Quality and Quantity of Rehabilitation Exercises Delivered By A 3-D Motion Controlled Camera: A Pilot Study

EGCN++: A New Fusion Strategy for Ensemble Learning in Skeleton-Based Rehabilitation Exercise Assessment

Skeleton-based exercise assessment focuses on evaluating the correctness or quality of an exercise performed by a subject. Skeleton data provide two groups of features (i.e., position and orientation), which existing methods have not fully harnessed. We previously proposed an ensemble-based graph convolutional network (EGCN) that considers both position and orientation features to construct a model-based approach. Integrating these types of features achieved better performance than available methods. However, EGCN lacked a fusion strategy across the data, feature, decision, and model levels. In this paper, we present an advanced framework, EGCN++, for rehabilitation exercise assessment. Based on EGCN, a new fusion strategy called MLE-PO is proposed for EGCN++; this technique considers fusion at the data and model levels. We conduct extensive cross-validation experiments and investigate the consistency between machine and human evaluations on three datasets: UI-PRMD, KIMORE, and EHE. Results demonstrate that MLE-PO outperforms other EGCN ensemble strategies and representative baselines. Furthermore, the MLE-PO's model evaluation scores are more quantitatively consistent with clinical evaluations than other ensemble strategies.

An online self-study mindfulness-based stress reduction course for people suffering from mental fatigue after an acquired brain injury

OBJECTIVE

The Mindfulness-Based Stress Reduction (MBSR) program has shown promising results for people suffering from mental fatigue after an acquired brain injury. The aim was to evaluate the feasibility of a MBSR program performed as an online self-study course for this group of people.

METHODS

Sixty participants who had suffered an acquired brain injury with lasting mental fatigue were randomized to an online MBSR course or to a waitlist control group. They answered self-report questionnaires before start and after the course.

RESULTS

Sixteen completed the MBSR program. With the repeated ANOVA no significant difference between groups was found, although there was a significant change in time (the repetition factor). The post-hoc paired t-test indicated a significant reduction and a large-to-median effect size in mental fatigue (p = 0.003, d = 0.896), depression (p = 0.038, d = 0.569) and anxiety (p = 0.030, d = 0.598) for the MBSR group. No significant changes were found for the control group.

CONCLUSION

An online self-study MBSR program for people suffering from mental fatigue after an acquired brain injury can be a feasible option for those suffering from less severe mental fatigue and emotional symptoms, while others may require a program adapted to their needs.

Exercise quantification from single camera view markerless 3D pose estimation

Sports physiotherapists and coaches are tasked with evaluating the movement quality of athletes across the spectrum of ability and experience. However, the accuracy of visual observation is low and existing technology outside of expensive lab-based solutions has limited adoption, leading to an unmet need for an efficient and accurate means to measure static and dynamic joint angles during movement, converted to movement metrics useable by practitioners. This paper proposes a set of pose landmarks for computing frequently used joint angles as metrics of interest to sports physiotherapists and coaches in assessing common strength-building human exercise movements. It then proposes a set of rules for computing these metrics for a range of common exercises (single and double drop jumps and counter-movement jumps, deadlifts and various squats) from anatomical key-points detected using video, and evaluates the accuracy of these using a published 3D human pose model trained with ground truth data derived from VICON motion capture of common rehabilitation exercises. Results show a set of mathematically defined metrics which are derived from the chosen pose landmarks, and which are sufficient to compute the metrics for each of the exercises under consideration. Comparison to ground truth data showed that root mean square angle errors were within 10° for all exercises for the following metrics: shin angle, knee varus/valgus and left/right flexion, hip flexion and pelvic tilt, trunk angle, spinal flexion lower/upper/mid and rib flare. Larger errors (though still all within 15°) were observed for shoulder flexion and ASIS asymmetry in some exercises, notably front squats and drop-jumps. In conclusion, the contribution of this paper is that a set of sufficient key-points and associated metrics for exercise assessment from 3D human pose have been uniquely defined. Further, we found generally very good accuracy of the Strided Transformer 3D pose model in predicting these metrics for the chosen set of exercises from a single mobile device camera, when trained on a suitable set of functional exercises recorded using a VICON motion capture system. Future assessment of generalization is needed.

