Musculoskeletal multibody models, such as those built in OpenSim, provide estimates of muscle, ground reaction and joint reaction forces during motor tasks, but represent each anatomical segment as a rigid body and therefore cannot resolve the stress and strain distribution within it. Finite element models capture that internal response, but require physiologically consistent boundary conditions that cannot be derived independently. Connecting the two descriptions is a recognised objective in computational biomechanics, and in the foot it has been pursued mainly for the plantar soft tissue; work directed at the interior of a single bone is less common, and it requires a quantity that is rarely transferred, namely the reactions at the joints through which that bone exchanges load with the ones adjacent to it. The aim of this work is to develop and verify a computational pipeline that couples a musculoskeletal simulation in OpenSim with a patient-specific finite element model of the calcaneus, built directly from computed tomography through the Cartesian grid finite element method (cgFEM). Muscle forces, the ground reaction with its measured free moment, and the subtalar and Chopart joint reactions are extracted at selected instants of the stance phase of gait, reconciled with the imaging domain through a landmark-initialised rigid registration, and applied to the finite element model, which returns the stress field within the bone. The stiffness of the articular interfaces, which cannot be measured in a subject, is obtained by solving an inverse problem in which the discrepancy between the computed reactions and the musculoskeletal joint reactions is minimised. The framework was applied to three instants of the stance phase of a walking trial, and a verification protocol was implemented alongside the pipeline to check the internal consistency of the transfer at each stage. The pipeline is automated from load extraction to the finite element solution and defined by a single configuration file, so that its application to a different subject, bone or motor task requires new inputs rather than new development.
Musculoskeletal multibody models, such as those built in OpenSim, provide estimates of muscle, ground reaction and joint reaction forces during motor tasks, but represent each anatomical segment as a rigid body and therefore cannot resolve the stress and strain distribution within it. Finite element models capture that internal response, but require physiologically consistent boundary conditions that cannot be derived independently. Connecting the two descriptions is a recognised objective in computational biomechanics, and in the foot it has been pursued mainly for the plantar soft tissue; work directed at the interior of a single bone is less common, and it requires a quantity that is rarely transferred, namely the reactions at the joints through which that bone exchanges load with the ones adjacent to it. The aim of this work is to develop and verify a computational pipeline that couples a musculoskeletal simulation in OpenSim with a patient-specific finite element model of the calcaneus, built directly from computed tomography through the Cartesian grid finite element method (cgFEM). Muscle forces, the ground reaction with its measured free moment, and the subtalar and Chopart joint reactions are extracted at selected instants of the stance phase of gait, reconciled with the imaging domain through a landmark-initialised rigid registration, and applied to the finite element model, which returns the stress field within the bone. The stiffness of the articular interfaces, which cannot be measured in a subject, is obtained by solving an inverse problem in which the discrepancy between the computed reactions and the musculoskeletal joint reactions is minimised. The framework was applied to three instants of the stance phase of a walking trial, and a verification protocol was implemented alongside the pipeline to check the internal consistency of the transfer at each stage. The pipeline is automated from load extraction to the finite element solution and defined by a single configuration file, so that its application to a different subject, bone or motor task requires new inputs rather than new development.
Bridging Multibody Dynamics and Finite Element Analysis: A Patient-Specific Coupling Framework for Musculoskeletal Foot Biomechanics
BRESCIANI, MARCO
2025/2026
Abstract
Musculoskeletal multibody models, such as those built in OpenSim, provide estimates of muscle, ground reaction and joint reaction forces during motor tasks, but represent each anatomical segment as a rigid body and therefore cannot resolve the stress and strain distribution within it. Finite element models capture that internal response, but require physiologically consistent boundary conditions that cannot be derived independently. Connecting the two descriptions is a recognised objective in computational biomechanics, and in the foot it has been pursued mainly for the plantar soft tissue; work directed at the interior of a single bone is less common, and it requires a quantity that is rarely transferred, namely the reactions at the joints through which that bone exchanges load with the ones adjacent to it. The aim of this work is to develop and verify a computational pipeline that couples a musculoskeletal simulation in OpenSim with a patient-specific finite element model of the calcaneus, built directly from computed tomography through the Cartesian grid finite element method (cgFEM). Muscle forces, the ground reaction with its measured free moment, and the subtalar and Chopart joint reactions are extracted at selected instants of the stance phase of gait, reconciled with the imaging domain through a landmark-initialised rigid registration, and applied to the finite element model, which returns the stress field within the bone. The stiffness of the articular interfaces, which cannot be measured in a subject, is obtained by solving an inverse problem in which the discrepancy between the computed reactions and the musculoskeletal joint reactions is minimised. The framework was applied to three instants of the stance phase of a walking trial, and a verification protocol was implemented alongside the pipeline to check the internal consistency of the transfer at each stage. The pipeline is automated from load extraction to the finite element solution and defined by a single configuration file, so that its application to a different subject, bone or motor task requires new inputs rather than new development.| File | Dimensione | Formato | |
|---|---|---|---|
|
Bresciani_Marco.pdf
accesso aperto
Dimensione
6.39 MB
Formato
Adobe PDF
|
6.39 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/116001