Inertial motion capture in conjunction with an artificial neural network can differentiate the gait patterns of hemiparetic stroke patients compared with able-bodied counterparts

Clinical gait analysis has proven to reduce uncertainties in selecting the appropriate quantity and type of treatment for patients with neuromuscular disorders. However, gait analysis as a clinical tool is under-utilised due to the limitations and cost of acquiring and managing data. To overcome these obstacles, inertial motion capture (IMC) recently emerged to counter the limitations attributed to other methods. This paper investigates the use of IMC for training and testing a back-propagation artificial neural network (ANN) for the purpose of distinguishing between hemiparetic stroke and able-bodied ambulation. Routine gait analysis was performed on 30 able-bodied control subjects and 28 hemiparetic stroke patients using an IMC system. An ANN was optimised to classify the two groups, achieving a repeatable network accuracy of 99.4%. It is concluded that an IMC system and appropriate computer methods may be useful for the planning and monitoring of gait rehabilitation therapy of stroke victims.     

Keep on your toes: gait initiation from toe-standing

Gait initiation from toe-standing is common in patients with upper motor neurone (UMN) pathology as well as in able-bodied subjects during certain dance and athletic situations. It is unclear whether balance problems in patients who toe-walk are due to the underlying pathology, or due to initiating gait from toe-standing. The aim of this study was to compare the biomechanics of gait initiation from toe-standing to that from heel-toe standing in healthy able-bodied subjects. Data were collected for three seconds prior to, and three seconds after, a visual signal to initiate gait. Ground reaction force and centre of pressure (COP) data were collected with an AMTI force platform, and electromyographic and kinematic data were collected from each limb with a Vicon motion analysis system. When initiating gait from toe-standing, there was a smaller backward displacement of the COP compared to heel-toe standing. In addition, greater forward momentum was generated, and there was an increase in gastrocnemius, rectus femoris and biceps femoris muscle activity. There were no differences in COP displacement or momentum generated in the mediolateral direction for the two conditions. Thus, initiating gait from toe-standing allows one to generate greater amounts of forward momentum but not at the expense of generating excessive stance-side momentum. This may be an advantageous method of initiating movement for dancers and athletes in certain situations. This work also suggests that balance problems in patients with UMN pathology are likely due to the underlying pathology and are not due to initiating gait from toe-standing.     

Artificial neural network model for the generation of muscle activation patterns for human locomotion.

Skilled locomotor behaviour requires information from various levels within the central nervous system (CNS). Mathematical models have permitted researchers to simulate various mechanisms in order to understand the organization of the locomotor control system. While it is difficult to adequately characterize the numerous inputs to the locomotor control system, an alternative strategy may be to use a kinematic movement plan to represent the complex inputs to the locomotor control system based on the possibility that the CNS may plan movements at a kinematic level. We propose the use of artificial neural network (ANN) models to represent the transformation of a kinematic plan into the necessary motor patterns. Essentially, kinematic representation of the actual limb movement was used as the input to an ANN model which generated the EMG activity of 8 muscles of the lower limb and trunk. Data from a wide variety of gait conditions was necessary to develop a robust model that could accommodate various environmental conditions encountered during everyday activity. A total of 120 walking strides representing normal walking and ten conditions where the normal gait was modified in terms of cadence, stride length, stance width or required foot clearance. The final network was assessed on its ability to predict the EMG activity on individual walking trials as well as its ability to represent the general activation pattern of a particular gait condition. The predicted EMG patterns closely matched those recorded experimentally, exhibiting the appropriate magnitude and temporal phasing required for each modification. Only 2 of the 96 muscle/gait conditions had RMS errors above 0.10, only 5 muscle/gait conditions exhibited correlations below 0.80 (most were above 0.90) and only 25 muscle/gait conditions deviated outside the normal range of muscle activity for more than 25% of the gait cycle. These results indicate the ability of single network ANNs to represent the transformation between a kinematic movement plan and the necessary muscle activations for normal steady state locomotion but they were also able to generate muscle activation patterns for conditions requiring changes in walking speed, foot placement and foot clearance. The abilities of this type of network have implications towards both the fundamental understanding of the control of locomotion and practical realizations of artificial control systems for use in rehabilitation medicine.     

