Validation of lower limb muscle activation estimated using musculoskeletal modeling against electromyography in the table tennis topspin forehand and backhand

This study aimed to validate the lower limb muscle activation, estimated using static optimization against electromyography (EMG), in the topspin forehand and backhand strokes. The secondary purpose was to compare the estimated activations of the major muscles/muscle groups between the forehand and backhand strokes. Eight male college table tennis players hit the cross-court topspin forehands and backhands with maximum effort. Stroke motions and ground reaction forces were measured using a motion capture system and two force plates. The EMG signals of the 16 lower-limb muscles were recorded using a wireless EMG system. The static optimization algorithm of OpenSim was applied to stroke motions to estimate lower limb muscle activation, which was compared to EMG activation. Of the seven muscles that showed maximum activation > 0.3 during the forehand, five showed a Pearson correlation coefficient > 0.3 Of the four muscles that showed maximum activation > 0.3 during the backhand, all four showed a Pearson correlation coefficient >0.3. However, some muscles, such as the bilateral gluteus medius muscles, showed a low correlation between estimated and EMG activation. A possible cause is the co-contraction of the relevant muscles. Concordance correlation coefficients were smaller than their respective Pearson correlation coefficients. This result reflects that EMG envelope (activation) is also an estimate of muscle activation and is subject to noise and confounding factors. Comparisons with additional independent measurements, such as ultrasound muscle images and instrumented joint loading, are necessary for more robust validation of the musculoskeletal modeling and muscle activation. The gluteus maximus and hamstrings on the playing side, and rectus femoris on the non-playing side exhibited higher activation during the forehand than during the backhand. The overall results suggest that the static optimization algorithm can adequately estimate lower-limb muscle activity during the topspin forehand and backhand strokes.