Master of Science
This thesis introduces an optimization framework for the scaling and registering of musculoskeletal models. Leveraging evolutionary algorithms, the framework navigates a highly dimensional and non-convex parameter space, exclusively using subject-specific motion-capture marker data. Monte Carlo simulations tested the framework's sensitivity to initial marker locations, demonstrating its robustness. The framework achieved an average RMS error of 0.77 cm in normal gait trials, representing a 44% improvement in RMS error over previous optimization methods (1.38 cm) and a 70% improvement over manual scaling (2.58 cm). This approach significantly reduces the need for expert intervention, enhances computational efficiency, and increases both accuracy and reproducibility for scaling musculoskeletal models.
Summary for Lay Audience
Musculoskeletal models are valuable tools for understanding human motion, especially in the realm of gait analysis. These computational models of the human skeletal and muscular system have become pivotal tools for gait analysis, surgical diagnostics, and rehabilitation. However, their effectiveness is often hampered by the laborious, manual process of scaling these generic models to fit individual subjects—a process that is not only time-consuming but also suffers from poor reproducibility. This thesis introduces a novel optimization framework that addresses the difficulties of musculoskeletal scaling. Previous optimization methods often use simplified models and synthetic data, limiting their real-world applicability. In contrast, our framework is designed to reliably scale a generic model to any subject using only motion-capture marker data. Using the subject's markers not only simplifies the process but allows us to analyze individuals with various gait pathologies and types of movements. The framework employs a unique two-level optimization approach and allows for a more systematic and objective scaling and registration process. By separating the optimization into two layers, we simplify what is traditionally a highly dimensional and complex problem. The result is a robust method that significantly enhances the accuracy and reproducibility of scaling models. One of the critical features of our framework is its computational efficiency, making it more realistic for in-depth gait analysis using more of the subject's data. This is particularly beneficial in clinical settings where quick, yet accurate, decision-making is often required. Moreover, the framework's modular structure allows for future users to adapt our approach to different types of movements and populations. In summary, this framework extends the field of biomechanical modeling by offering a robust, automated, and objective framework for scaling and registering musculoskeletal models. By significantly reducing the need for expert intervention and increasing both accuracy and reproducibility, this work presents promise for future development of high-level tools in musculoskeletal modeling and gait analysis.
Redgrift, Adam Garry, "Differential Evolution: A Global Optimization Framework For Marker Registration and Scaling Musculoskeletal Models" (2023). Electronic Thesis and Dissertation Repository. 9910.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Available for download on Monday, November 17, 2025