Search

Joint absence in people with upper-limb-difference leads to compensatory motions. Such compensation has long been a topic of study, but typically only for a single object/user layout, which is unlikely to spatially generalize. We seek to understand how motion varies over a planar workspace for different target orientations and wrist mobility conditions. We therefore present a study that records arm and torso pose during grasping of 49 equally spaced cylindrical targets. Furthermore, we seek to validate the research practice of using wrist-immobilizing bypass sockets on able-bodied participants to simulate prostheses without wrists. Participants were 2 transradial amputees and 7 able-bodied individuals who conducted the study with and without wrist braces, generating 2450 trajectories. Heat-maps illustrate variation over the workspace in Mean Joint Angle, Range of Joint Motion and Distance Travelled by Body Segment. Results indicate that greater wrist restriction primarily exacerbated shoulder internal rotation and elbow flexion, not the trunk. We observed that bypass sockets do not fully simulate amputee behavior. Furthermore, amputee reaching with their intact limb is different to the reaching motion of normative participants, implying that transradial limb-difference affects both sides of the body. Differences in participant behavior were also observed between horizontal and vertical target orientations.

Motion style transfer, which aims to transfer the style from a source motion to the target while keeping its content, has recently gained considerable attention. Some existing works have shown promising results but required labeled data for supervised training, limiting their applicability. In this paper, we present a novel self-supervised learning method for motion style transfer. Specifically, we cast the problem into a contrastive learning framework, which disentangles the human motion representation into a content code and a style code, and the result can be generated by compositing the style code of source motion and the content code of target motion. To encourage better code disentanglement and composition, we investigate InfoNCE loss and Triplet loss in a self-supervised manner. This framework aims at generating reasonable motions while guaranteeing the disentanglement of the latent codes. Comprehensive experiments have been conducted over the benchmark datasets and demonstrated our superior performance over state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Gait monitoring using hip joint angles offers a promising approach for person identification, leveraging the capabilities of smartphone inertial measurement units (IMUs). This study investigates the use of smartphone IMUs to extract hip joint angles for distinguishing individuals based on their gait patterns. The data were collected from 10 healthy subjects (8 males, 2 females) walking on a treadmill at 4 km/h for 10 min. A sensor fusion technique that combined accelerometer, gyroscope, and magnetometer data was used to derive meaningful hip joint angles. We employed various machine learning algorithms within the WEKA environment to classify subjects based on their hip joint pattern and achieved a classification accuracy of 88.9%. Our findings demonstrate the feasibility of using hip joint angles for person identification, providing a baseline for future research in gait analysis for biometric applications. This work underscores the potential of smartphone-based gait analysis in personal identification systems. [ABSTRACT FROM AUTHOR]

Kumite is a karate sparring competition in which two players face off and perform offensive and defensive techniques. Depending on the players, there may be preliminary actions (hereinafter referred to as "pre-actions"), such as pulling the arms or legs, lowering the shoulders, etc., just before a technique is performed. Since the presence of a pre-action allows the opponent to know the timing of the technique, it is important to reduce pre-actions in order to improve the kumite. However, it is difficult for beginners and intermediate players to accurately identify their pre-actions and to improve them through practice. Therefore, this study aims to construct a practice support system that enables beginners and intermediate players to understand their pre-actions. In this paper, we focus on the forefist punch, one of kumite's punching techniques. We propose a method to estimate the presence or absence of a pre-action based on the similarity between the acceleration data of an arbitrary forefist punch and a previously prepared dataset consisting of acceleration data of the forefist punch without a pre-action. We found that the proposed method can estimate the presence or absence of a pre-action in an arbitrary forefist punch with an accuracy of 86%. We also developed KARATECH as a system to support the practice of reducing pre-actions using the proposed method. KARATECH shows the presence or absence of pre-actions through videos and graphs. The evaluation results confirmed that the group using KARATECH had a lower pre-action rate. [ABSTRACT FROM AUTHOR]