Physical Activity in Cerebral Palsy: A Current State Study

This document analyzes a survey conducted in three geographical areas in Spain, focusing on centers for individuals with cerebral palsy (CP). The study aims to determine the adherence rate to recommended physical activity guidelines, assess if there is a decline in interest in physical activity over time, identify the stage at which this decline occurs, and explore potential mechanisms, tools, or strategies to sustain long-term engagement in regular physical activity for this population. The 36-item questionnaire comprises multiple-choice, open-ended, and Likert scale-type questions. Data were collected on physical activity frequency and duration, daily living activities, and demographics. Statistical analysis identified patterns and relationships between variables. Findings reveal that only a 17.6% meets the World Health Organization (WHO) recommendations regarding regular physical activity (RPA), decreasing in frequency or number of days a week, (3.7 d/w to 2.9 d/w; p < 0.01) and duration (50.5 min/d to 45.2 min/d; p < 0.001) with age, especially for those with higher Gross Motor Function Classification System (GMFCS) mobility levels. Obesity slightly correlates with session duration (ρ = -0.207; p < 0.05), not mobility limitations. Gender has no significant impact on mobility, communication, or physical activity, while age affects variables such as body mass index (BMI) and engagement (p < 0.01). A substantial proportion follows regular physical activities based on health professionals' advice, with interest decreasing with age. To improve adherence, focusing on sports-oriented goals, group sessions, and games is recommended. These findings emphasize the importance of personalized programs, particularly for older individuals and those with greater mobility limitations.

A deep learning system to monitor and assess rehabilitation exercises in home-based remote and unsupervised conditions

In the domain of physical rehabilitation, the progress in machine learning and the availability of cost-effective motion capture technologies have paved the way for innovative systems capable of capturing human movements, automatically analyzing recorded data, and evaluating movement quality. This study introduces a novel, economically viable system designed for monitoring and assessing rehabilitation exercises. The system enables real-time evaluation of exercises, providing precise insights into deviations from correct execution. The evaluation comprises two significant components: range of motion (ROM) classification and compensatory pattern recognition. To develop and validate the effectiveness of the system, a unique dataset of 6 resistance training exercises was acquired. The proposed system demonstrated impressive capabilities in motion monitoring and evaluation. Notably, we achieved promising results, with mean accuracies of 89% for evaluating ROM-class and 98% for classifying compensatory patterns. By complementing conventional rehabilitation assessments conducted by skilled clinicians, this cutting-edge system has the potential to significantly improve rehabilitation practices. Additionally, its integration in home-based rehabilitation programs can greatly enhance patient outcomes and increase access to high-quality care.

UCO Physical Rehabilitation: New Dataset and Study of Human Pose Estimation Methods on Physical Rehabilitation Exercises

Physical rehabilitation plays a crucial role in restoring motor function following injuries or surgeries. However, the challenge of overcrowded waiting lists often hampers doctors' ability to monitor patients' recovery progress in person. Deep Learning methods offer a solution by enabling doctors to optimize their time with each patient and distinguish between those requiring specific attention and those making positive progress. Doctors use the flexion angle of limbs as a cue to assess a patient's mobility level during rehabilitation. From a Computer Vision perspective, this task can be framed as automatically estimating the pose of the target body limbs in an image. The objectives of this study can be summarized as follows: (i) evaluating and comparing multiple pose estimation methods; (ii) analyzing how the subject's position and camera viewpoint impact the estimation; and (iii) determining whether 3D estimation methods are necessary or if 2D estimation suffices for this purpose. To conduct this technical study, and due to the limited availability of public datasets related to physical rehabilitation exercises, we introduced a new dataset featuring 27 individuals performing eight diverse physical rehabilitation exercises focusing on various limbs and body positions. Each exercise was recorded using five RGB cameras capturing different viewpoints of the person. An infrared tracking system named OptiTrack was utilized to establish the ground truth positions of the joints in the limbs under study. The results, supported by statistical tests, show that not all state-of-the-art pose estimators perform equally in the presented situations (e.g., patient lying on the stretcher vs. standing). Statistical differences exist between camera viewpoints, with the frontal view being the most convenient. Additionally, the study concludes that 2D pose estimators are adequate for estimating joint angles given the selected camera viewpoints.

