The interaction between muscle pathophysiology, body mass, walking speed and ankle foot orthosis stiffness on walking energy cost: a predictive simulation study.

BACKGROUND: The stiffness of a dorsal leaf AFO that minimizes walking energy cost in people with plantarflexor weakness varies between individuals. Using predictive simulations, we studied the effects of plantarflexor weakness, passive plantarflexor stiffness, body mass, and walking speed on the optimal AFO stiffness for energy cost reduction. METHODS: We employed a planar, nine degrees-of-freedom musculoskeletal model, in which for validation maximal strength of the plantar flexors was reduced by 80%. Walking simulations, driven by minimizing a comprehensive cost function of which energy cost was the main contributor, were generated using a reflex-based controller. Simulations of walking without and with an AFO with stiffnesses between 0.9 and 8.7 Nm/degree were generated. After validation against experimental data of 11 people with plantarflexor weakness using the Root-mean-square error (RMSE), we systematically changed plantarflexor weakness (range 40-90% weakness), passive plantarflexor stiffness (range: 20-200% of normal), body mass (+ 30%) and walking speed (range: 0.8-1.2 m/s) in our baseline model to evaluate their effect on the optimal AFO stiffness for energy cost minimization. RESULTS: Our simulations had a RMSE < 2 for all lower limb joint kinetics and kinematics except the knee and hip power for walking without AFO. When systematically varying model parameters, more severe plantarflexor weakness, lower passive plantarflexor stiffness, higher body mass and walking speed increased the optimal AFO stiffness for energy cost minimization, with the largest effects for severity of plantarflexor weakness. CONCLUSIONS: Our forward simulations demonstrate that in individuals with bilateral plantarflexor the necessary AFO stiffness for walking energy cost minimization is largely affected by severity of plantarflexor weakness, while variation in walking speed, passive muscle stiffness and body mass influence the optimal stiffness to a lesser extent. That gait deviations without AFO are overestimated may have exaggerated the required support of the AFO to minimize walking energy cost. Future research should focus on improving predictive simulations in order to implement personalized predictions in usual care. Trial Registration Nederlands Trial Register 5170. Registration date: May 7th 2015.  http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=5170.

Interacting effects of AFO stiffness, neutral angle and footplate stiffness on gait in case of plantarflexor weakness: A predictive simulation study.

To maximize effects of dorsal leaf ankle foot orthoses (AFOs) on gait in people with bilateral plantarflexor weakness, the AFO properties should be matched to the individual. However, how AFO properties interact regarding their effect on gait function is unknown. We studied the interaction of AFO bending stiffness with neutral angle and footplate stiffness on the effect of bending stiffness on walking energy cost, gait kinematics and kinetics in people with plantarflexor weakness by employing predictive simulations. Our simulation framework consisted of a planar 11 degrees of freedom model, containing 11 muscles activated by a reflex-based neuromuscular controller. The controller was optimized by a comprehensive cost function, predominantly minimizing walking energy cost. The AFO bending and footplate stiffness were modelled as torsional springs around the ankle and metatarsal joint. The neutral angle of the AFO was defined as the angle in the sagittal plane at which the moment of the ankle torsional spring was zero. Simulations without AFO and with AFO for 9 bending stiffnesses (0-14 Nm/degree), 3 neutral angles (0-3-6 degrees dorsiflexion) and 3 footplate stiffnesses (0-0.5-2.0 Nm/degree) were performed. When changing neutral angle towards dorsiflexion, a higher AFO bending stiffness minimized energy cost of walking and normalized joint kinematics and kinetics. Footplate stiffness mainly affected MTP joint kinematics and kinetics, while no systematic and only marginal effects on energy cost were found. In conclusion, the interaction of the AFO bending stiffness and neutral angle in bilateral plantarflexor weakness, suggests that these should both be considered together when matching AFO properties to the individual patient.

Evaluating cost function criteria in predicting healthy gait.

