Fig 1.
The proposed lower-extremity musculoskeletal model of people with an osseointegrated unilateral transfemoral amputation (of the right leg but without loss of generality) wearing a generic transfemoral prosthesis during level ground walking in the OpenSim simulation environment.
Table 1.
The participants’ characteristics.
Fig 2.
(A) Generic OpenSim musculoskeletal model of a healthy subject with 23 DOF and 92 musculotendon units (gait2392_simbody.osim). (B) The generic OpenSim musculoskeletal model of people with an unilateral transfemoral amputation wearing a generic prosthesis with 19 DOF, 76 musculotendon units, and two actuators, as proposed in this study. The muscles in yellow are transected and anchored to the transected femur bone, while the muscles in grey are removed in the proposed model.
Table 2.
Joint description of the OpenSim healthy subject model Gait2392.
Table 3.
Joint description of the proposed OpenSim transfemoral amputee model.
Fig 3.
Model topology of osseointegrated transfemoral amputee model in OpenSim.
The subtlar_l and mtp_l are marked in red to represent the locked DOF.
Table 4.
The mass of the body segments, center of mass (x, y, z), and their inertial properties (Ixx, Iyy, Izz) of the generic unscaled osseointegrated transfemoral model.
Fig 4.
The workflow to perform the forward dynamics simulation in OpenSim of the proposed.
The unscaled model is shown on the left. The cyan boxes are the toolbox available in OpenSim. The blue boxes are the input and settings files. The orange boxes are the generated outputs. The grey box contains the ground reaction forces.
Fig 5.
The Opensim CMC is modified to perform the forward dynamic simulation of the proposed model.
The control architecture relies on an optimal control in which the cost function accounts for the muscles excitations and the two actuators at the knee and ankle joints of the prosthesis.
Fig 6.
The residual forces generated by the (reserve) actuators at some joints of the subject-specific models of the two osseointegrated participants (OI-TFA-1 and OI-TFA-2).
Fig 7.
Angles in the sagittal plane for the ankle, knee, and hip joints during level ground walking as a result of the inverse kinematic simulation.
Red, blue, black color represents the participant OI-TFA1, OI-TFA-2, and AB-1, respectively. The continuous thick lines show the mean joint angles of the amputated leg for the participants with transfemoral amputation, and of the right leg for the healthy participant. The dashed thick lines show the contralateral limb. The shaded region represents the standard deviation of the joint angles over five gait cycles.
Fig 8.
(left) The relationship between the normalized ankle joint moment and ankle joint angle and (prosthetic limb for participants OI-TFA1 and OI-TFA2, the right leg for AB-1) as the results of the inverse dynamic simulation. (right) The relationship between the normalized knee joint moment and knee joint angle (the prosthetic limb for participants OI-TFA1 and OI-TFA2, and right leg for AB-1) as the results of the inverse dynamic simulation.
Fig 9.
Angles in the sagittal plane for the ankle, knee, and hip joints during level ground walking as the result of the modified computed muscle control simulation for one gait cycle for the contralateral leg of OI-TFA-1 (red) and OI-TFA-2 (blue), and for the right leg of AB-1 (black).
Fig 10.
Muscle forces for the hip flexion and extension of the amputated leg during a gait cycle.
For the AB-1, the results from the right leg are shown.
Fig 11.
Total power produced by hip flexors and extensors of the amputated leg during a gait cycle.
For the AB-1, the results from the right leg are shown.
Fig 12.
Muscle forces for the ankle plantarflexion and dorsiflexion of the contralateral limb during a gait cycle.
For the AB-1, the results from the right leg are shown.
Fig 13.
Total power produced by the ankle plantarflexors and dorsiflexors of the contralateral limb during a gait cycle.
For the AB-1, the results from the right leg are shown.