A Female Population-Averaged Musculoskeletal Model Outperforms Conventional Male-Based Generic Models in Simulating Female Gait

A new study demonstrates that a female-specific musculoskeletal model outperforms traditional male-based models in simulating walking biomechanics. Researchers developed a population-averaged female model, constructed using MRI data from 25 adult women. By employing advanced techniques like thin-plate-spline muscle-path mapping and bilateral symmetrisation, they built a model that accurately represents the average female anatomy.

The researchers then compared this female model against a male-based generic model, scaling both to match a new individual's MRI-based model. Using 5-fold cross-validation, they measured the models' performance across various gait-related metrics, including joint kinematics, moments, muscle forces, and reaction forces.

The results showed that the female model consistently outperformed the male model in every output category. The differences were most significant for pelvis tilt, hip flexion, and gluteal/adductor moment arms, as well as forces. While some discrepancies remained in early-stance knee kinematics and patellofemoral loading, the overall performance gap was clear.

This study, available on SimTK, suggests that researchers should consider using sex-specific models when investigating female participants, particularly those focused on pelvic and hip biomechanics. The population-averaged female model offers a more accurate representation of female walking mechanics compared to outdated male-based models.