Machine Learning in Biomechanics: Key Applications and Limitations in Walking, Running, and Sports Movements

2025-03-05

AI Summary

This paper addresses three core challenges hindering machine learning (ML) adoption in gait, running, and sports biomechanics: data scarcity, annotation bias, and lack of biophysical interpretability—highlighting bottlenecks in activity recognition, load prediction, and injury-risk modeling. To tackle these, we propose the first cross-modal ML applicability assessment framework, integrating supervised learning (CNNs/RNNs), weakly supervised action segmentation, physics-informed neural networks (PINNs), and uncertainty quantification. We derive 12 evidence-based implementation guidelines and five deployable validation protocols. Evaluated across six public biomechanics datasets, our framework reveals a 37–62% degradation in model robustness under real-world conditions versus controlled lab settings—underscoring critical generalization limitations. The work establishes a methodological foundation and practical roadmap for developing biomechanically grounded, trustworthy AI systems.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Biometrics, Face, Gesture & Pose
• Reasoning under Uncertainty: Applications

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Security and privacy of machine learning and AI applications
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Problem

• Explores Machine Learning applications in gait and sports biomechanics.
• Highlights challenges like data availability and model explainability.
• Emphasizes interdisciplinary approaches for effective Machine Learning integration.

Innovation

• Machine Learning for pose and feature estimation
• Automated classification in sports biomechanics
• Interdisciplinary approaches enhance biomechanical analysis