Gait2Hip-60: A Unified Deep Learning Benchmark for Predicting Hip Muscle Forces and Joint Moments from Multi-Cadence Gait Kinematics
Researchers have developed a deep learning benchmark, Gait2Hip-60, to predict hip muscle forces and joint moments from gait kinematics. The study compared LSTM, Transformer, and Mamba models, finding that the Transformer model achieved the best performance in predicting these parameters from healthy adults. While the Transformer model showed moderate predictive ability in a small cohort of patients with osteonecrosis of the femoral head, further validation is needed for clinical application. AI
IMPACT This research introduces a novel deep learning approach for biomechanical analysis, potentially improving clinical diagnostics and rehabilitation strategies.