A Contrastive Learning Network for Performance Metric and Assessment of Physical Rehabilitation Exercises

Human activity analysis in the legal monitoring environment plays an important role in the physical rehabilitation field, as it helps patients with physical injuries improve their postoperative conditions and reduce their medical costs. Recently, several deep learning-based action quality assessment (AQA) frameworks have been proposed to evaluate physical rehabilitation exercises. However, most of them treat this problem as a simple regression task, which requires both the action instance and its score label as input. This approach is limited by the fact that the annotations in this field usually consist of healthy or unhealthy labels rather than quality scores provided by professional physicians. Additionally, most of these methods cannot provide informative feedback on a patient's motion defects, which weakens their practical application. To address these problems, we propose a multi-task contrastive learning framework to learn subtle and critical differences from skeleton sequences to deal with the performance metric and AQA problems of physical rehabilitation exercises. Specifically, we propose a performance metric network that takes triplets of training samples as input for score generation. For the AQA task, the same contrast learning strategy is used, but pairwise training samples are fed into the action quality assessment network for score prediction. Notably, we propose quantifying the deviation of the joint attention matrix between different skeleton sequences and introducing it into the loss function of our learning network. It is proven that considering both score prediction loss and joint attention deviation loss improves physical exercises AQA performance. Furthermore, it helps to obtain informative feedback for patients to improve their motion defects by visualizing the joint attention matrix's difference. The proposed method is verified on the UI-PRMD and KIMORE datasets. Experimental results show that the proposed method achieves state-of-the-art performance.

A Low-Cost System Using a Big-Data Deep-Learning Framework for Assessing Physical Telerehabilitation: A Proof-of-Concept

The consolidation of telerehabilitation for the treatment of many diseases over the last decades is a consequence of its cost-effective results and its ability to offer access to rehabilitation in remote areas. Telerehabilitation operates over a distance, so vulnerable patients are never exposed to unnecessary risks. Despite its low cost, the need for a professional to assess therapeutic exercises and proper corporal movements online should also be mentioned. The focus of this paper is on a telerehabilitation system for patients suffering from Parkinson's disease in remote villages and other less accessible locations. A full-stack is presented using big data frameworks that facilitate communication between the patient and the occupational therapist, the recording of each session, and real-time skeleton identification using artificial intelligence techniques. Big data technologies are used to process the numerous videos that are generated during the course of treating simultaneous patients. Moreover, the skeleton of each patient can be estimated using deep neural networks for automated evaluation of corporal exercises, which is of immense help to the therapists in charge of the treatment programs.

Autonomous modeling of repetitive movement for rehabilitation exercise monitoring

BACKGROUND

Insightful feedback generation for daily home-based stroke rehabilitation is currently unavailable due to the inefficiency of exercise inspection done by therapists. We aim to produce a compact anomaly representation that allows a therapist to pay attention to only a few specific sections in a long exercise session record and boost their efficiency in feedback generation.

METHODS

This study proposes a data-driven technique to model a repetitive exercise using unsupervised phase learning on an artificial neural network and statistical learning on principal component analysis (PCA). After a model is built on a set of normal healthy movements, the model can be used to extract a sequence of anomaly scores from a movement of the same prescription.

RESULTS

The method not only works on a standard marker-based motion capture system but also performs well on a more compact and affordable motion capture system based-on Kinect V2 and wrist-worn inertial measurement units that can be used at home. An evaluation of four different exercises shows its potential in separating anomalous movements from normal ones with an average area under the curve (AUC) of 0.9872 even on the compact motion capture system.

