A comparison of task and muscle specific isometric submaximal electromyography data normalization techniques

Abstract

The submaximal, constrained nature of joystick manipulation makes it difficult to select an appropriate technique for upper limb electromyography (EMG) normalization. The purpose of this study was to determine an appropriate submaximal isometric normalization method to quantify EMG from shoulder muscle activation in hydraulic-actuation joystick operators that could later be implemented in field settings. Surface EMG data were collected from the upper trapezius, posterior deltoid, and anterior deltoid of seventeen subjects while operating a hydraulic-actuation joystick. EMG data were normalized using two techniques: muscle specific (mRVC) and task specific (three joystick positions: start-tRVC_Start, middle-tRVC_Middle and end-tRVC_End). No significant differences (p ≪ 0.05) were observed for intersubject coefficient of variation (CV) between normalization procedures (mRVC, tRVC_Start, tRVC_Middle tRVC_End, un-normalized). These equivocal findings do not favour the use of any one of the submaximal normalization procedures over another. However, though not statistically significant, the un-normalized (0.68 ± 0.15) CVs were lower than those of normalized ensembles (0.96 ± 0.24) suggesting that for constrained, submaximal tasks, it may not be necessary to normalize EMG. Although this analysis was applied to upper limb EMG during joystick manipulation, the results have potential application to other submaximal upper limb tasks which are constrained and repetitive in nature thus including many assembly line jobs.