Search

Clinical motion analysis involves quantitative measurement of gait patterns to identify gait anomalies that currently or have the potential to impact function, activities of daily living and participation. Clinical motion analysis services are equipped with motion capture technology and comprise specialised staff who deliver 3-dimensional motion analysis services to children and adults who present with varying levels of gait impairment. Data is then used to inform intervention recommendations to clinicians with a view to maintaining independent, functional and pain free walking (or appropriate mobility). The ANZ-CMAG (established in 2013) identified a need to establish recommendations to assist in standardising practice guidelines for both current and new clinical motion analysis services within the region. The group serves to promote collaboration between services in quality assurance processes, clinical practices, data sets and research activities. The clinical practice recommendations described in this paper cover: i) requirements for a motion analysis service (including staffing, facilities and equipment), ii) patient assessments (requirements, clinical information and data gathered, reporting and interpretation of patient data), iii) quality assurance processes (including motion capture system / biomechanical models & limitations, marker placement, data storage / record keeping, creation of normative dataset); iv) helpful resources. Better outcomes for children and adults with gait deviations is dependent upon accurate measurement and evaluation of walking and requires input from multidisciplinary clinical teams with specialist knowledge and skills. The ANZ-CMAG hopes these clinical practice recommendations are beneficial to motion analysis services with an aim to improve clinical practices, patient outcomes, and support research collaboration., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier B.V. All rights reserved.)

Background: The Gait Assessment and Intervention Tool (GAIT) provides a valid and reliable gait quality measure for adults with stroke, however a minimal clinically important difference (MCID) is yet to be determined., Research Question: What is the GAIT MCID in community dwelling adults with a stroke diagnosis?, Methods: The GAIT scores for a consecutive series of 63 adults with stroke, referred for video gait assessment within an outpatient rehabilitation program, were retrospectively identified from a gait laboratory database. Patients were classified by Functional Ambulation Category (FAC) and had walking speed measured, which classified them into one of three walking speed-based ambulatory levels (I.e. household, limited community or community ambulator). Linear regression models were fitted to assess the association between GAIT score and FAC level and GAIT score and ambulatory level. MCIDs were determined based on estimates calculated in the two models RESULTS: The FAC of patients ranged from 3 to 5. GAIT score was negatively correlated with FAC level and ambulatory level (ρ = -0.73, p < 0.001 and ρ =-0.69, p < 0.001, respectively). Pairwise comparisons from the models showed absolute mean differences between estimated GAIT scores of 10.84 (95% confidence interval 7.59-14.09) for changes from FAC level 3-4 and 12.13 (8.90-15.36) for household to limited community ambulator, and 5.90 (3.44-8.37) for changes from FAC level 4-5 and 4.39 (2.01-6.76) for limited community to community ambulator. The proposed MCID for FAC level 3 or household ambulators is 11.48, and for FAC level 4 and 5 or limited community/community ambulators is 5.19., Significance: The proposed MCIDs represent real changes in gait quality measured by the GAIT for adults with stroke who exhibit lower or higher functional mobility levels. The MCIDs will assist clinicians and researchers using the tool to determine if meaningful change in gait quality has taken place for adults with stroke undergoing rehabilitation., (Crown Copyright © 2021. Published by Elsevier B.V. All rights reserved.)