This study presents a machine learning model for predicting lumbar spine moments using data from low-cost sensors, with the ultimate aim of developing a control strategy for waist-active exoskeleton devices. The limitation of sparse features in low-cost insoles was addressed by leveraging a source model constructed based on data acquired from the high-precision Pedar-X device, employing a transfer learning technique. The model's performance saw significant improvement through a training approach that incorporated high-precision commercial insole data and fine-tuning with low-cost insole data. In comparison to the conventional model, this method resulted in a noteworthy 7% enhancement in performance, achieving an rRMSE of approximately 12% and a correlation coefficient of 0.9 in lumbar joint moment prediction. If the model can demonstrate real-time efficacy and effectiveness across various operations in future applications, it holds substantial potential for deployment as an active exoskeleton device for the waist. [ABSTRACT FROM AUTHOR]

In this paper, we present a dataset that takes 2D and 3D human pose keypoints estimated from images and relates them to the location of 3D anatomical landmarks. The dataset contains 51,051 poses obtained from 71 persons in A-Pose while performing 7 movements (walking, running, squatting, and four types of jumping). These poses were scanned to build a collection of 3D moving textured meshes with anatomical correspondence. Each mesh in that collection was used to obtain the 3D locations of 53 anatomical landmarks, and 48 images were created using virtual cameras with different perspectives. 2D pose keypoints from those images were obtained using the MediaPipe Human Pose Landmarker, and their corresponding 3D keypoints were calculated by linear triangulation.The dataset consists of a folder for each participant containing two Track Row Column (TRC) files and one JSON file for each movement sequence. One TRC file is used to store the 3D data of the triangulated 3D keypoints while the other contains the 3D anatomical landmarks. The JSON file is used to store the 2D keypoints and the calibration parameters of the virtual cameras. The anthropometric characteristics of the participants are annotated in a single CSV file.These data are intended to be used in developments that require the transformation of existing human pose solutions in computer vision into biomechanical applications or simulations. This dataset can also be used in other applications related to training neural networks for human motion analysis and studying their influence on anthropometric characteristics.

Gait monitoring using hip joint angles offers a promising approach for person identification, leveraging the capabilities of smartphone inertial measurement units (IMUs). This study investigates the use of smartphone IMUs to extract hip joint angles for distinguishing individuals based on their gait patterns. The data were collected from 10 healthy subjects (8 males, 2 females) walking on a treadmill at 4 km/h for 10 min. A sensor fusion technique that combined accelerometer, gyroscope, and magnetometer data was used to derive meaningful hip joint angles. We employed various machine learning algorithms within the WEKA environment to classify subjects based on their hip joint pattern and achieved a classification accuracy of 88.9%. Our findings demonstrate the feasibility of using hip joint angles for person identification, providing a baseline for future research in gait analysis for biometric applications. This work underscores the potential of smartphone-based gait analysis in personal identification systems.

Kumite is a karate sparring competition in which two players face off and perform offensive and defensive techniques. Depending on the players, there may be preliminary actions (hereinafter referred to as “pre-actions”), such as pulling the arms or legs, lowering the shoulders, etc., just before a technique is performed. Since the presence of a pre-action allows the opponent to know the timing of the technique, it is important to reduce pre-actions in order to improve the kumite. However, it is difficult for beginners and intermediate players to accurately identify their pre-actions and to improve them through practice. Therefore, this study aims to construct a practice support system that enables beginners and intermediate players to understand their pre-actions. In this paper, we focus on the forefist punch, one of kumite’s punching techniques. We propose a method to estimate the presence or absence of a pre-action based on the similarity between the acceleration data of an arbitrary forefist punch and a previously prepared dataset consisting of acceleration data of the forefist punch without a pre-action. We found that the proposed method can estimate the presence or absence of a pre-action in an arbitrary forefist punch with an accuracy of 86%. We also developed KARATECH as a system to support the practice of reducing pre-actions using the proposed method. KARATECH shows the presence or absence of pre-actions through videos and graphs. The evaluation results confirmed that the group using KARATECH had a lower pre-action rate.