Design of a robotic gait trainer using spring over muscle actuators for ankle stroke rehabilitation

Repetitive task training is an effective form of rehabilitation for people suffering from debilitating injuries of stroke. We present the design and working concept of a robotic gait trainer (RGT), an ankle rehabilitation device for assisting stroke patients during gait. Structurally based on a tripod mechanism, the device is a parallel robot that incorporates two pneumatically powered, double-acting, compliant, spring over muscle actuators as actuation links which move the ankle in dorsiflex ion/plantarflexion and inversion/eversion. A unique feature in the tripod design is that the human anatomy is part of the robot, the first fixed link being the patient's leg. The kinematics and workspace of the tripod device have been analyzed determining its range of motion. Experimental gait data from an able-bodied person wearing the working RGT prototype are presented.     

Evaluation of lower limb cross planar kinetic connectivity signatures post-stroke

Following stroke, aberrant three dimensional multijoint gait impairments emerge that present in kinematic asymmetries such as circumduction. A precise pattern of cross-planar coordination may underlie abnormal hemiparetic gait as several studies have underscored distinctive neural couplings between medio-lateral control and sagittal plane progression during walking. Here we investigate potential neuromechanical constraints governing abnormal multijoint coordination post-stroke. 15 chronic monohemispheric stroke patients and 10 healthy subjects were recruited. Coupled torque production patterns were assessed using a volitional isometric torque generation task where subjects matched torque targets for a primary joint in 4 directions while receiving visual feedback of the magnitude and direction of the torque. Secondary torques at other lower limb joints were recorded without subject feedback. We find that common features of cross-planar connectivity in stroke subjects include statistically significant frontal to sagittal plane kinetic coupling that overlay a common sagittal plane coupling in healthy subjects. Such coupling is independent of proximal or distal joint control and limb biomechanics. Principal component analysis of the stroke aggregate kinetic signature reveals unique abnormal frontal plane coupling features that explain a larger percentage of the total torque coupling variance. This study supports the idea that coupled cross-planar kinetic outflow between the lower limb joints uniquely emerges during pathological control of frontal plane degrees of freedom resulting in a generalized extension of the limb. It remains to be seen if a pattern of lower limb motor outflow that is centrally mediated contributes to abnormal hemiparetic gait.     

Joint moment work during the stance-to-swing transition in hemiparetic subjects

Following stroke many individuals are left with neurological and functional deficits, including hemiparesis, which impair their ability to walk. Our previous work reported that propulsion of the paretic leg during pre-swing is impaired and may limit gait speed and knee flexion during swing. To elucidate the mechanism of this impairment, we assessed the mechanical work produced by the hip, knee, and ankle moments during pre-swing of the paretic limb in a group of stroke subjects and compared it with the work produced by non-disabled controls walking at similar speeds. Kinematic and kinetic gait data were collected from 23 hemiparetic and 10 control subjects. The hemiparetic subjects walked at their self-selected speeds. The controls walked at their self-selected and two or three slower speeds. Even when compared to controls walking at slow speeds, ankle plantarflexor work during pre-swing was greatly reduced (-0.136+/-0.062J/kg) in the hemiparetic subjects. Differences in hip (+0.006+/-0.020J/kg) and knee (+0.040+/-0.026J/kg) moment work partially offset the reduction in ankle work, but net joint moment work was still significantly reduced (-0.088+/-0.056J/kg). The reduction in work accounts for the low energy of the paretic limb at the stance-to-swing transition previously reported. Future investigation is needed to determine if targeted training of the plantarflexors in the paretic limb improves swing-phase function and locomotor performance in hemiparetic individuals.     

An Artificial Neural Network Estimation of Gait Balance Control in the Elderly Using Clinical Evaluations

The use of motion analysis to assess balance is essential for determining the underlying mechanisms of falls during dynamic activities. Clinicians evaluate patients using clinical examinations of static balance control, gait performance, cognition, and neuromuscular ability. Mapping these data to measures of dynamic balance control, and the subsequent categorization and identification of community dwelling elderly fallers at risk of falls in a quick and inexpensive manner is needed. The purpose of this study was to demonstrate that given clinical measures, an artificial neural network (ANN) could determine dynamic balance control, as defined by the interaction of the center of mass (CoM) with the base of support (BoS), during gait. Fifty-six elderly adults were included in this study. Using a feed-forward neural network with back propagation, combinations of five functional domains, the number of hidden layers and error goals were evaluated to determine the best parameters to assess dynamic balance control. Functional domain input parameters included subject characteristics, clinical examinations, cognitive performance, muscle strength, and clinical balance performance. The use of these functional domains demonstrated the ability to quickly converge to a solution, with the network learning the mapping within 5 epochs, when using up to 30 hidden nodes and an error goal of 0.001. The ability to correctly identify the interaction of the CoM with BoS demonstrated correlation values up to 0.89 (P<.001). On average, using all clinical measures, the ANN was able to estimate the dynamic CoM to BoS distance to within 1 cm and BoS area to within 75 cm². Our results demonstrated that an ANN could be trained to map clinical variables to biomechanical measures of gait balance control. A neural network could provide physicians and patients with a cost effective means to identify dynamic balance issues and possible risk of falls from routinely collected clinical examinations.     