CONCLUSIONS

The proposed processing technique has the potential to help clinicians in providing high-quality feedback for telerehabilitation in a more scalable way.

A Study on Physical Exercise and General Mobility in People with Cerebral Palsy: Health through Costless Routines

Sedentary behavior (SB) is a common problem that may produce health issues in people with cerebral palsy (CP). When added to a progressive reduction in motor functions over time, SB can lead to higher percentages of body fat, muscle stiffness and associated health issues in this population. Regular physical activity (RPA) may prevent the loss of motor skills and reduce health risks. In this work, we analyzed data collected from 40 people (20 children and teenagers, and 20 adults) who attend two specialist centers in Seville to obtain an up-to-date picture regarding the practice of RPA in people with CP. Roughly 60% of the participants showed mostly mid/severe mobility difficulties, while 38% also had communicative issues. Most of the participants performed light-intensity physical activity (PA) at least once or twice a week and, in the majority of cases, had a neutral or positive attitude to exercising. In the Asociación Sevillana de Parálisis Cerebral (ASPACE) sample test, the higher the International Classification of Functioning, Disability and Health (ICF), the higher the percentage of negative responses to doing exercise. Conversely, in the Centro Específico de Educación Especial Mercedes Sanromá (CEEEMS), people likes PA but slightly higher ratios of positive responses were found at Gross Motor Function Classification System (GMFCS) levels V and II, agreeing with the higher personal engagement of people at those levels. We have also performed a literature review regarding RPA in CP and the use of low-cost equipment. As a conclusion, we found that RPA produces enormous benefits for health and motor functions, whatever its intensity and duration. Costless activities such as walking, running or playing sports; exercises requiring low-cost equipment such as elastic bands, certain smartwatches or video-games; or therapies with animals, among many others, have all demonstrated their suitability for such a purpose.

App-Based Feedback for Rehabilitation Exercise Correction in Patients With Knee or Hip Osteoarthritis: Prospective Cohort Study

Background

The use of digital therapeutic solutions for rehabilitation of conditions such as osteoarthritis provides scalable access to rehabilitation. Few validated technological solutions exist to ensure supervision of users while they exercise at home. Motion Coach (Kaia Health GmbH) provides audiovisual feedback on exercise execution in real time on conventional smartphones.

Objective

We hypothesized that the interrater agreement between physiotherapists and Motion Coach would be noninferior to physiotherapists’ interrater agreement for exercise evaluations in a cohort with osteoarthritis.

Methods

Patients diagnosed with osteoarthritis of the knee or hip were recruited at a university hospital to perform a set of 6 exercises. Agreement between Motion Coach and 2 physiotherapists’ corrections for segments of the exercises were compared using Cohen κ and percent agreement.

Results

Participants (n=24) were enrolled and evaluated. There were no significant differences between interrater agreements (Motion Coach app vs physiotherapists: percent agreement 0.828; physiotherapist 1 vs physiotherapist 2: percent agreement 0.833; P<.001). Age (70 years or under, older than 70 years), gender (male, female), or BMI (30 kg/m² or under, greater than 30 kg/m²) subgroup analysis revealed no detectable difference in interrater agreement. There was no detectable difference in levels of interrater agreement between Motion Coach vs physiotherapists and between physiotherapists in any of the 6 exercises.

Conclusions

The results demonstrated that Motion Coach is noninferior to physiotherapist evaluations. Interrater agreement did not differ between 2 physiotherapists or between physiotherapists and the Motion Coach app. This finding was valid for all investigated exercises and subgroups. These results confirm the ability of Motion Coach to detect user form during exercise and provide valid feedback to users with musculoskeletal disorders.