Objectives: The objectives of this study were to evaluate whether botulinum toxin type A (BoNT-A) treatment for lower limb spasticity leads to patient goal attainment and identify factors associated with positive goal attainment and to assess the effect of BoNT-A treatment on patients' gait., Design: Retrospective cohort study between June 2014 and February 2019., Setting: Public outpatient spasticity clinic in a tertiary hospital., Participants: Thirty patients (N=30; 50% female; average age, 50.5y) with lower limb spasticity of heterogenous etiologies (96.7% cerebral±spinal origin and 3.3% isolated spinal origin); 73.3% (N=22) of patients had previously received BoNT-A treatment., Interventions: BoNT-A injection to lower limb muscles., Main Outcome Measures: The primary outcome measure was goal attainment measured using Goal Attainment Scaling. The Modified Ashworth Scale (MAS) was used to assess spasticity. Gait was characterized by spatiotemporal parameters., Results: Fifty-six treatment episodes were analyzed and showed that BoNT-A treatment resulted in a significant reduction in spasticity (pretreatment MAS=3.18±0.73; posttreatment MAS=2.27±0.89; P <.001) with no associated change in gait parameters. Logistic regression revealed that most patients (74.1%) achieved all of their goals, with younger patients having a high likelihood of goal attainment regardless of their gait profile identified by latent profile analysis of the gait parameters. Patients considered to have a low functioning gait profile demonstrated a significantly greater likelihood of goal attainment than patients with the other gait profiles combined (odds ratio, 45.6; 95% confidence interval, 1.3-1602.1; P =.036). Chronic spasticity and pretreatment severity of spasticity (MAS) and its reduction were not associated with likelihood of goal attainment., Conclusions: The success and efficacy of BoNT-A treatment in improving patient perceived gait quality and reducing the negative symptoms of spasticity were best measured using Goal Attainment Scaling. The study emphasizes the importance of measuring patient goals as a clinical outcome. Gait parameters were most informative when used collectively to classify patients based on their overall gait profile, which assisted in identifying differences between patients' likelihood of goal attainment after treatment., (© 2021 The Authors.)

Purpose: To assess the orthotic and therapeutic effects of prolonged use of functional electrical stimulation (FES) on fatigue induced gait patterns in people with Multiple Sclerosis (MS)., Method: Thirteen people with MS completed 3D gait analysis with FES off and on, before and after a fatiguing 6-minute walk, at baseline and after 8 weeks of use of FES., Results: Eleven participants completed all testing. An orthotic effect on gait was not evident on first use of FES. However, therapeutic effects on gait after 8 weeks use were generally positive, including increases in walking speed due to improved neuromuscular control and power generated at the hip and ankle of the more affected limb. The action of FES alone was not sufficient to overcome all fatigue related deficits in gait but there was evidence 8 weeks use of FES can ameliorate some fatigue effects on lower limb kinetics, including benefits to ankle mechanics involved in generating power around push-off during stance., Conclusions: Eight-weeks of FES can benefit the gait pattern of people with MS under non-fatigued and fatigued conditions. Implications for rehabilitation In some people with MS prolonged use of FES may be necessary before observing positive orthotic effects. Improvements in the neuromuscular control of the more affected lower limb may develop with prolonged use of FES in people with MS. Only some therapeutic benefits of FES are maintained during fatigued walking in people with MS. FES may be considered as a gait retraining device as well as an orthotic intervention for people with MS.

Objective: To determine whether normalizing spatial-temporal gait data for walking speed obtained from multiple walking trials leads to differences in gait variability parameters associated with a history of falling in people with transtibial amputations., Design: Cross-sectional study., Setting: Rehabilitation center., Participants: People with unilateral transtibial amputations (N=45; mean age ± SD, 60.5±13.7y; 35 men [78%]) were recruited., Interventions: Not applicable., Main Outcome Measures: Participants completed 10 consecutive walking trials using an instrumented walkway system. Primary gait parameters were walking speed and step-length, step-width, step-time, and swing-time variability. A retrospective 12-month fall history was obtained from participants., Results: Sixteen amputees (36%) were classified as fallers. Variation in gait speed across the 10 walking trials was 2.9% (range, 1.1%-12.1%). Variability parameters of normalized gait data were significantly different from variability parameters of nonnormalized data (all P<.01). For nonnormalized data, fallers had greater amputated limb step-time (P=.02), step-length (P=.02), swing-time (P=.05), and step-width (P=.03) variability and nonamputated limb step-length (P=.04) and step-width (P=.01) variability. For normalized data, only 3 variability parameters were significantly greater for fallers. These were amputated limb step-time (P=.05), step-length (P=.02), and step-width (P=.01) variability., Conclusions: Normalizing spatial-temporal gait data for walking speed before calculating gait variability parameters may aid in discerning variability parameters related to falls histories in people with transtibial amputations. This may help focus on the initial rehabilitation efforts of amputees with a fall history., (Copyright © 2015 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.)