Effects of Quadriceps Muscle Fatigue on Stiff-Knee Gait in Patients with Hemiparesis

The relationship between neuromuscular fatigue and locomotion has never been investigated in hemiparetic patients despite the fact that, in the clinical context, patients report to be more spastic or stiffer after walking a long distance or after a rehabilitation session. The aim of this study was to evaluate the effects of quadriceps muscle fatigue on the biomechanical gait parameters of patients with a stiff-knee gait (SKG). Thirteen patients and eleven healthy controls performed one gait analysis before a protocol of isokinetic quadriceps fatigue and two after (immediately after and after 10 minutes of rest). Spatiotemporal parameters, sagittal knee and hip kinematics, rectus femoris (RF) and vastus lateralis (VL) kinematics and electromyographic (EMG) activity were analyzed. The results showed that quadriceps muscle weakness, produced by repetitive concentric contractions of the knee extensors, induced an improvement of spatiotemporal parameters for patients and healthy subjects. For the patient group, the increase in gait velocity and step length was associated with i) an increase of sagittal hip and knee flexion during the swing phase, ii) an increase of the maximal normalized length of the RF and VL and of the maximal VL lengthening velocity during the pre-swing and swing phases, and iii) a decrease in EMG activity of the RF muscle during the initial pre-swing phase and during the latter 2/3 of the initial swing phase. These results suggest that quadriceps fatigue did not alter the gait of patients with hemiparesis walking with a SKG and that neuromuscular fatigue may play the same functional role as an anti-spastic treatment such as botulinum toxin-A injection. Strength training of knee extensors, although commonly performed in rehabilitation, does not seem to be a priority to improve gait of these patients.     

Fatigue compensation during FES using surface EMG

Muscle fatigue limits the effectiveness of FES when applied to regain functional movements in spinal cord injured (SCI) individuals. The stimulation intensity must be manually increased to provide more force output to compensate for the decreasing muscle force due to fatigue. An artificial neural network (ANN) system was designed to compensate for muscle fatigue during functional electrical stimulation (FES) by maintaining a constant joint angle. Surface electromyography signals (EMG) from electrically stimulated muscles were used to determine when to increase the stimulation intensity when the muscle’s output started to drop.

In two separate experiments on able-bodied subjects seated in hard back chairs, electrical stimulation was continuously applied to fatigue either the biceps (during elbow flexion) or the quadriceps muscle (during leg extension) while recording the surface EMG. An ANN system was created using processed surface EMG as the input, and a discrete fatigue compensation control signal, indicating when to increase the stimulation current, as the output. In order to provide training examples and test the systems’ performance, the stimulation current amplitude was manually increased to maintain constant joint angles. Manual stimulation amplitude increases were required upon observing a significant decrease in the joint angle. The goal of the ANN system was to generate fatigue compensation control signals in an attempt to maintain a constant joint angle.

On average, the systems could correctly predict 78.5% of the instances at which a stimulation increase was required to maintain the joint angle. The performance of these ANN systems demonstrates the feasibility of using surface EMG feedback in an FES control system.     

Measuring movement irregularity in the upper motor neurone syndrome using normalised average rectified jerk.