Design of a Low-Cost, Wearable Device for Kinematic Analysis in Physical Therapy Settings

BACKGROUND

Unsupervised home exercise is a major component of physical therapy (PT). This study proposes an inexpensive, inertial measurement unit-based wearable device to capture kinematic data to facilitate exercise. However, conveying and interpreting kinematic data to non-experts poses a challenge due to the complexity and background knowledge required that most patients lack.

OBJECTIVES

The objectives of this study were to identify key user interface and user experience features that would likely improve device adoption and assess participant receptiveness toward the device.

METHODS

Fifty participants were recruited to perform nine upper extremity exercises while wearing the device. Prior to exercise, participants completed an orientation of the device, which included examples of software graphics with exercise data. Surveys that measured receptiveness toward the device, software graphics, and ergonomics were given before and after exercise.

RESULTS

Participants were highly receptive to the device with 90% of the participants likely to use the device during PT. Participants understood how the simple kinematic data could be used to aid exercise, but the data could be difficult to comprehend with more complex movements. Devices should incorporate wireless sensors and emphasize ease of wear.

CONCLUSION

Device-guided home physical rehabilitation can allow for individualized treatment protocols and improve exercise self-efficacy through kinematic analysis. Future studies should implement clinical testing to evaluate the impact a wearable device can have on rehabilitation outcomes.

Evaluation of Machine Learning Models for Classifying Upper Extremity Exercises Using Inertial Measurement Unit-Based Kinematic Data

The amount of home-based exercise prescribed by a physical therapist is difficult to monitor. However, the integration of wearable inertial measurement unit (IMU) devices can aid in monitoring home exercise by analyzing exercise biomechanics. The objective of this study is to evaluate machine learning models for classifying nine different upper extremity exercises, based upon kinematic data captured from an IMU-based device. Fifty participants performed one compound and eight isolation exercises with their right arm. Each exercise was performed ten times for a total of 4500 trials. Joint angles were calculated using IMUs that were placed on the hand, forearm, upper arm, and torso. Various machine learning models were developed with different algorithms and train-test splits. Random forest models with flattened kinematic data as a feature had the greatest accuracy (98.6%). Using triaxial joint range of motion as the feature set resulted in decreased accuracy (91.9%) with faster speeds. Accuracy did not decrease below 90% until training size was decreased to 5% from 50%. Accuracy decreased (88.7%) when splitting data by participant. Upper extremity exercises can be classified accurately using kinematic data from a wearable IMU device. A random forest classification model was developed that quickly and accurately classified exercises. Sampling frequency and lower training splits had a modest effect on performance. When the data were split by subject stratification, larger training sizes were required for acceptable algorithm performance. These findings set the basis for more objective and accurate measurements of home-based exercise using emerging healthcare technologies.

A review of computational approaches for evaluation of rehabilitation exercises

Recent advances in data analytics and computer-aided diagnostics stimulate the vision of patient-centric precision healthcare, where treatment plans are customized based on the health records and needs of every patient. In physical rehabilitation, the progress in machine learning and the advent of affordable and reliable motion capture sensors have been conducive to the development of approaches for automated assessment of patient performance and progress toward functional recovery. The presented study reviews computational approaches for evaluating patient performance in rehabilitation programs using motion capture systems. Such approaches will play an important role in supplementing traditional rehabilitation assessment performed by trained clinicians, and in assisting patients participating in home-based rehabilitation. The reviewed computational methods for exercise evaluation are grouped into three main categories: discrete movement score, rule-based, and template-based approaches. The review places an emphasis on the application of machine learning methods for movement evaluation in rehabilitation. Related work in the literature on data representation, feature engineering, movement segmentation, and scoring functions is presented. The study also reviews existing sensors for capturing rehabilitation movements and provides an informative listing of pertinent benchmark datasets. The significance of this paper is in being the first to provide a comprehensive review of computational methods for evaluation of patient performance in rehabilitation programs.