Gait analysis is important in a variety of applications such as animation, healthcare, and virtual reality. So far, high-cost experimental setups employing special cameras, markers, and multiple wearable sensors have been used for indoor human pose-tracking and gait-analysis purposes. Since locomotive activities such as walking are rhythmic and exhibit a kinematically constrained motion, fewer wearable sensors can be employed for gait and pose analysis. One of the core parts of gait analysis and pose-tracking is lower-limb-joint angle estimation. Therefore, this study proposes a neural network-based lower-limb-joint angle-estimation method from a single inertial sensor unit. As proof of concept, four different neural-network models were investigated, including bidirectional long short-term memory (BLSTM), convolutional neural network, wavelet neural network, and unidirectional LSTM. Not only could the selected network affect the estimation results, but also the sensor placement. Hence, the waist, thigh, shank, and foot were selected as candidate inertial sensor positions. From these inertial sensors, two sets of lower-limb-joint angles were estimated. One set contains only four sagittal-plane leg-joint angles, while the second includes six sagittal-plane leg-joint angles and two coronal-plane leg-joint angles. After the assessment of different combinations of networks and datasets, the BLSTM network with either shank or thigh inertial datasets performed well for both joint-angle sets. Hence, the shank and thigh parts are the better candidates for a single inertial sensor-based leg-joint estimation. Consequently, a mean absolute error (MAE) of 3.65° and 5.32° for the four-joint-angle set and the eight-joint-angle set were obtained, respectively. Additionally, the actual leg motion was compared to a computer-generated simulation of the predicted leg joints, which proved the possibility of estimating leg-joint angles during walking with a single inertial sensor unit. [ABSTRACT FROM AUTHOR]

This study presents a machine learning model for predicting lumbar spine moments using data from low-cost sensors, with the ultimate aim of developing a control strategy for waist-active exoskeleton devices. The limitation of sparse features in low-cost insoles was addressed by leveraging a source model constructed based on data acquired from the high-precision Pedar-X device, employing a transfer learning technique. The model’s performance saw significant improvement through a training approach that incorporated high-precision commercial insole data and fine-tuning with low-cost insole data. In comparison to the conventional model, this method resulted in a noteworthy 7% enhancement in performance, achieving an rRMSE of approximately 12% and a correlation coefficient of 0.9 in lumbar joint moment prediction. If the model can demonstrate real-time efficacy and effectiveness across various operations in future applications, it holds substantial potential for deployment as an active exoskeleton device for the waist.

Human motion analysis using inertial measurement units (IMUs) has recently been shown to provide accuracy similar to the gold standard, optical motion capture, but at lower costs and while being less restrictive and time-consuming. However, IMU-based motion analysis requires precise knowledge of the orientations in which the sensors are attached to the body segments. This knowledge is commonly obtained via time-consuming and error-prone anatomical calibration based on precisely defined poses or motions. In the present work, we propose a self-calibrating approach for magnetometer-free joint angle tracking that is suitable for joints with two degrees of freedom (DoF), such as the elbow, ankle, and metacarpophalangeal finger joints. The proposed methods exploit kinematic constraints in the angular rates and the relative orientations to simultaneously identify the joint axes and the heading offset. The experimental evaluation shows that the proposed methods are able to estimate plausible and consistent joint axes from just ten seconds of arbitrary elbow joint motion. Comparison with optical motion capture shows that the proposed methods yield joint angles with similar accuracy as a conventional IMU-based method while being much less restrictive. Therefore, the proposed methods improve the practical usability of IMU-based motion tracking in many clinical and biomedical applications. [ABSTRACT FROM AUTHOR]

