Login / Signup

A subject-specific musculoskeletal model to predict the tibiofemoral contact forces during daily living activities.

Li ZhangGeng LiuYuzhou YanBing HanHui LiJianbing MaXupeng Wang
Published in: Computer methods in biomechanics and biomedical engineering (2022)
Accurate prediction of tibiofemoral contact force (TFCF) during daily living activities is significant for understanding the initiation, progression, and treatment of knee osteoarthritis (KOA). However, the diversity of target activities, prediction accuracy, and computational efficiency of the current musculoskeletal simulations need to be further improved. In this study, a subject-specific musculoskeletal model considered the tibiofemoral alignment, medial-lateral contact locations, secondary tibiofemoral and all patellofemoral motions, and knee ligaments was proposed to predict the TFCFs during the five daily activities (normal walking, sit-to-stand, stand-to-sit, stair ascent, and stair descent) in OpenSim software. The standing lower-limbs-full-length radiograph, local radiograph of knee joint, motion capture data, and force plate data of eighteen subjects were acquired as the input data of the musculoskeletal model. The results showed good agreements of TFCFs between the predictions based on our proposed musculoskeletal model and the in-vivo measurements based on instrumented knee implants during the five daily activities (RMSE: 0.16 ∼ 0.31 BW, R 2 : 0.88 ∼ 0.97, M: -0.11 ∼ -0.02, P: 0.03 ∼ 0.10, and C: 0.04 ∼ 0.14). Additionally, the order of the peak total and lateral TFCFs from low to high was normal walking, stair ascent and stand-to-sit, and stair descent and sit-to-stand ( P  < 0.05), and the peak medial TFCF was stand-to-sit, sit-to-stand, normal walking, stair ascent and stair descent ( P  < 0.05). The outcomes of this study are valuable for further understanding the knee biomechanics during daily living activities and providing theoretical guidance for the treatments of KOA.
Keyphrases
  • knee osteoarthritis
  • total knee arthroplasty
  • physical activity
  • electronic health record
  • big data
  • single molecule
  • type diabetes
  • anterior cruciate ligament
  • machine learning
  • skeletal muscle
  • combination therapy