Movement irregularity is a feature of the upper motor neurone (UMN) syndrome which is difficult to measure. Average rectified jerk (ARJ) has been proposed as a measure of this movement irregularity, but ARJ depends upon the duration of movement. Since movements may be slower in UMN patients, duration dependence compromises ARJ as a measure of irregularity. A normalisation technique is proposed that generates a measure of movement irregularity which is independent of movement duration: normalised average rectified jerk (NARJ). This paper presents a validation of NARJ in the UMN syndrome. Nine control subjects, nine left hemiparetic stroke patients and nine right hemiparetic stroke patients were studied. Test movements comprised elbow extension/flexion in the horizontal plane; these were recorded with an electro-goniometer and accelerometer. The effectiveness of the normalisation technique has been demonstrated using trajectories of various durations; some of these were artificially generated from participants' trajectories, in order to preserve the movement profile. The variability of NARJ and ARJ have been compared in a sample of control subjects. NARJ has been criterion validated by correlation with expert subjective rating of irregularity in a heterogeneous set of trajectories. Construct validity has been tested by discrimination between movements of control subjects, left hemiparetic stroke patients and right hemiparetic stroke patients. When comparing trajectories of identical profile but two-fold difference in movement duration, NARJ differed only 2.6% whereas ARJ differed 706%. NARJ was less reproducible in healthy participants than ARJ: median non-parametric coefficients of variation for repeated movements were 55% and 41%, respectively. Spearman rank correlation coefficient for NARJ and expert rating was 0.92 (p<0.01). NARJ measurements on right hemiparetic patients differ significantly from those made on the control group (p<0.02); corresponding ARJ measurements do not attain statistical significance. NARJ is a valid measure of movement irregularity in the UMN syndrome.     

Neural network estimation of balance control during locomotion

Gait patterns of the elderly are often adjusted to accommodate for reduced function in the balance control system and a general reduction in skeletal muscle strength. Recent studies have demonstrated that measures related to motion of whole body center of mass (COM) can distinguish elderly individuals with balance impairment from healthy peers. Accurate COM estimation requires a multiple-segment anthropometric model, which may restrict its broad application in assessment of dynamic instability. Although temporal-distance measures and electromyography have been used in evaluation of overall gait function and determination of gait dysfunction, no studies have examined the use of gait measurements in predicting COM motion during gait. The purpose of this study was to demonstrate the effectiveness of an artificial neural network (ANN) model in mapping gait measurements onto COM motion in the frontal plane. Data from 40 subjects of varied age and balance impairment were entered into a 3-layer feed-forward model with back-propagated error correction. Bootstrap re-sampling was used to enhance the generalization accuracy of the model, using 20 re-sampling trials. The ANN model required minimal processing time (5 epochs, with 20 hidden units) and accurately mapped COM motion (R-values up to 0.89). As training proportion and number of hidden units increased, so did model accuracy. Overall, this model appears to be effective as a mapping tool for estimating balance control during locomotion. With easily obtained gait measures as input and a simple, computationally efficient architecture, the model may prove useful in clinical scenarios where electromyography equipment exists.     

Postural and dynamic balance while walking in adults with incomplete spinal cord injury

The purpose of this study was to characterize balance in individuals with and without an incomplete spinal cord injury (ISCI) during the single support phase of gait. Thirty-four individuals (17 with a ISCI, 17 able-bodied) walked at their self-selected walking speed. Among those, eighteen individuals (9 with ISCI, 9 able-bodied) with a similar walking speed were also analyzed. Stabilizing and destabilizing forces quantified balance during the single support phase of gait. The biomechanical factors included in the equation of the stabilizing and destabilizing forces served as explanatory factors. Individuals with ISCI had a lower stabilizing force and a higher destabilizing force compared to able-bodied individuals. The main explanatory factors of the forces extracted from the equations were the speed of the center of mass (maximal stabilizing force) and the distance between the center of pressure and the base of support (minimal destabilizing force). Only the minimal destabilizing force was significantly different among subgroups with a similar walking speed. The stabilizing and destabilizing forces suggest that individuals with ISCI were more stable than able-bodied, which was achieved by walking more slowly – which decrease the speed of the center of mass – and keeping the center of pressure away from the margin of the base of support in order to maintain balance within their range of physical ability.     

Integration of wavelet transform with PCA and ANN for metabolomics data-mining

PCA (principal components analysis) and ANN (artificial neural network) are two broadly used pattern recognition methods in metabolomics data-mining. Yet their limitations sometimes are great obstacles for researchers. In this paper the wavelet transform (WT) method was used to integrate with PCA and ANN to improve their performance in manipulating metabolomics data. A dataset was decomposed by wavelets and then reconstructed. The "hard thresholding" algorithm was used, through which the detail information was discarded, and the entire "metabolomics image" reconstructed on the significant information. It was supposed that the most relevant information was captured after this process. It was found that, thanks to its ability in denoising data, the WT method could significantly improve the performance of the non-linear essence-extracting method ANN in classifying samples; further integration of WT with PCA showed that WT could greatly enhance the ability of PCA in distinguishing one group of samples from another and also its ability in identifying potential biomarkers. The results highlighted WT as a promising resolution in bridging the gap between huge bytes of data and the instructive biological information.     