Understanding the fine-grained temporal structure of human actions and its semantic interpretation is beneficial to many real-world tasks, such as sports movements, rehabilitation exercises, and daily-life activities analysis. Current action segmentation methods mainly rely on deep neural networks to derive feature embedding of actions from motion data, while works on analyzing human actions in fine-granularity are still lacking due to the lack of clear and generic definitions of subactions and related datasets. On the other hand, the motion representations obtained in current action segmentation methods lack semantic or mathematical interpretability that can be used to evaluate action/subaction similarity in quantitative motion analysis. Toward the goal of fine-grained, interpretable, scalable, and efficient action segmentation, we propose a novel unsupervised action segmentation and distributed representation framework based on intuitive motion primitives defined on pose data. Metrics for comprehensive evaluation of the unsupervised fine-grained action segmentation task performance are proposed, and both public and self-constructed datasets are adopted in the experiments. The results show that the proposed method has good performance and generality across different subjects, datasets, and application scenarios. [ABSTRACT FROM AUTHOR]

Although industrial shoulder exoskeletons have undergone rapid advancement, their acceptance by industrial workers is limited owing to the misalignment and interference between the exoskeletal frame and the wearer's body and bulkiness of the frames. Several joint mechanisms have been developed to offset misalignments; however, none of the existing systems can simultaneously alleviate the interference and bulkiness problems. Furthermore, the reduction in the misalignments in terms of forces generated at the human–robot interface has not been experimentally verified. Therefore, in this study, design optimization was performed to address the various factors that limit the use of the existing industrial shoulder exoskeletons. Upper body motions were captured and converted into a target trajectory for the exoskeleton to follow. The optimal prismatic–revolute–revolute joint configuration was derived and used to manufacture a skeletal mock-up, which was used to perform experiments. The misalignments of the optimized configuration in the considered motions were 67% lower than those for the conventional joint configuration. Furthermore, the interaction forces were negligible (1.35 N), with a maximum reduction of 61.8% compared to those of conventional configurations. [ABSTRACT FROM AUTHOR]

Crouch gait is one of the most common gait abnormalities; it is usually caused by cerebral palsy. There are few works related to the modeling of crouch gait kinematics, crouch gait analysis, and visualization in both the workspace and joint space. In this work, we present a quaternion-based method to solve the forward kinematics of the position of the lower limbs during walking. For this purpose, we propose a modified eight-DoF human skeletal model. Using this model, we present a geometric method to calculate the gait inverse kinematics. Both methods are applied for gait analysis over normal, mild, and severe crouch gaits, respectively. A metric-based comparison of workspace and joint space for the three gaits for a gait cycle is conducted. In addition, gait visualization is performed using Autodesk Maya for the three anatomical planes. The obtained results allow us to determine the capabilities of the proposed methods to assess the performance of crouch gaits, using a normal pattern as a reference. Both forward and inverse kinematic methods could ultimately be applied in rehabilitation settings for the diagnosis and treatment of diseases derived from crouch gaits or other types of gait abnormalities. [ABSTRACT FROM AUTHOR]

Human locomotion is an imperative topic to be conversed among researchers. Predicting the human motion using multiple techniques and algorithms has always been a motivating subject matter. For this, different methods have shown the ability of recognizing simple motion patterns. However, predicting the dynamics for complex locomotion patterns is still immature. Therefore, this article proposes unique methods including the calibration-based filter algorithm and kinematic-static patterns identification for predicting those complex activities from fused signals. Different types of signals are extracted from benchmarked datasets and pre-processed using a novel calibration-based filter for inertial signals along with a Bessel filter for physiological signals. Next, sliding overlapped windows are utilized to get motion patterns defined over time. Then, polynomial probability distribution is suggested to decide the motion patterns natures. For features extraction based kinematic-static patterns, time and probability domain features are extracted over physical action dataset (PAD) and growing old together validation (GOTOV) dataset. Further, the features are optimized using quadratic discriminant analysis and orthogonal fuzzy neighborhood discriminant analysis techniques. Manifold regularization algorithms have also been applied to assess the performance of proposed prediction system. For the physical action dataset, we achieved an accuracy rate of 82.50% for patterned signals. While, the GOTOV dataset, we achieved an accuracy rate of 81.90%. As a result, the proposed system outdid when compared to the other state-of-the-art models in literature.