A Deep Learning Framework for Assessing Physical Rehabilitation Exercises

Computer-aided assessment of physical rehabilitation entails evaluation of patient performance in completing prescribed rehabilitation exercises, based on processing movement data captured with a sensory system. Despite the essential role of rehabilitation assessment toward improved patient outcomes and reduced healthcare costs, existing approaches lack versatility, robustness, and practical relevance. In this paper, we propose a deep learning-based framework for automated assessment of the quality of physical rehabilitation exercises. The main components of the framework are metrics for quantifying movement performance, scoring functions for mapping the performance metrics into numerical scores of movement quality, and deep neural network models for generating quality scores of input movements via supervised learning. The proposed performance metric is defined based on the log-likelihood of a Gaussian mixture model, and encodes low-dimensional data representation obtained with a deep autoencoder network. The proposed deep spatio-temporal neural network arranges data into temporal pyramids, and exploits the spatial characteristics of human movements by using sub-networks to process joint displacements of individual body parts. The presented framework is validated using a dataset of ten rehabilitation exercises. The significance of this work is that it is the first that implements deep neural networks for assessment of rehabilitation performance.

Assessment of physical rehabilitation movements through dimensionality reduction and statistical modeling

The article proposes a method for evaluation of the consistency of human movements within the context of physical therapy and rehabilitation. Captured movement data in the form of joint angular displacements in a skeletal human model is considered in this work. The proposed approach employs an autoencoder neural network to project the high-dimensional motion trajectories into a low-dimensional manifold. Afterwards, a Gaussian mixture model is used to derive a parametric probabilistic model of the density of the movements. The resulting probabilistic model is employed for evaluation of the consistency of unseen motion sequences based on the likelihood of the data being drawn from the model. The approach is validated on two physical rehabilitation movements.

A quantitative method for evaluation of 6 degree of freedom virtual reality systems

Modern virtual reality systems such as the HTC Vive enable users to be immersed in a virtual world. Validation of the HTC Vive and other contemporaneous systems for use in clinic, research, and industry applications will assure users and developers that games and applications made for these systems are accurate representations of the real world. The purpose of this study was to develop a standardized method for testing the translational and rotational capabilities of VR systems such as the HTC Vive. The translational and rotational capabilities of the HTC Vive were investigated using an industry grade robot arm and a gold standard motion capture system. It was found that the average difference between reported translational distances traveled was 0.74 ± 0.42 mm for all room-scale calibration trials and 0.63 ± 0.27 mm for all standing calibration trials. The mean difference in angle rotated was 0.46 ± 0.46° for all room-scale calibration trials and 0.66 ± 0.40° for all standing calibration trials. When tested using human movement, the average difference in distance traveled was 3.97 ± 3.37 mm. Overall, the HTC Vive shows promise as a tool for clinic, research, and industry and its controllers can be accurately tracked in a variety of situations. The methodology used for this study can easily be replicated for other VR systems so that direct comparisons can be made as new systems become available.

Would a thermal sensor improve arm motion classification accuracy of a single wrist-mounted inertial device?

Background

Inertial Measurement Unit (IMU)-based wearable sensors have found common use to track arm activity in daily life. However, classifying a high number of arm motions with single IMU-based systems still remains a challenging task. This paper explores the possibility to increase the classification accuracy of these systems by incorporating a thermal sensor. Increasing the number of arm motions that can be classified is relevant to increasing applicability of single-device wearable systems for a variety of applications, including activity monitoring for athletes, gesture control for video games, and motion classification for physical rehabilitation patients. This study explores whether a thermal sensor can increase the classification accuracy of a single-device motion classification system when evaluated with healthy participants. The motions performed are reproductions of exercises described in established rehabilitation protocols.

Methods

A single wrist-mounted device was built with an inertial sensor and a thermal sensor. This device was worn on the wrist, was battery powered, and transmitted data over Bluetooth to computer during recording. A LabVIEW Graphical User Interface (GUI) instructed the user to complete 24 different arm motions in a pre-randomized order. The received data were pre-processed, and secondary features were calculated on these data. These features were processed with Principal Component Analysis (PCA) for dimensionality reduction and then several machine learning models were applied to select the optimal model based on speed and accuracy. To test the effectiveness of the scheme, 11 healthy subjects participated in the trials.