Reorganization of primary motor cortex (M1) is well-described in long-term lower limb amputees. In contrast cortical reorganization during the rehabilitation period after amputation is poorly understood. Thirteen transtibial amputees and 13 gender matched control participants of similar age were recruited. Transcranial magnetic stimulation was used to assess corticomotor and intracortical excitability of M1 bilaterally. Neurophysiological assessments were conducted at admission, prosthetic casting, first walk and discharge. Gait variability at discharge was assessed as a functional measure. Compared to controls, amputees had reduced short-latency intracortical inhibition (SICI) for the ipsilateral M1 at admission (p = 0.01). Analysis across rehabilitation revealed SICI was reduced for the contralateral M1 at first walk compared to discharge (p = 0.003). For the ipsilateral M1 both short and long-latency intracortical inhibition were reduced at admission (p < 0.05) and prosthetic casting (p < 0.02). Analysis of the neurophysiology and gait function revealed several interesting relationships. For the contralateral M1, reduced inhibition at admission (p = 0.04) and first walk (p = 0.05) was associated with better gait function. For the ipsilateral M1, reduced inhibition at discharge (p = 0.05) was associated with poor gait function. This study characterized intracortical excitability in rehabilitating amputees. A dichotomous relationship between reduced intracortical inhibition for each M1 and gait function was observed at different times. Intracortical inhibition may be an appropriate cortical biomarker of gait function in lower limb amputees during rehabilitation, but requires further investigation. Understanding M1 intracortical excitability of amputees undertaking prosthetic rehabilitation provides insight into brain reorganization in the sub-acute post-amputation period and may guide future studies seeking to improve rehabilitation outcomes.

Ipsilateral primary motor cortex (M1) reorganisation after unilateral lower-limb amputation may degrade function of the amputated limb. We hypothesised unilateral lower-limb amputees would have a bilateral increase in corticomotor excitability, and increased excitability of ipsilateral M1 would be associated with increased step-time variability during gait. Twenty transtibial amputees (16 male) aged 60.1 years (range 45-80 years), and 20 age- and gender-matched healthy adult controls were recruited. Single-pulse transcranial magnetic stimulation assessed corticomotor excitability. Two indices of corticomotor excitability were calculated. An index of corticospinal excitability (ICE) determined relative excitability of ipsilateral and contralateral corticomotor projections to alpha-motoneurons innervating the quadriceps muscle (QM) of the amputated limb. A laterality index (LI) assessed relative excitability of contralateral projections from each hemisphere. Spatial-temporal gait analysis was performed to calculate step-time variability. Amputees had lower ICE values, indicating relatively greater excitability of ipsilateral corticomotor projections than controls (P = 0.04). A lower ICE value was associated with increased step-time variability for amputated (P = 0.04) and non-amputated limbs (P = 0.02). This association suggests corticomotor projections from ipsilateral M1 to alpha-motoneurons innervating the amputated limb QM may interfere with gait. Cortical excitability in amputees was not increased bilaterally, contrary to our hypothesis. There was no difference in excitability of contralateral M1 between amputees and controls (P = 0.10), and no difference in LI (P = 0.71). It appears both hemispheres control one QM, with predominance of contralateral corticomotor excitability in healthy adults. Following lower-limb amputation, putative ipsilateral corticomotor excitability is relatively increased in some amputees and may negatively impact on function., (© 2014 Federation of European Neuroscience Societies and John Wiley & Sons Ltd.)

Objectives: To determine whether a knee brace incorporating inflatable air bladders can alter the net peak external knee adduction moment in persons with medial compartment knee osteoarthritis., Design: Prospective cohort study., Setting: Motion analysis laboratory., Participants: Subjects (n = 18) diagnosed with knee osteoarthritis as defined by the Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association., Methods: Instrumented gait analysis was performed while subjects walked with and without the knee brace. When subjects wore the knee brace, the air bladders were either uninflated or inflated to 7 psi. The net external knee adduction moment was obtained by subtracting the abduction moment produced by the knee brace (estimated using a finite element analysis model) from the external knee adduction moment (estimated using a camera-based motion analysis system)., Main Outcome Measurements: The net external knee adduction moment was compared across all testing conditions., Results: A 7.6% decrease in net peak external knee adduction moment was observed when subjects wore the knee brace uninflated compared with when they did not wear the brace. Inflation of the bladders to 7 psi led to a 26.0% decrease in net peak external knee adduction moment., Conclusions: The results of the study suggest that the effects of an unloading knee brace may be enhanced by incorporating inflatable air bladders into the design of the brace, thus leading to an improved correction of the excessive peak external knee adduction moment observed in patients with medial compartment knee osteoarthritis., (Copyright © 2013. Published by Elsevier Inc.)