Accurate predictive simulations of human gait rely on optimisation criteria to solve the system's redundancy. Defining such criteria is challenging, as the objectives driving the optimization of human gait are unclear. This study evaluated how minimising various physiologically-based criteria (i.e., cost of transport, muscle activity, head stability, foot-ground impact, and knee ligament use) affects the predicted gait, and developed and evaluated a combined, weighted cost function tuned to predict healthy gait. A generic planar musculoskeletal model with 18 Hill-type muscles was actuated using a reflex-based, parameterized controller. First, the criteria were applied into the base simulation framework separately. The gait pattern predicted by minimising each criterion was compared to experimental data of healthy gait using coefficients of determination (R2) and root mean square errors (RMSE) averaged over all biomechanical variables. Second, the optimal weighted combined cost function was created through stepwise addition of the criteria. Third, performance of the resulting combined cost function was evaluated by comparing the predicted gait to a simulation that was optimised solely to track experimental data. Optimising for each of the criteria separately showed their individual contribution to distinct aspects of gait (overall R2: 0.37-0.56; RMSE: 3.47-4.63 SD). An optimally weighted combined cost function provided improved overall agreement with experimental data (overall R2: 0.72; RMSE: 2.10 SD), and its performance was close to what is maximally achievable for the underlying simulation framework. This study showed how various optimisation criteria contribute to synthesising gait and that careful weighting of them is essential in predicting healthy gait.

Validation of forward simulations to predict the effects of bilateral plantarflexor weakness on gait.

BACKGROUND: Bilateral plantarflexor muscle weakness is a common impairment in many neuromuscular diseases. However, the way in which severity of plantarflexor weakness affects gait in terms of walking energy cost and speed is not fully understood. Predictive simulations are an attractive alternative to human experiments as simulations allow systematic alterations in muscle weakness. However, simulations of pathological gait have not yet been validated against experimental data, limiting their applicability. RESEARCH QUESTION: Our first aim was to validate a predictive simulation framework for walking with bilateral plantarflexor weakness by comparing predicted gait against experimental gait data of patients with bilateral plantarflexor weakness. Secondly, we aimed to evaluate how incremental levels of bilateral plantarflexor weakness affect gait. METHODS: We used a planar musculoskeletal model with 9 degrees of freedom and 9 Hill-type muscle-tendon units per leg. A state-dependent reflex-based controller optimized for a function combining energy cost, muscle activation squared and head acceleration was used to simulate gait. For validation, strength of the plantarflexors was reduced by 80 % and simulated gait compared with experimental data of 16 subjects with bilateral plantarflexor weakness. Subsequently, strength of the plantarflexors was reduced stepwise to evaluate its effect on gait kinematics and kinetics, walking energy cost and speed. RESULTS: Simulations with 80 % weakness matched well with experimental hip and ankle kinematics and kinetics (R > 0.64), but less for knee kinetics (R < 0.55). With incremental strength reduction, especially beyond a reduction of 60 %, the maximal ankle moment and power decreased. Walking energy cost and speed showed a strong quadratic relation (R2>0.82) with plantarflexor strength. SIGNIFICANCE: Our simulation framework predicted most gait changes due to bilateral plantarflexor weakness, and indicates that pathological gait features emerge especially when bilateral plantarflexor weakness exceeds 60 %. Our framework may support future research into the effect of pathologies or assistive devices on gait.

The Human Body Model versus conventional gait models for kinematic gait analysis in children with cerebral palsy.

With the rise of biofeedback in gait training in cerebral palsy there is a need for real-time measurements of gait kinematics. The Human Body Model (HBM) is a recently developed model, optimized for the real-time computing of kinematics. This study evaluated differences between HBM and two commonly used models for clinical gait analysis: the Newington Model, also known as Plug-in-Gait (PiG), and the calibrated anatomical system technique (CAST). Twenty-five children with cerebral palsy participated. 3D instrumented gait analyses were performed in three laboratories across Europe, using a comprehensive retroreflective marker set comprising three models: HBM, PiG and CAST. Gait kinematics from the three models were compared using statistical parametric mapping, and RMSE values were used to quantify differences. The minimal clinically significant difference was set at 5°. Sagittal plane differences were mostly less than 5°. For frontal and transverse planes, differences between all three models for almost all segment and joint angles exceeded the value of minimal clinical significance. Which model holds the most accurate information remains undecided since none of the three models represents a ground truth. Meanwhile, it can be concluded that all three models are equivalent in representing sagittal plane gait kinematics in clinical gait analysis.

The validity and usability of an eight marker model for avatar-based biofeedback gait training.

