Neuromechanical synergy patterns explain metabolic efficiency differences during the sit-to-walk transition
摘要
The sit-to-walk (STW) transition is a key component of the Timed Up and Go test, yet the neuromuscular and cross-modal coordination features associated with metabolic cost during STW remain incompletely understood. This study developed a simulation-informed synergy analysis framework combining OpenSim–Umberger metabolic estimation, non-negative matrix factorization (NMF), and mixed-matrix factorization (MMF). Seventy-one healthy participants performed STW trials with kinematics, ground reaction forces, and eight-channel surface electromyography recorded. Model-derived metabolic cost was estimated using participant-specific musculoskeletal simulations, and an STW oxygen-consumption agreement analysis in a validation subsample supported the metabolic estimation workflow. Participants were stratified into residual-based metabolic-cost groups after adjustment for body mass, height, and sex. Across groups, joint kinematics and simulation-derived muscle-level outputs were broadly comparable. NMF identified localized differences in temporal activation coefficients after correction for the number of synergies, with group effects confined to early and late time windows rather than extending across the full analyzed cycle. Exploratory NMF weight comparisons further suggested potential differences in selected muscle weights. MMF characterized non-negative temporal coefficients and cross-modal EMG–biomechanical weight patterns; corrected analyses of MMF temporal coefficients identified limited, localized differences in selected MMF synergies, and exploratory weight comparisons indicated localized differences in selected EMG and biomechanical channel–synergy combinations. These findings suggest that residual-based metabolic-cost phenotypes during STW are associated with localized synergy-related features under broadly similar external movement patterns. The proposed framework provides a healthy-reference and hypothesis-generating approach for examining subtle internal coordination features linked to model-estimated metabolic cost.