The relationship between EMG and muscle force changes with muscle fatigue, making interpretation of load sharing between muscles over time challenging. The purpose of this investigation was to evaluate the efficacy of normalizing EMG data to repeated, static, submaximal exertions to mitigate the fatigue artifact in EMG amplitude. Participants completed simulated repetitive work tasks, in 60-second work cycles, until exhaustion and surface EMG was recorded from 11 muscles. Every 12 minutes, participants completed a series of 4 submaximal reference exertions. Reference exertion EMG data were used in 6 normalizing methods including 1 standard (normalized to initial reference exertion) and 5 novel methods: (i) Fatigue Only, (ii) Linear Model, (iii) Cubic Model, (iv) Points Forward, and (v) Points Forward/Backward. EMG data were normalized to each novel methods and results were compared to the Standard Method. The significant differences between the novel methods and the Standard Method were dependent on the muscle and the number of time points in the analysis. Correlation analysis showed that the predicted cubic model points correlated better to the actual data points than the linear predicted values. This novel method to create “fatigue debiased” ratios may better reflect the changing muscular loads during repetitive work. This method was evaluated with a novel data set examining the effects of repetitive shoulder exertions, in multiple axes, on load sharing in the shoulder over time. The normalizing method was effective at distinguishing between the effects of fatigue artifact on EMG amplitude and load sharing between muscles over time.
Listed In: Biomechanics

Background: Numerous studies have described 3D kinematics, 3D kinetics and electromyography (EMG) of the lower limb during quasi-static or dynamic squatting activities. However there is only little information on the comparison of these two squatting conditions. Only one study compared these activities in terms of 3D kinematics, but no information was available on 3D kinetics and EMG. The purpose of this study was to compare simultaneous recordings of 3D kinematics, 3D kinetics and EMG of the lower limb during quasi-static and fast dynamic squats. Methods: Ten subjects were recruited. 3D knee kinematics was recorded with a motion capture system, 3D kinetics was recorded with a force plate, and EMG of 8 muscles was recorded with surface electrodes. Each subject performed a quasi-static squat and several fast dynamic squats from 0° to 70° of knee flexion. Findings: Mean differences between quasi-static and dynamic squats were 1.6° for rotations, 1.8 mm for translations, 38 N ground reaction forces (2.1 % of subjects’ body weight), 6 Nm for torques, 13.0 mm for center of pressure, and 7 µV for EMG (6.3% of the maximum dynamic electromyographic activities ). Some significant differences (P < 0.05) were found in anterior-posterior translation, vertical forces and EMG. Interpretation: All differences found between quasi-static and fast dynamic squats can be considered small. 69.5% of the compared data were equivalent. In conclusion, this study show for the first time that quasi-static and dynamic squatting activities are comparable in terms of 3D kinematics, 3D kinetics and EMG.

Listed In: Biomechanical Engineering, Biomechanics, Gait, Orthopedic Research, Posturography