BACKGROUND: Virtual reality presents a platform for therapeutic gaming, and incorporation of immersive biofeedback on gait may enhance outcomes in rehabilitation. Time is limited in therapeutic practice, therefore any potential gait training tool requires a short set up time, while maintaining clinical relevance and accuracy. The aim of this study was to develop, validate, and establish the usability of an avatar-based application for biofeedback-enhanced gait training with minimal set up time. METHODS: A simplified, eight marker model was developed using eight passive markers placed on anatomical landmarks. This allowed for visualisation of avatar-based biofeedback on pelvis kinematics, hip and knee sagittal angles in real-time. Retrospective gait analysis data from typically developing children (n = 41) and children with cerebral palsy (n = 25), were used to validate eight marker model. Gait outcomes were compared to the Human Body Model using statistical parametric mapping. Usability for use in clinical practice was tested in five clinical rehabilitation centers with the system usability score. FINDINGS: Gait outcomes of Human Body Model and eight marker model were comparable, with small differences in gait parameters. The discrepancies between models were <5°, except for knee extension where eight marker model showed significantly less knee extension, especially towards full extension. The application was considered of 'high marginal acceptability' (system usability score, mean 68 (SD 13)). INTERPRETATION: Gait biofeedback can be achieved, to acceptable accuracy for within-session gait training, using an eight marker model. The application may be considered usable and implemented for use in patient populations undergoing gait training.

Precision of determining bone pose and marker position in the foot and lower leg from computed tomography scans: How low can we go in radiation dose?

Computed tomography (CT) imaging can be used to determine bone pose, sometimes combined with skin-mounted markers. For this specific application, a lower radiation dose than the conventional clinical dose might suffice. This study aims to determine how lowering the radiation dose of a CT-scan of the ankle and foot affects the precision of detecting bone pose and marker position. Radiation dose is proportional to tube charge. Hence, an adult cadaveric leg was scanned 10 times at four different tube charges (150, 75, 50 and 20 mAs) with a Philips Brilliance 64 CT scanner. Precision of detecting bone and marker position at 50 mAs was not significantly different from 75 mAs and from the clinically used 150 mAs, but higher than 20 mAs. Furthermore, no differences of the precision in detecting bone orientation were found. These results indicate that the radiation dose can be reduced by a factor 3 compared to the clinically usual radiation dose, without affecting the precision of detecting bone pose and marker position in the foot and ankle.

European consensus on the concepts and measurement of the pathophysiological neuromuscular responses to passive muscle stretch.

BACKGROUND AND PURPOSE: To support clinical decision-making in central neurological disorders, a physical examination is used to assess responses to passive muscle stretch. However, what exactly is being assessed is expressed and interpreted in different ways. A clear diagnostic framework is lacking. Therefore, the aim was to arrive at unambiguous terminology about the concepts and measurement around pathophysiological neuromuscular response to passive muscle stretch. METHODS: During two consensus meetings, 37 experts from 12 European countries filled online questionnaires based on a Delphi approach, followed by plenary discussion after rounds. Consensus was reached for agreement ≥75%. RESULTS: The term hyper-resistance should be used to describe the phenomenon of impaired neuromuscular response during passive stretch, instead of for example 'spasticity' or 'hypertonia'. From there, it is essential to distinguish non-neural (tissue-related) from neural (central nervous system related) contributions to hyper-resistance. Tissue contributions are elasticity, viscosity and muscle shortening. Neural contributions are velocity dependent stretch hyperreflexia and non-velocity dependent involuntary background activation. The term 'spasticity' should only be used next to stretch hyperreflexia, and 'stiffness' next to passive tissue contributions. When joint angle, moment and electromyography are recorded, components of hyper-resistance within the framework can be quantitatively assessed. CONCLUSIONS: A conceptual framework of pathophysiological responses to passive muscle stretch is defined. This framework can be used in clinical assessment of hyper-resistance and will improve communication between clinicians. Components within the framework are defined by objective parameters from instrumented assessment. These parameters need experimental validation in order to develop treatment algorithms based on the aetiology of the clinical phenomena.

Optimal calibration of instrumented treadmills using an instrumented pole.