Background and Purpose: Individuals with amputations are a core group in Australian rehabilitation units that have a long index length of stay. The Repatriation General Hospital (RGH) offers general rehabilitation services to the population of Southern Adelaide (a population of 350,000) and includes an on-site prosthetic manufacturing facility. Using a physiotherapy database at the RGH, we sought to answer the following questions: What are the demographic and clinical characteristics of patients admitted for lower limb prosthetic rehabilitation over 15 years? What are the times to rehabilitation outcomes? How have these changed over 15 years with changes in service delivery?, Methods: This paper is a retrospective observational study using a physiotherapy clinical database (1996-2010) of 531 consecutive individuals with lower limb amputation at one South Australian hospital (RGH). There were two changes in service delivery: 1) a multidisciplinary interim prosthetic programme (IPP) introduced in 1998 and 2) removable rigid dressings (RRDs) introduced in 2000. Outcome measures were patient demographics, clinical characteristics and time to rehabilitation outcome markers., Results: Mean age was 68 years (standard deviation [SD]: 15), with 69% male, 80% dysvascular and 68% transtibial. The overall median inpatient rehabilitation length of stay (RLOS) was 39 days (interquartile range [IQR]: 26-57). Individuals with amputation entering rehabilitation each year had a higher number of co-morbidities (β: 0.08; 95% confidence interval: 0.05-0.11). Introduction of the IPP was associated with a significant reduction in time to initial prosthetic casting, independent walking and inpatient RLOS. Introduction of RRDs was associated with a significant reduction in time to wound healing, initial prosthetic casting and independent walking., Conclusions: Individuals with amputation were typically elderly dysvascular men with transtibial amputations. Introduction of the IPP and RRDs successfully reduced time to rehabilitation outcomes including independent walking, an outcome that is rarely reported but is of significance to patients and physiotherapists., (Copyright © 2012 John Wiley & Sons, Ltd.)

The purpose of this study was to evaluate the extent to which residual shank length affects long jump performance of elite athletes with a unilateral transtibial amputation. Sixteen elite, male, long jumpers with a transtibial amputation were videoed while competing in major championships (World Championships 1998, 2002 and Paralympic Games, 2004). The approach, take-off, and landing of each athlete's best jump was digitized to determine residual and intact shank lengths, jump distance, and horizontal and vertical velocity of center of mass at touchdown. Residual shank length ranged from 15 cm to 38 cm. There were weak, nonsignificant relationships between residual shank length and (a) distance jumped (r = 0.30), (b) horizontal velocity (r = 0.31), and vertical velocity (r = 0.05). Based on these results, residual shank length is not an important determinant of long jump performance, and it is therefore appropriate that all long jumpers with transtibial amputation compete in the same class. The relationship between residual shank length and key performance variables was stronger among athletes that jumped off their prosthetic leg (N = 5), and although this result must be interpreted cautiously, it indicates the need for further research.

Background: Prefabricated orthotic devices are currently designed to fit a range of patients and therefore they do not provide individualized comfort and function. Custom-fit orthoses are superior to prefabricated orthotic devices from both of the above-mentioned standpoints. However, creating a custom-fit orthosis is a laborious and time-intensive manual process performed by skilled orthotists. Besides, adjustments made to both prefabricated and custom-fit orthoses are carried out in a qualitative manner. So both comfort and function can potentially suffer considerably. A computerized technique for fabricating patient-specific orthotic devices has the potential to provide excellent comfort and allow for changes in the standard design to meet the specific needs of each patient., Methods: In this paper, 3D laser scanning is combined with rapid prototyping to create patient-specific orthoses. A novel process was engineered to utilize patient-specific surface data of the patient anatomy as a digital input, manipulate the surface data to an optimal form using Computer Aided Design (CAD) software, and then download the digital output from the CAD software to a rapid prototyping machine for fabrication., Results: Two AFOs were rapidly prototyped to demonstrate the proposed process. Gait analysis data of a subject wearing the AFOs indicated that the rapid prototyped AFOs performed comparably to the prefabricated polypropylene design., Conclusions: The rapidly prototyped orthoses fabricated in this study provided good fit of the subject's anatomy compared to a prefabricated AFO while delivering comparable function (i.e. mechanical effect on the biomechanics of gait). The rapid fabrication capability is of interest because it has potential for decreasing fabrication time and cost especially when a replacement of the orthosis is required.