Automatic emotion recognition is active research in analyzing human's emotional state over the past decades. It is still a challenging task in computer vision and artificial intelligence due to its high intra-class variation. The main advantage of emotion recognition is that a person's emotion can be recognized even if he is extreme away from the surveillance monitoring since the camera is far away from the human; it is challenging to identify the emotion with facial expression alone. This scenario works better by adding visual body clues (facial actions, hand posture, body gestures). The body posture can powerfully convey the emotional state of a person in this scenario. This paper analyses the frontal view of human body movements, visual expressions, and body gestures to identify the various emotions. Initially, we extract the motion information of the body gesture using dense optical flow models. Later the high-level motion feature frames are transferred to the pre-trained convolutional neural network (CNN) models to recognize the 17 various emotions in Geneva multimodal emotion portrayals (GEMEP) dataset. In the experimental results, AlexNet exhibits the architecture's effectiveness with an overall accuracy rate of 96.63% for the GEMEP dataset is better than raw frames and 94% for visual geometry group-19 VGG-19, and 93.35% for VGG-16 respectively. This shows that the dense optical flow method performs well using transfer learning for recognizing emotions. [ABSTRACT FROM AUTHOR]

The calculation of the helical axis, a.k.a. screw axis, is a functional technique that was introduced for the characterization of the motion and the stability of a human joint. Examples are its applications in the design of prostheses and its use for evaluating the joint performance in post-operatory follow-up. The typical way of studying the variations in the helical axis is to instantaneously compare it to some reference. The reference is typically assumed as (i) an anatomical or geometrical reference (e.g., the condyle to condyle axis or an anatomical plane); (ii) a functional reference, i.e., some axis calculated in a functional way. Calculating the helical axis means determining its orientation and its position, based on the recorded motion of the joint. This paper reviewed the calculation methods of the helical axis, its clinical applications, and the most relevant findings. The operative equations of the most common procedures were clearly and synthetically illustrated. More in detail, the focus of this review was set on the calculation of (i) the instantaneous helical axis; (ii) the finite helical axis; (iii) the average helical axis; (iv) a functional coordinate system attached to the helical axis; and (v) the analysis of the time variations of helical axis. The calculation of those quantities was implemented in MATLAB and the code was proposed as supplementary material. The calculation of the discussed quantities was demonstrated on a sample dataset. [ABSTRACT FROM AUTHOR]

Methods of studying human motion in computer vision systems can be divided into two types. These are analysis in twodimensional and three-dimensional space. The former uses a single camera image and/ or multiple body sensors. Such an approach leads to a rapid accumulation of error and, consequently, low accuracy of the figure representation. Multiple cameras are usually used in the case of three-dimensional space analysis, while the objects are represented as sets of volumetric elements. Despite the high accuracy of this method, it is associated with high computational complexity and internal network load. The purpose of the paper is to develop a model using a single camera, while approaching threedimensional space analysis methods in terms of accuracy. In this paper a human figure is represented as a skeleton. The skeleton is described by an acyclic connected graph. The general structure of a human figure is analyzed. Fifteen basic points are selected. Physical and logical connections between them were studied and mathematically described. The velocity and spatial characteristics of the points and connections outline the general dynamics of motion. The study describes a model of human motion and gives the option for model construction on the example of a particular image. The developed algorithm for collection and analysis of information estimates relative locations and velocity characteristics of the graph elements. The model can be used for acquisition of information about the reference dynamics of human movements. In case of detecting major differences between the reference and the reality, the behavior is defined as deviant. Thus, the obtained algorithm can be applied in computer vision systems for detection and analysis of human movements.