Calibration of instrumented treadmills is imperative for accurate measurement of ground reaction forces and center of pressure (COP). A protocol using an instrumented pole has been shown to considerably increase force and COP accuracy. This study examined how this protocol can be further optimized to maximize accuracy, by varying the measurement time and number of spots, using nonlinear approaches to calculate the calibration matrix and by correcting for potential inhomogeneity in the distribution of COP errors across the treadmill's surface. The accuracy increased with addition of spots and correction for the inhomogeneous distribution across the belt surface, decreased with reduction of measurement time, and did not improve by including nonlinear terms. Most of these methods improved the overall accuracy only to a limited extent, suggesting that the maximal accuracy is approached given the treadmill's inherent mechanical limitations. However, both correction for position dependence of the accuracy as well as its optimization within the walking area are found to be valuable additions to the standard calibration process.

Sensitivity of subject-specific models to Hill muscle-tendon model parameters in simulations of gait.

Subject-specific musculoskeletal (MS) models of the lower extremity are essential for applications such as predicting the effects of orthopedic surgery. We performed an extensive sensitivity analysis to assess the effects of potential errors in Hill muscle-tendon (MT) model parameters for each of the 56 MT parts contained in a state-of-the-art MS model. We used two metrics, namely a Local Sensitivity Index (LSI) and an Overall Sensitivity Index (OSI), to distinguish the effect of the perturbation on the predicted force produced by the perturbed MT parts and by all the remaining MT parts, respectively, during a simulated gait cycle. Results indicated that sensitivity of the model depended on the specific role of each MT part during gait, and not merely on its size and length. Tendon slack length was the most sensitive parameter, followed by maximal isometric muscle force and optimal muscle fiber length, while nominal pennation angle showed very low sensitivity. The highest sensitivity values were found for the MT parts that act as prime movers of gait (Soleus: average OSI=5.27%, Rectus Femoris: average OSI=4.47%, Gastrocnemius: average OSI=3.77%, Vastus Lateralis: average OSI=1.36%, Biceps Femoris Caput Longum: average OSI=1.06%) and hip stabilizers (Gluteus Medius: average OSI=3.10%, Obturator Internus: average OSI=1.96%, Gluteus Minimus: average OSI=1.40%, Piriformis: average OSI=0.98%), followed by the Peroneal muscles (average OSI=2.20%) and Tibialis Anterior (average OSI=1.78%) some of which were not included in previous sensitivity studies. Finally, the proposed priority list provides quantitative information to indicate which MT parts and which MT parameters should be estimated most accurately to create detailed and reliable subject-specific MS models.

TLEM 2.0 - a comprehensive musculoskeletal geometry dataset for subject-specific modeling of lower extremity.

When analyzing complex biomechanical problems such as predicting the effects of orthopedic surgery, subject-specific musculoskeletal models are essential to achieve reliable predictions. The aim of this paper is to present the Twente Lower Extremity Model 2.0, a new comprehensive dataset of the musculoskeletal geometry of the lower extremity, which is based on medical imaging data and dissection performed on the right lower extremity of a fresh male cadaver. Bone, muscle and subcutaneous fat (including skin) volumes were segmented from computed tomography and magnetic resonance images scans. Inertial parameters were estimated from the image-based segmented volumes. A complete cadaver dissection was performed, in which bony landmarks, attachments sites and lines-of-action of 55 muscle actuators and 12 ligaments, bony wrapping surfaces, and joint geometry were measured. The obtained musculoskeletal geometry dataset was finally implemented in the AnyBody Modeling System (AnyBody Technology A/S, Aalborg, Denmark), resulting in a model consisting of 12 segments, 11 joints and 21 degrees of freedom, and including 166 muscle-tendon elements for each leg. The new TLEM 2.0 dataset was purposely built to be easily combined with novel image-based scaling techniques, such as bone surface morphing, muscle volume registration and muscle-tendon path identification, in order to obtain subject-specific musculoskeletal models in a quick and accurate way. The complete dataset, including CT and MRI scans and segmented volume and surfaces, is made available at http://www.utwente.nl/ctw/bw/research/projects/TLEMsafe for the biomechanical community, in order to accelerate the development and adoption of subject-specific models on large scale. TLEM 2.0 is freely shared for non-commercial use only, under acceptance of the TLEMsafe Research License Agreement.

Energy exchange between subject and belt during treadmill walking.

