Fig 1.
Conceptual drawings of (a) the pre-limb loss model and (b) the post-limb loss model, with the right ankle muscles removed and replaced with a transtibial prosthesis.
Fig 2.
Prosthetic ankle stiffness vs. metabolic cost.
Metabolic cost of walking at 1.3 m/s using the post-limb loss model with no strength loss and a passive prosthesis, with different prosthetic ankle stiffnesses. Horizontal line is the metabolic cost of the pre-limb loss simulation at this same speed.
Fig 3.
Metabolic cost and stride duration for simulations of walking at five different speeds (circles) using (a) the pre-limb loss model and (b) the post-limb loss model. The squares and error bars are means and standard deviations for human experimental data [10] with (a) Service Members with two intact limbs and (b) Service Members with unilateral transtibial limb loss walking with a passive transtibial prosthesis.
Fig 4.
Gait kinematics (hip, knee, and ankle angles) and kinetics (horizontal and vertical ground reaction forces, HGRF and VGRF) of Subject 01’s walking simulations with the pre-limb loss model (columns 1 and 4), and with the post-limb loss model using the passive prosthesis (columns 2 and 5) and the powered prosthesis (columns 3 and 6). The post-limb loss model had no strength loss. The stride begins at heel-strike of the right/residual limb. The shaded area in each panel is ± two standard deviations around the mean of experimental data for N = 14 healthy adults without limb loss [33].
Fig 5.
Example muscle forces and excitations.
Muscle forces (solid lines, % max isometric force) and excitations (dashed lines) of Subject 01’s walking simulations with the pre-limb loss model (column 1) and with the post-limb loss model using the passive prosthesis (columns 2 and 3). The post-limb loss model had no strength loss. The stride begins at heel-strike of the right/residual limb. The thick bars at the top of each panel are typical “on/off” timing derived from experimental electromyograms [43,44].
Fig 6.
Box plots of tracking errors for different prosthesis types, passive prosthetic ankle stiffness, and level of strength loss. Panel (a) is simulations of bilateral strength loss. Panel (b) is simulations of unilateral strength loss, on the side of the residual limb. The symbol within each box is the mean.
Fig 7.
Changes in gross metabolic cost.
Box plots of changes in metabolic cost of walking at 1.3 m/s (relative to the pre-limb loss simulation) for different prosthesis types, passive prosthetic ankle stiffness, and level of strength loss. Panel (a) is simulations of bilateral strength loss. Panel (b) is simulations of unilateral strength loss, on the side of the residual limb. The symbol within each box is the mean. * indicates the change was significantly greater than zero (p < 0.001, d > 1.0, B > 100).
Fig 8.
Changes in muscle metabolic costs.
Changes in the metabolic cost of individual muscles in the model (mean±SD) between the pre- and post-limb loss simulations when using the passive prosthesis with generic stiffness and 10% strength loss.