Results

Average personalized classification model accuracies of 93.55% were obtained for 11 healthy participants. Generalized model accuracies of 82.5% indicated that the device can classify arm motions on a user without prior training. The addition of a thermal sensor significantly increased classification accuracy of a single wrist-mounted inertial device, from 75 to 93.55%, (F(1,20) = 90.53, p = 7.25e−09).

Conclusion

This study found that the addition of the thermal sensor improved the classification accuracy of 24 arm motions from 75 to 93.55% for a single-device system. Our results provide evidence that a single device can be used to classify a relatively large number of arm motions from arm rehabilitation protocols. While this study provides a conceptual proof-of-concept with a healthy population, additional investigation is required to evaluate the performance of this system for specific applications, such as activity classification for physically affected stroke survivors undergoing home-based rehabilitation.

Electronic supplementary material

The online version of this article (10.1186/s12938-019-0677-7) contains supplementary material, which is available to authorized users.

Generative Adversarial Networks for Generation and Classification of Physical Rehabilitation Movement Episodes

This article proposes a method for mathematical modeling of human movements related to patient exercise episodes performed during physical therapy sessions by using artificial neural networks. The generative adversarial network structure is adopted, whereby a discriminative and a generative model are trained concurrently in an adversarial manner. Different network architectures are examined, with the discriminative and generative models structured as deep subnetworks of hidden layers comprised of convolutional or recurrent computational units. The models are validated on a data set of human movements recorded with an optical motion tracker. The results demonstrate an ability of the networks for classification of new instances of motions, and for generation of motion examples that resemble the recorded motion sequences.

A Data Set of Human Body Movements for Physical Rehabilitation Exercises

The article presents University of Idaho - Physical Rehabilitation Movement Data (UI-PRMD) - a publically available data set of movements related to common exercises performed by patients in physical rehabilitation programs. For the data collection, 10 healthy subjects performed 10 repetitions of different physical therapy movements, with a Vicon optical tracker and a Microsoft Kinect sensor used for the motion capturing. The data are in a format that includes positions and angles of full-body joints. The objective of the data set is to provide a basis for mathematical modeling of therapy movements, as well as for establishing performance metrics for evaluation of patient consistency in executing the prescribed rehabilitation exercises.

Metrics for Performance Evaluation of Patient Exercises during Physical Therapy

OBJECTIVE

The article proposes a set of metrics for evaluation of patient performance in physical therapy exercises.

METHODS

Taxonomy is employed that classifies the metrics into quantitative and qualitative categories, based on the level of abstraction of the captured motion sequences. Further, the quantitative metrics are classified into model-less and model-based metrics, in reference to whether the evaluation employs the raw measurements of patient performed motions, or whether the evaluation is based on a mathematical model of the motions. The reviewed metrics include root-mean square distance, Kullback Leibler divergence, log-likelihood, heuristic consistency, Fugl-Meyer Assessment, and similar.

RESULTS

The metrics are evaluated for a set of five human motions captured with a Kinect sensor.

CONCLUSION

The metrics can potentially be integrated into a system that employs machine learning for modelling and assessment of the consistency of patient performance in home-based therapy setting. Automated performance evaluation can overcome the inherent subjectivity in human performed therapy assessment, and it can increase the adherence to prescribed therapy plans, and reduce healthcare costs.

Mathematical Modeling and Evaluation of Human Motions in Physical Therapy Using Mixture Density Neural Networks

OBJECTIVE

The objective of the proposed research is to develop a methodology for modeling and evaluation of human motions, which will potentially benefit patients undertaking a physical rehabilitation therapy (e.g., following a stroke or due to other medical conditions). The ultimate aim is to allow patients to perform home-based rehabilitation exercises using a sensory system for capturing the motions, where an algorithm will retrieve the trajectories of a patient's exercises, will perform data analysis by comparing the performed motions to a reference model of prescribed motions, and will send the analysis results to the patient's physician with recommendations for improvement.