Variations in Kinematics during Clinical Gait Analysis in Stroke Patients

In addition to changes in spatio-temporal and kinematic parameters, patients with stroke exhibit fear of falling as well as fatigability during gait. These changes could compromise interpretation of data from gait analysis. The aim of this study was to determine if the gait of hemiplegic patients changes significantly over successive gait trials. Forty two stroke patients and twenty healthy subjects performed 9 gait trials during a gait analysis session. The mean and variability of spatio-temporal and kinematic joint parameters were analyzed during 3 groups of consecutive gait trials (1–3, 4–6 and 7–9). Principal component analysis was used to reduce the number of variables from the joint kinematic waveforms and to identify the parts of the gait cycle which changed during the gait analysis session. The results showed that i) spontaneous gait velocity and the other spatio-temporal parameters significantly increased, and ii) gait variability decreased, over the last 6 gait trials compared to the first 3, for hemiplegic patients but not healthy subjects. Principal component analysis revealed changes in the sagittal waveforms of the hip, knee and ankle for hemiplegic patients after the first 3 gait trials. These results suggest that at the beginning of the gait analysis session, stroke patients exhibited phase of adaptation,characterized by a “cautious gait” but no fatigue was observed.     

Dynamic Stability and Risk of Tripping during the Timed Up and Go Test in Hemiparetic and Healthy Subjects

Background

The Timed Up and Go (TUG) test is often used to estimate risk of falls. Foot clearance and displacement of the center of mass (COM), which are related to risk of tripping and dynamic stability have never been evaluated during the TUG. Accurate assessment of these parameters using instrumented measurements would provide a comprehensive assessment of risk of falls in hemiparetic patients. The aims of this study were to analyze correlations between TUG performance time and displacement of the COM and foot clearance in patients with stroke-related hemiparesis and healthy subjects during the walking and turning sub-tasks of the TUG and to compare these parameters between fallers and non-fallers.

Methods

29 hemiparetic patients and 25 healthy subjects underwent three-dimensional gait analysis during the TUG test. COM and foot clearance were analyzed during the walking and turning sub-tasks of the TUG.

Results

Lateral displacement of the COM was greater and faster during the walking sub-tasks and vertical displacement of the COM was greater during the turn in the patients compared to the healthy subjects (respectively p<0.01 and p<0.05). Paretic foot clearance was greater during walking and displacement of the COM was slower during the turn in the patients (p<0.01). COM displacement and velocity during the turn were correlated with TUG performance in the patients, however, vertical COM displacement was not. These correlations were significant in the healthy subjects. There were no differences between COM parameters or foot clearance in fallers and non-fallers.

Discussion and Conclusion

Hemiparetic patients are less stable than healthy subjects, but compensate with a cautious gait to avoid tripping. Instrumented analysis of the TUG test appears relevant for the assessment of dynamic stability in hemiparetic patients, providing more information than straight-line gait.     

Combining muscle synergies and biomechanical analysis to assess gait in stroke patients

The understanding of biomechanical deficits and impaired neural control of gait after stroke is crucial to prescribe effective customized treatments aimed at improving walking function. Instrumented gait analysis has been increasingly integrated into the clinical practice to enhance precision and inter-rater reliability for the assessment of pathological gait. On the other hand, the analysis of muscle synergies has gained relevance as a novel tool to describe the neural control of walking. Since muscle synergies and gait analysis capture different but equally important aspects of walking, we hypothesized that their combination can improve the current clinical tools for the assessment of walking performance.To test this hypothesis, we performed a complete bilateral, lower limb biomechanical and muscle synergies analysis on nine poststroke hemiparetic patients during overground walking. Using stepwise multiple regression, we identified a number of kinematic, kinetic, spatiotemporal and synergy-related features from the paretic and non-paretic side that, combined together, allow to predict impaired walking function better than the Fugl-Meyer Assessment score. These variables were time of peak knee flexion, VAF_total values, duration of stance phase, peak of paretic propulsion and range of hip flexion. Since these five variables describe important biomechanical and neural control features underlying walking deficits poststroke, they may be feasible to drive customized rehabilitation therapies aimed to improve walking function.This paper demonstrates the feasibility of combining biomechanical and neural-related measures to assess locomotion performance in neurologically injured individuals.     

