Tibiofemoral Joint Contact Force Estimation Based on OpenSim Musculoskeletal Modeling
ObjectiveTo develop a full-body musculoskeletal model based on OpenSim, which has a knee joint structure with double contact points and three degrees of freedom, and to estimate the knee joint load, so as to provide a reliable basis for precise diagnosis of the patients with knee osteoarthritis and...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Editorial Office of Rehabilitation Medicine
2025-02-01
|
| Series: | 康复学报 |
| Subjects: | |
| Online Access: | http://kfxb.publish.founderss.cn/thesisDetails#10.3724/SP.J.1329.2025.01015 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | ObjectiveTo develop a full-body musculoskeletal model based on OpenSim, which has a knee joint structure with double contact points and three degrees of freedom, and to estimate the knee joint load, so as to provide a reliable basis for precise diagnosis of the patients with knee osteoarthritis and formulation of customized rehabilitation plan in clinic.MethodsAuthorized public dataset from Stanford University, published by the Shiley Center for Orthopaedic Research & Education at Scripps Clinic in La Jolla, California was used, and four subjects who had undergone total knee arthroplasty and implanted the pressure testing instruments were included. Based on the generic model gait2392 in the OpenSim model library, a new model was built, which contained 28 joints and 43 degrees of freedom, adjusted the knee joint structure to have 3 directional degrees of freedom in flexion-extension, adduction-abduction, and internal-external rotation, added an upper limb structure with 4 joints and 7 degrees of freedom in a single arm capable of torque actuation, and adjusted the contact point of the tibiofemoral joint from a single to a double contact point.Through the model scaling, inverse kinematics, inverse dynamics analysis, residual reduction processing and muscle control and analysis, the medial and lateral tibiofemoral contact forces were estimated during walking gait, comprehensively considering the contributions of external forces and internal muscle strength. Matlab 2018b was used for statistical analysis. The validity of the musculoskeletal model established in this study was verified by calculating the Pearson's correlation coefficient, root mean square error, and standard deviation between the model estimation results and the measurement results of the implanted prosthesis.ResultsA total of 21 complete gait trials data were obtained from the 4 subjects. The average gait cycle durations for subjects JW, DM, SC, and PS were (1.18±0.03), (1.18±0.08), (1.09±0.02), and (1.14±0.04) s. The average correlation coefficients between the estimated contact force and the measured values of medial, lateral, and total contact forces were (0.921±0.079), (0.817±0.084) and (0.930±0.066), and the average root mean square errors were (0.336±0.146), (0.332±0.043), and (0.442±0.160) BW. The average root mean square errors of the estimated and measured peak contact forces of the medial and lateral sides were (0.43±0.25) and (0.34±0.24) BW, and the errors at the time of peak contact force onset were (44.09±34.66) and (67.52±61.19) ms.ConclusionThe established full-body OpenSim musculoskeletal model of the knee joint with two contact points and three degrees of freedom can be used to generate muscle-driven gait simulations and obtain reliable contact forces of the tibiofemoral joint, which provides an important basis for accurate monitoring of tibiofemoral joint load, formulation of clinical rehabilitation programs and continuous improvement of prosthesis design. |
|---|---|
| ISSN: | 2096-0328 |