Passive-elastic foot prostheses cannot produce net work. Consequently, passive-elastic foot prostheses are limited in their ability to enable a biologically-realistic gait pattern in transtibial amputees. This shortcoming results in difficulties in balance and walking and leads to high levels of oxygen consumption during locomotion. A powered prosthesis has the potential for overcoming these problems and allowing transtibial amputees to achieve a biologically-realistic gait pattern. In this study, we compared the effects of the Ceterus by Össur, a traditional passive-elastic prosthesis, with those of the PowerFoot Biom (iWalk, Cambridge, MA), a recently-developed powered prosthesis. Gait biomechanics and metabolic cost were compared in a group of 5 transtibial amputees during level-ground walking. The results provided preliminary evidence that the use of a powered prosthesis leads to a decrease in the level of oxygen consumption during ambulation due to improvements in ankle kinematics and kinetics primarily during late stance. An average decrease in oxygen consumption of 8.4% was observed during the study when subjects used the PowerFoot compared to the Ceterus. An average increase of 54% was observed in the peak ankle power generation during late stance. Our results suggest that powered prostheses have the potential for significantly improving ambulation in transtibial amputees.

Objective: To evaluate gait biomechanics after training with a virtual reality (VR) system and to elucidate underlying mechanisms that contributed to the observed functional improvement in gait speed and distance., Design: A single blind randomized control study., Setting: Gait analysis laboratory in a rehabilitation hospital and the community., Participants: Fifteen men and three women with hemiparesis caused by stroke., Interventions: Subjects trained on a six-degree of freedom force-feedback robot interfaced with a VR simulation. Subjects were randomized to either a VR group (n=9) or non-VR group (NVR, n=9). Training was performed three times a week for 4 weeks for approximately 1h each visit., Main Outcome Measures: Kinematic and kinetic gait parameters., Results: Subjects in the VR group demonstrated a significantly larger increase in ankle power generation at push-off as a result of training (p=0.036). The VR group had greater change in ankle ROM post-training (19.5%) as compared to the NVR group (3.3%). Significant differences were found in knee ROM on the affected side during stance and swing, with greater change in the VR group. No significant changes were observed in kinematics or kinetics of the hip post-training., Conclusions: These findings are encouraging because they support the potential for recovery of force and power of the lower extremity for individuals with chronic hemiparesis. It is likely that the effects of training included improved motor control at the ankle, which enabled the cascade of changes that produced the functional improvements seen after training., (Copyright 2010 Elsevier B.V. All rights reserved.)

Improved treatments and rehabilitation for cerebral palsy (CP) have led to an increased number of children with CP surviving into adulthood. Adults with CP show an increased prevalence of pain, fatigue, and musculoskeletal dysfunction, leading to a decrease in ambulatory function. Recent work has demonstrated the potential benefits of intensive task-specific gait training, including the use of robotic-driven gait orthoses, on motor recovery in children with CP. In contrast, reports of interventions aimed at improving motor function in adults with CP are lacking. This case study reports on the outcomes of a 6-week intervention of robotic-assisted gait training administered to a 52-year-old woman with right hemiplegia attributable to CP. Improvements were noted in balance, walking speed, and time to negotiate stairs at posttraining and follow-up. Gait analysis showed an increase in step length and a reduction in the period of double support. In conclusion, robotic-assisted gait training may be beneficial in enhancing locomotor function in adults with CP.

The clinical management of children with cerebral palsy (CP) relies on monitoring changes in the severity of gait abnormalities and on planning appropriate clinical interventions. Currently available technology does not make it possible to perform clinical gait evaluations as often as it would be desirable from a clinical standpoint. The use of wearable technology (e.g. a sensorized shoe) could provide an effective means to monitor changes in the severity of gait abnormalities in children with CP. In this paper, we studied a group of children with CP who showed an equinus (i.e. toe-walking) gait pattern, a gait abnormality often observed in children with CP. The aim of the study was to determine the feasibility of relying upon a sensorized shoe to assess changes in the severity of toe-walking. We demonstrated that it is possible to use features extracted from the center of pressure trajectory and ankle kinematics to predict the severity of toe-walking. Our results motivate the development and clinical testing of a sensorized shoe to assess changes in gait patterns that accompany the development, and the response to clinical interventions, of children with CP.