Gait analysis is important in a variety of applications such as animation, healthcare, and virtual reality. So far, high-cost experimental setups employing special cameras, markers, and multiple wearable sensors have been used for indoor human pose-tracking and gait-analysis purposes. Since locomotive activities such as walking are rhythmic and exhibit a kinematically constrained motion, fewer wearable sensors can be employed for gait and pose analysis. One of the core parts of gait analysis and pose-tracking is lower-limb-joint angle estimation. Therefore, this study proposes a neural network-based lower-limb-joint angle-estimation method from a single inertial sensor unit. As proof of concept, four different neural-network models were investigated, including bidirectional long short-term memory (BLSTM), convolutional neural network, wavelet neural network, and unidirectional LSTM. Not only could the selected network affect the estimation results, but also the sensor placement. Hence, the waist, thigh, shank, and foot were selected as candidate inertial sensor positions. From these inertial sensors, two sets of lower-limb-joint angles were estimated. One set contains only four sagittal-plane leg-joint angles, while the second includes six sagittal-plane leg-joint angles and two coronal-plane leg-joint angles. After the assessment of different combinations of networks and datasets, the BLSTM network with either shank or thigh inertial datasets performed well for both joint-angle sets. Hence, the shank and thigh parts are the better candidates for a single inertial sensor-based leg-joint estimation. Consequently, a mean absolute error (MAE) of 3.65° and 5.32° for the four-joint-angle set and the eight-joint-angle set were obtained, respectively. Additionally, the actual leg motion was compared to a computer-generated simulation of the predicted leg joints, which proved the possibility of estimating leg-joint angles during walking with a single inertial sensor unit.

Abnormal movement of the head and neck is a typical symptom of Cervical Dystonia (CD). Accurate scoring on the severity scale is of great significance for treatment planning. The traditional scoring method is to use a protractor or contact sensors to calculate the angle of the movement, but this method is time-consuming, and it will interfere with the movement of the patient. In the recent outbreak of the coronavirus disease, the need for remote diagnosis and treatment of CD has become extremely urgent for clinical practice. To solve these problems, we propose a multi-view vision based CD severity scale scoring method, which detects the keypoint positions of the patient from the frontal and lateral images, and finally scores the severity scale by calculating head and neck motion angles. We compared the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) subscale scores calculated by our vision based method with the scores calculated by a neurologist trained in dyskinesia. An analysis of the correlation coefficient was then conducted. Intra-class correlation (ICC)(3,1) was used to measure absolute accuracy. Our multi-view vision based CD severity scale scoring method demonstrated sufficient validity and reliability. This low-cost and contactless method provides a new potential tool for remote diagnosis and treatment of CD. [ABSTRACT FROM AUTHOR]

Humans learn movements naturally, but it takes a lot of time and training to achieve expert performance in motor skills. In this review, we show how modern technologies can support people in learning new motor skills. First, we introduce important concepts in motor control, motor learning and motor skill learning. We also give an overview about the rapid expansion of machine learning algorithms and sensor technologies for human motion analysis. The integration between motor learning principles, machine learning algorithms and recent sensor technologies has the potential to develop AI-guided assistance systems for motor skill training. We give our perspective on this integration of different fields to transition from motor learning research in laboratory settings to real world environments and real world motor tasks and propose a stepwise approach to facilitate this transition. [ABSTRACT FROM AUTHOR]

Despite its applications in automatic video editing and automatic music composition, the problem of music recommendation from dance motions has seldom been explored. In order to solve this problem, this work proposes a deep music recommendation algorithm based on dance motion analysis and evaluate it through quantitative measures. For quantitative evaluation, this work implements a LSTM-AE based music recommendation method which learns the correspondences between motion and music. In experiments, the two methodologies are compared and the motion analysis based methods outperform their rival by large margins. This work also proposes a quantitative measure of accurately recommended music genre. The proposed motion analysis based method achieves a recommendation accuracy of 91.3% using late fusion of joint and limb features.