METHODS

The modeling approach employs an artificial neural network, consisting of layers of recurrent neuron units and layers of neuron units for estimating a mixture density function over the spatio-temporal dependencies within the human motion sequences. Input data are sequences of motions related to a prescribed exercise by a physiotherapist to a patient, and recorded with a motion capture system. An autoencoder subnet is employed for reducing the dimensionality of captured sequences of human motions, complemented with a mixture density subnet for probabilistic modeling of the motion data using a mixture of Gaussian distributions.

RESULTS

The proposed neural network architecture produced a model for sets of human motions represented with a mixture of Gaussian density functions. The mean log-likelihood of observed sequences was employed as a performance metric in evaluating the consistency of a subject's performance relative to the reference dataset of motions. A publically available dataset of human motions captured with Microsoft Kinect was used for validation of the proposed method.

CONCLUSION

The article presents a novel approach for modeling and evaluation of human motions with a potential application in home-based physical therapy and rehabilitation. The described approach employs the recent progress in the field of machine learning and neural networks in developing a parametric model of human motions, by exploiting the representational power of these algorithms to encode nonlinear input-output dependencies over long temporal horizons.

Are Virtual Rehabilitation Technologies Feasible Models to Scale an Evidence-Based Fall Prevention Program? A Pilot Study Using the Kinect Camera

Background

Falls in older adults are a significant public health issue. Interventions have been developed and proven effective to reduce falls in older adults, but these programs typically last several months and can be resource intensive. Virtual rehabilitation technologies may offer a solution to bring these programs to scale. Off-the-shelf and custom exergames have demonstrated to be a feasible adjunct to rehabilitation with older adults. However, it is not known if older adults will be able or willing to use a virtual rehabilitation technology to participate in an evidence-based fall prevention program. To have the greatest impact, virtual rehabilitation technologies need to be acceptable to older adults from different backgrounds and level of fall risk. If these technologies prove to be a feasible option, they offer a new distribution channel to disseminate fall prevention programs.

Objective

Stand Tall (ST) is a virtual translation of the Otago Exercise Program (OEP), an evidence-based fall prevention program. Stand Tall was developed using the Virtual Exercise Rehabilitation Assistant (VERA) software, which uses a Kinect camera and a laptop to deliver physical therapy exercise programs. Our purpose in this pilot study was to explore if ST could be a feasible platform to deliver the OEP to older adults from a variety of fall risk levels, education backgrounds, and self-described level of computer expertise.

Methods

Adults age 60 and over were recruited to participate in a one-time usability study. The study included orientation to the program, navigation to exercises, and completion of a series of strength and balance exercises. Quantitative analysis described participants and the user experience.

Results

A diverse group of individuals participated in the study. Twenty-one potential participants (14 women, 7 men) met the inclusion criteria. The mean age was 69.2 (± 5.8) years, 38% had a high school education, 24% had a graduate degree, and 66% classified as “at risk for falls”. Eighteen participants agreed they would like to use ST to help improve their balance, and 17 agreed or strongly agreed they would feel confident using the system in either the senior center or the home. Thirteen participants felt confident they could actually set up the system in their home. The mean System Usability Scale (SUS) score was 65.5 ± 21.2 with a range of 32.5 to 97.5. Ten participants scored ST as an above average usability experience compared to other technologies and 5 participants scored a less than optimal experience. Exploratory analysis revealed no significant relationships between user experience, education background, self-described computer experience, and fall risk.

Conclusions

Results support the virtual delivery of the OEP by a Kinect camera and an avatar may be acceptable to older adults from a variety of backgrounds. Virtual technologies, like Stand Tall, could offer an efficient and effective approach to bring evidence-based fall prevention programs to scale to address the problem of falls and fall-related injuries. Next steps include determining if similar or better outcomes are achieved by older adults using the virtual OEP, Stand Tall, compared to the standard of care.