An analysis of the take-off technique in transtibial amputee high jump was performed on two athletes (both using intact limb take-off) competing in the high jump finals of the 2004 Paralympic Games. Two digital video cameras were used to film the event with the data later digitized and reconstructed using standard 3D direct linear transformation (DLT) procedures. Some similarities with non-amputee high jump technique were noted in that centre of mass height was low at touch-down (TD), there was a similar magnitude of negative vertical velocity at TD, and most of the vertical velocity generated occurred in the first half of the take-off phase. However, both transtibial amputee athletes exhibited a slower horizontal approach velocity, a lower positive vertical take-off velocity, a more upright position at touch-down and a greater range of motion of the hip throughout the take-off phase compared to what is known about non-amputee high jump technique. These differences may be associated with constraints of taking off from the prosthetic limb on the previous step, resulting in having to adopt a different posture at touch-down compared to non-amputees. Understanding transtibial amputee high jump technique and the differences compared to what is known about non-amputee technique has implications for coaching and improving performance in prosthetic sport.

Purpose: The purpose of this study was to investigate whether female lower-limb amputees conform to the established long-jump model and to compare the kinematics of the approach and take-off phases for elite female transfemoral and transtibial amputee long jumpers., Methods: Eight female transfemoral and nine female transtibial amputee athletes were videotaped (sagittal plane movements at 50 Hz) from third-to-last step to take-off during the 2004 Paralympic Games long-jump finals. After digitizing and reconstruction of 2D coordinates, key variables were calculated at each stride and during contact with the take-off board. Additionally, approach speed during the run-up of each jump was recorded (100 Hz) using a laser Doppler device (LDM 300 C Sport, Jenoptik Laser, Jena, Germany)., Results: The transfemoral amputees had a consistently higher center of mass height on the last three steps before take-off than the transtibial amputees. However, at touch-down onto the take-off board, they lowered their center of mass excessively so that from touch-down to take-off, they were actually lower than the transtibial amputees. This resulted in a greater negative vertical velocity at touch-down and may have inversely affected their jump performance., Conclusion: Female transtibial athletes conformed to the long-jump model, although adaptations to this technique were displayed. Female transfemoral athletes, however, exhibited no relationship between take-off speed and distance jumped, which may be attributable to their excessive lowering of their center-of-mass height at touch-down onto the take-off board. It is recommended that coaches and athletes proceed with caution when trying to replicate techniques used by able-bodied athletes because adaptations to the constraints of a prosthesis should be considered.

Functional recovery of an impaired gait pattern is a common goal for stroke patients in their rehabilitation. Robotic and mechatronic devices offer a means of facilitating and enhancing gait retraining practices undertaken by clinicians. A new active knee orthosis has been developed for gait retraining of stroke patients that may fulfil this role. Since this device is newly developed, it is important to determine its impact on the walking patterns of healthy individuals before exploring its use in gait retraining of stroke patients. The aim of this study was to analyze adaptations in gait mechanics of healthy subjects due to the added mass of the knee orthosis while worn uni-laterally and bi-laterally. In our preliminary tests we observed significant deviations from normal gait patterns when the knee orthosis was worn uni-laterally. Conversely, minor gait deviations were seen when the knee orthosis was worn bi-laterally. This suggests that a bilateral configuration may be more suited for gait retraining purposes.

Spasticity in stroke patients interferes with coordinated muscle firing patterns of the lower extremity leading to gait abnormalities. The goal of this study was to improve ankle function during walking by augmenting treadmill gait retraining with a visual EMG biofeedback technique. Eight stroke patients who could ambulate between 0.5 and 0.9 m/s participated in the study. The training consisted of 12 sessions of treadmill walking during which the activity of the tibialis anterior and gastrocnemius lateralis muscles of the affected side was displayed on a computer screen. Targets were shown to indicate to the subject when to activate the monitored muscles. Gait evaluations were performed before and after the training period to test the hypothesis that ankle mechanics improved following the intervention. Improvements in gait function were characterized by changes in temporal gait parameters and lower extremity kinematics and kinetics. Subjects showed an increase in gait speed, time of single leg support on the affected side, ankle power generation at push-off and a reduction in knee extensor moment. These results indicate that treadmill gait retraining augmented via visual EMG-biofeedback facilitates improvements in hemiparetic gait.