Estimating 3D kinematics and kinetics from virtual inertial sensor data through musculoskeletal movement simulations

• DOI: 10.3389/fbioe.2024.1285845
• Publication Date: 2024-04-02T05:29:33Z
• Authors: Marlies Nitschke; Eva Dorschky; Sigrid Leyendecker; Bjoern M. Eskofier; Anne D. Koelewijn
• Keywords: Inverse dynamics; Motion Capture; Ground reaction force

Portable measurement systems using inertial sensors enable motion capture outside the lab, facilitating longitudinal and large-scale studies in natural environments. However, estimating 3D kinematics kinetics from data for a comprehensive biomechanical movement analysis is still challenging. Machine learning models or stepwise approaches performing Kalman filtering, inverse kinematics, dynamics can lead to inconsistencies between kinetics. We investigated reconstruction of arbitrary running motions sensor optimal control simulations full-body musculoskeletal models. To evaluate feasibility proposed method, we used marker tracking created optical as reference computing virtual such that desired solution was known exactly. generated by formulating problems tracked acceleration angular velocity while minimizing effort without requiring task constraint an initial state. approach, reconstructed three trials each straight running, curved v-cut 10 participants. compared estimated signals variables simulations. The closely, resulting low mean root squared deviations pelvis translation (≤20.2 mm), angles (≤1.8 deg), ground reaction forces (≤1.1 BW%), joint moments (≤0.1 BWBH%), muscle (≤5.4 BW%) high coefficients multiple correlation all <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="m1"><mml:mrow><mml:mo stretchy="false">(</mml:mo><mml:mrow><mml:mo>≥</mml:mo><mml:mn>0.99</mml:mn></mml:mrow><mml:mo stretchy="false">)</mml:mo></mml:mrow></mml:math> . Accordingly, our results showed could reconstruct individual motions. led mutually dynamically consistent kinetics, which allows researching causal chains, example, analyze anterior cruciate ligament injury prevention. Our work proved approach data. When future with measured data, location alignment on segment must be estimated, soft-tissue artifacts are potential error sources. Nevertheless, demonstrated simulation highly promising analysis.