Treadmill walking aims to simulate overground walking, but intra-stride belt speed variations of treadmills result in some interaction between treadmill and subject, possibly obstructing this aim. Especially in self-paced treadmill walking, in which the belt speed constantly adjusts to the subject, these interactions might affect the gait pattern significantly. The aim of this study was to quantify the energy exchange between subject and treadmill, during the fixed speed (FS) and self-paced (SP) modes of treadmill walking. Eighteen subjects walked on a dual-belt instrumented treadmill at both modes. The energy exchange was calculated as the integration of the product of the belt speed deviation and the fore-aft ground reaction force over the stride cycle. The total positive energy exchange was 0.44 J/stride and the negative exchange was 0.11 J/stride, which was both less than 1.6% of the performed work on the center of mass. Energy was mainly exchanged from subject to treadmill during both the braking and propulsive phase of gait. The two treadmill modes showed a similar pattern of energy exchange, with a slightly increased energy exchange during the braking phase of SP walking. It is concluded that treadmill walking is only mildly disturbed by subject-belt interactions when using instrumented treadmills with adequate belt control.

Effects of adding a virtual reality environment to different modes of treadmill walking.

Differences in gait between overground and treadmill walking are suggested to result from imposed treadmill speed and lack of visual flow. To counteract this effect, feedback-controlled treadmills that allow the subject to control the belt speed along with an immersive virtual reality (VR) have recently been developed. We studied the effect of adding a VR during both fixed speed (FS) and self-paced (SP) treadmill walking. Nineteen subjects walked on a dual-belt instrumented treadmill with a simple endless road projected on a 180° circular screen. A main effect of VR was found for hip flexion offset, peak hip extension, peak knee extension moment, knee flexion moment gain and ankle power during push off. A consistent interaction effect between VR and treadmill mode was found for 12 out of 30 parameters, although the differences were small and did not exceed 50% of the within subject stride variance. At FS, the VR seemed to slightly improve the walking pattern towards overground walking, with for example a 6.5mm increase in stride length. At SP, gait became slightly more cautious by adding a VR, with a 9.1mm decrease in stride length. Irrespective of treadmill mode, subjects rated walking with the VR as more similar to overground walking. In the context of clinical gait analysis, the effects of VR are too small to be relevant and are outweighed by the gains of adding a VR, such as a more stimulating experience and possibility of augmenting it by real-time feedback.

Evaluation of a morphing based method to estimate muscle attachment sites of the lower extremity.

To generate subject-specific musculoskeletal models for clinical use, the location of muscle attachment sites needs to be estimated with accurate, fast and preferably automated tools. For this purpose, an automatic method was used to estimate the muscle attachment sites of the lower extremity, based on the assumption of a relation between the bone geometry and the location of muscle attachment sites. The aim of this study was to evaluate the accuracy of this morphing based method. Two cadaver dissections were performed to measure the contours of 72 muscle attachment sites on the pelvis, femur, tibia and calcaneus. The geometry of the bones including the muscle attachment sites was morphed from one cadaver to the other and vice versa. For 69% of the muscle attachment sites, the mean distance between the measured and morphed muscle attachment sites was smaller than 15 mm. Furthermore, the muscle attachment sites that had relatively large distances had shown low sensitivity to these deviations. Therefore, this morphing based method is a promising tool for estimating subject-specific muscle attachment sites in the lower extremity in a fast and automated manner.

Self-paced versus fixed speed treadmill walking.

Instrumented treadmills are increasingly used in gait research, although the imposed walking speed is suggested to affect gait performance. A feedback-controlled treadmill that allows subjects to walk at their preferred speed, i.e. functioning in a self-paced (SP) mode, might be an attractive alternative, but could disturb gait through accelerations of the belt. We compared SP with fixed speed (FS) treadmill walking, and also considered various feedback modes. Nineteen healthy subjects walked on a dual-belt instrumented treadmill. Spatio-temporal, kinematic and kinetic gait parameters were derived from both the average stride patterns and stride-to-stride variability. For 15 out of 70 parameters significant differences were found between SP and FS. These differences were smaller than 1cm, 1°, 0.2 Nm and 0.2 W/kg for respectively stride length and width, joint kinematics, moments and powers. Since this is well within the normal stride variability, these differences were not considered to be clinically relevant, indicating that SP walking is not notably affected by belt accelerations. The long-term components of walking speed variability increased during SP walking (43%, p<0.01), suggesting that SP allows for more natural stride variability. Differences between SP feedback modes were predominantly found in the timescales of walking speed variability, while the gait pattern was similar between modes. Overall, the lack of clinically significant differences in gait pattern suggests that SP walking is a suitable alternative to fixed speed treadmill walking in gait analysis.

