue. Eight healthy individuals participated in the experiment, and the erector spinae muscle was used to collect the signals. All participants performed sustained isometric contractions at various combinations of three loads (0, 25 and 50% MVC), three flexion angles (0, 22.5 and 45°) and five twisting angles (−30, −15, 0, 15, and 30°). The signal collected for 20 s in each experimental task was analyzed using the five window sizes (250, 500, 750, 1000, and 2000 ms). The stationarity of the signal was examined using reverse and modified reverse arrangements tests. We determined the stationary level of the signal to be significantly affected by the window size. It was found that a window size of 750 ms was the optimal window size of the five considered in this study.Relevance to industryThis study focuses on suggesting an operative method to precisely measure the muscle fatigue in a lifting task. The results of this study will help researchers evaluate the risk of low back pain caused by local muscle fatigue measured by electromyography (EMG).

► We examined EMG signal stationarity to enhance the quality of frequency analysis.
► We tested the effect of load, flexion, twisting and window size on the stationarity.
► Signal stationarity was significantly affected by the window size and load.
► The 750-ms window size was the reasonable choice for stable signal.