Validation of gait event detection by centre of pressure during target stepping in healthy and paretic gait

BackgroundTarget-stepping paradigms are increasingly used to assess and train gait adaptability. Accurate gait-event detection (GED) is key to locating targets relative to the ongoing step cycle as well as measuring foot-placement error. In the current literature GED is either based on kinematics or centre of pressure (CoP), and both have been previously validated with young healthy individuals. However, CoP based GED has not been validated for stroke survivors who demonstrate altered CoP pattern.MethodsYoung healthy adults and individuals affected by stroke stepped to targets on a treadmill, while gait events were measured using three detection methods; verticies of CoP cyclograms, and two kinematic criteria, (1) vertical velocity and position and of the heel marker, (2) anterior velocity and position of the heel and toe marker, were used. The percentage of unmatched gait events was used to determine the success of the GED method. The difference between CoP and kinematic GED methods were tested with two one sample (two-tailed) t-tests against a reference value of zero. Differences between group and paretic and non-paretic leg were tested with a repeated measures ANOVA.ResultsThe kinematic method based on vertical velocity only detected about 80% of foot contact events on the paretic side in stroke survivors while the method on anterior velocity was more successful in both young healthy adults as stroke survivors (3% young healthy and 7% stroke survivors unmatched). Both kinematic methods detected gait events significantly earlier than CoP GED (p < 0.001) except for foot contact in stroke survivors based on the vertical velocity.ConclusionsCoP GED may be more appropriate for gait analyses of SS than kinematic methods; even when walking and varying steps.     

Prediction of effect of pegylated interferon alpha-2b plus ribavirin combination therapy in patients with chronic hepatitis C infection

Treatment with pegylated interferon alpha-2b (PEGIFN) plus ribavirin (RBV) is standard therapy for patients with chronic hepatitis C. Although the effectiveness, patients with high titres of group Ib hepatitis C virus (HCV) respond poorly compared to other genotypes. At present, we cannot predict the effect in an individual. Previous studies have used traditional statistical analysis by assuming a linear relationship between clinical features, but most phenomena in the clinical situation are not linearly related. The aim of this study is to predict the effect of PEG IFN plus RBV therapy on an individual patient level using an artificial neural network system (ANN). 156 patients with HCV group 1b from multiple centres were treated with PEGIFN (1.5 μg/kg) plus RBV (400-1000 mg) for 48 weeks. Data on the patients' demographics, laboratory tests, PEGIFN, and RBV doses, early viral responses (EVR), and sustained viral responses were collected. Clinical data were randomly divided into training data set and validation data set and analyzed using multiple logistic regression analysis (MLRs) and ANN to predict individual outcomes. The sensitivities of predictive expression were 0.45 for the MLRs models and 0.82 for the ANNs and specificities were 0.55 for the MLR and 0.88 for the ANN. Non-linear relation analysis showed that EVR, serum creatinine, initial dose of Ribavirin, gender and age were important predictive factors, suggesting non-linearly related to outcome. In conclusion, ANN was more accurate than MLRs in predicting the outcome of PEGIFN plus RBV therapy in patients with group 1b HCV.     

An upper extremity kinematic model for evaluation of hemiparetic stroke

Quantification of rehabilitation progress is necessary for accurately assessing clinical treatments. A three-dimension (3D) upper extremity (UE) kinematic model was developed to obtain joint angles of the trunk, shoulder and elbow using a Vicon motion analysis system. Strict evaluation confirmed the system's accuracy and precision. As an example of application, the model was used to evaluate the upper extremity movement of eight hemiparetic stroke patients with spasticity, while completing a set of reaching tasks. Main outcome measures include kinematic variables of movement time, range of motion, peak angular velocity, and percentage of reach where peak velocity occurs. The model computed motion patterns in the affected and unaffected arms. The unaffected arm showed a larger range of motion and higher angular velocity than the affected arm. Frequency analysis (power spectrum) demonstrated lower frequency content for elbow angle and angular velocity in the affected limb when compared to the unaffected limb. The model can accurately quantify UE arm motion, which may aid in the assessment and planning of stroke rehabilitation, and help to shorten recovery time.