A simple controller for the prediction of three-dimensional gait.

The objective of this study is to investigate the potential of forward dynamic modeling in predicting the functional outcome of complicated orthopedic procedures involving relocation or removal of muscles or correction osteotomies in the lower extremities. For this purpose, we developed a torque actuated forward dynamics based three-dimensional model of gait, that extends the previous reported work of Van der Kooij et al. (2003). The mechanical properties are scaled to the subject and lateral stability is provided by an 'offset plus proportional' controller (Hof, 2008). Kinematic constraints are formulated based on three independent gait descriptors and implemented in an optimization algorithm. The computational effort is small (1min per gait cycle on a 1GHz processor) and the control scheme generates symmetric and cyclic gait based on the desired gait descriptors. An interface with the inverse dynamics based AnyBody Modeling System, a musculoskeletal modeling tool, provides insight in muscle activities. The proposed control scheme is robust against mediolateral perturbations. The predictive capacity of the model is evaluated by simulating pathological gait by means of weakening the hip abductors, and the model is able to predict some of the trends of compensatory strategies in such a perturbed mechanical system.

Sensitivity of subject-specific models to errors in musculo-skeletal geometry.

Subject-specific musculo-skeletal models of the lower extremity are an important tool for investigating various biomechanical problems, for instance the results of surgery such as joint replacements and tendon transfers. The aim of this study was to assess the potential effects of errors in musculo-skeletal geometry on subject-specific model results. We performed an extensive sensitivity analysis to quantify the effect of the perturbation of origin, insertion and via points of each of the 56 musculo-tendon parts contained in the model. We used two metrics, namely a Local Sensitivity Index (LSI) and an Overall Sensitivity Index (OSI), to distinguish the effect of the perturbation on the predicted force produced by only the perturbed musculo-tendon parts and by all the remaining musculo-tendon parts, respectively, during a simulated gait cycle. Results indicated that, for each musculo-tendon part, only two points show a significant sensitivity: its origin, or pseudo-origin, point and its insertion, or pseudo-insertion, point. The most sensitive points belong to those musculo-tendon parts that act as prime movers in the walking movement (insertion point of the Achilles Tendon: LSI=15.56%, OSI=7.17%; origin points of the Rectus Femoris: LSI=13.89%, OSI=2.44%) and as hip stabilizers (insertion points of the Gluteus Medius Anterior: LSI=17.92%, OSI=2.79%; insertion point of the Gluteus Minimus: LSI=21.71%, OSI=2.41%). The proposed priority list provides quantitative information to improve the predictive accuracy of subject-specific musculo-skeletal models.

The effect of ankle foot orthosis stiffness on the energy cost of walking: a simulation study.

BACKGROUND: In stroke and multiple sclerosis patients, gait is frequently hampered by a reduced ability to push-off with the ankle caused by weakness of the plantar-flexor muscles. To enhance ankle push-off and to decrease the high energy cost of walking, spring-like carbon-composite Ankle Foot Orthoses are frequently prescribed. However, it is unknown what Ankle Foot Orthoses stiffness should be used to obtain the most efficient gait. The aim of this simulation study was to gain insights into the effect of variation in Ankle Foot Orthosis stiffness on the amount of energy stored in the Ankle Foot Orthosis and the energy cost of walking. METHODS: We developed a two-dimensional forward-dynamic walking model with a passive spring at the ankle representing the Ankle Foot Orthosis and two constant torques at the hip for propulsion. We varied Ankle Foot Orthosis stiffness while keeping speed and step length constant. FINDINGS: We found an optimal stiffness, at which the energy delivered at the hip joint was minimal. Energy cost decreased with increasing energy storage in the ankle foot orthosis, but the most efficient gait did not occur with maximal energy storage. With maximum storage, push-off occurred too late to reduce the impact of the contralateral leg with the floor. Maximum return prior to foot strike was also suboptimal, as push-off occurred too early and its effects were subsequently counteracted by gravity. The optimal Ankle Foot Orthosis stiffness resulted in significant push-off timed just prior to foot strike and led to greater ankle plantar-flexion velocity just before contralateral foot strike. INTERPRETATION: Our results suggest that patient energy cost might be reduced by the proper choice of Ankle Foot Orthosis stiffness.