Author(s): XU, Y., GU, Y., Institution: NINGBO UNIVERSITY, Country: CHINA, Abstract-ID: 412

INTRODUCTION:
Several biomechanical models for skeletal muscles have seen extensive use. Nonetheless, the model employed in sports biomechanics relied on a non-dynamic equation and was only applicable under quasi-static conditions. This research seeks to develop a skeletal muscle model based on the semi-phenomenological model (SPM) of a sarcomere and employ it in dynamic modeling utilizing the results of kinetic calculations in OpenSim. Subsequently, it aims to validate the precision and dependability of the SPM by comparing joint moment calculations based on the SPM in OpenSim and gravitational resis
METHODS:
In this study, the biceps femoris long head was selected as the subject. Kinetic data, sEMG signals, and external joint moments were recorded, collected, computed, and analyzed. The One-dimensional Statistical Parameter Mapping (SPM1D) algorithm was utilized to compare the net knee bending moment and mechanical output calculated in the SPM and OpenSim models through a paired t-test at a significance level of 0.001. The calculated actual knee bending moment served as the reference standard. Correlation analysis was conducted to compare the average joint moments and average mechanical work output calculated by OpenSim tools with the mass of body segments and the sarcomere SPM at each time step.
RESULTS:
The SPM significantly decreased the error in joint rotation moment but did not show a significant reduction in the error in the calculation of mechanical work output at each time step. The SPM demonstrated an advantage in calculating instantaneous power, with a negligible difference compared to real instantaneous power output, and improved the accuracy in computing total mechanical work output by reducing the error rate to 43.1%.

CONCLUSION:
The results showed that the SPM-based biomechanical algorithm significantly reduced errors in joint rotation moment calculations compared to the traditional Hill-type model in OpenSim, particularly in dynamic contraction conditions. The SPM also exhibited a better transient response, recovering and declining muscle force more quickly. These advantages can be attributed to the SPMs modeling principles, which assume equal time constants for muscle activation and inactivation, resulting in faster muscle force predictions. However, when it comes to the calculation of mechanical work output, both the Hill-type model in OpenSim and the SPM showed no statistically significant difference in error. This aligns with the principles of both models and the studys experimental design. It is worth noting that the complexity of muscle contraction energy output is influenced by factors like action potential frequency, movement type, and muscle activation timing. While the SPM has shown effectiveness in previous experiments, caution should be exercised when applying these results, as the validation data came from insect muscles. In conclusion, the skeletal muscle modeling based on the SPM of sarcomere appears to be a better approach for biomechanical research.