We consider the problem of implementing a running platform control system for organizing a comfortable motion of a person in virtual reality with the possibility of adapting the speed of the platform depending on the nature of the user's motion; this will increase the comfort and degree of immersion in the virtual environment. The article uses a controlled running platform (implemented on the basis of a unidirectional treadmill) and a human positioning system based on virtual reality sensors (trackers), which ensures sufficient frequency and accuracy of human positioning. Software has been developed for the platform that can collect data on the user's motion and compare various control algorithms. A set of platform control functions, including those developed in the framework of this study, were formalized, compared, and evaluated. In the course of experimental studies, the optimal control functions were determined for the considered running platform. The advantages and disadvantages of each function are revealed. The scientific novelty of the study lies in the implementation of new running platform control functions to increase its adaptability to user actions by reducing the average deviation of a person from the starting position and hence the negative effects of oscillation and inertia. The practical significance lies in the improvement of the process of pairing virtual reality with devices simulating human motion (in the framework of this study, a unidirectional treadmill) and the development of appropriate software. The results obtained can be adapted to omnidirectional running platforms. [ABSTRACT FROM AUTHOR]

Functional methods usually allow for a flexible and accurate representation of joint kinematics and are increasingly implemented both for clinical and biomechanics research purposes. This paper presents a quantitative comparison between two widely adopted methods for functional axis estimation, that is, the helical axis theory and the symmetrical axis of rotation approach (SARA). To this purpose, a multi-body model was developed to simulate the lower limb of a subject. This model was designed to reproduce different motion patterns, that is, by selecting the active degrees of freedom of the simulated ankle joint. Thanks to virtual markers attached to each segment, the multi-body model was used to generate simulated motion capture data that were then analyzed by instantaneous helical axes and SARA algorithms. To achieve a synthetic representation of joint kinematics, a mean helical axis and an average SARA functional axis were estimated, along with dispersion parameters and rms distance data that were used to quantitatively assess the performance of each method. The sensitivity of each algorithm to different combinations of range and speed of motion, scattering of marker clusters, sampling rate, and additive noise on markers' trajectories, was finally evaluated. [ABSTRACT FROM AUTHOR]

This work describes a novel methodology for creating exergames on an edge-native platform with the integration of multiple deep neural networks. A prototype of the platform, which includes capabilities for innovative gameplay and advanced user interactivity, has been implemented and deployed in a real-world scenario. At core of the proposed methodology is the ad hoc training of classifiers for posture classification which can be dynamically adapted to the specific requirements of the usage scenario, operational and environmental conditions allowing for real-time identification of events and advanced game control. The proposed solution is ideal for individual consumers in a home environment since is supports by-design edge platforms minimizing the cost of the system and enabling in parallel the communication with state-of-the-art hardware (i.e., GPUs, TPUs, computer boards) for real-time operation. The proposed system allows the collection and analysis of game data, which can be exploited by specialized personnel in rehabilitation centers or for other purposes in the areas of healthcare and assisted living. [ABSTRACT FROM AUTHOR]

Data representation learning is one of the most important problems in machine learning. Unsupervised representation learning becomes meritorious as it has no necessity of label information with observed data. Due to the highly time-consuming learning of deep-learning models, there are many machine-learning models directly adapting well-trained deep models that are obtained in a supervised and end-to-end manner as feature abstractors to distinct problems. However, it is obvious that different machine-learning tasks require disparate representation of original input data. Taking human action recognition as an example, it is well known that human actions in a video sequence are 3-D signals containing both visual appearance and motion dynamics of humans and objects. Therefore, the data representation approaches with the capabilities to capture both spatial and temporal correlations in videos are meaningful. Most of the existing human motion recognition models build classifiers based on deep-learning structures such as deep convolutional networks. These models require a large quantity of training videos with annotations. Meanwhile, these supervised models cannot recognize samples from the distinct dataset without retraining. In this article, we propose a new 3-D deconvolutional network (3DDN) for representation learning of high-dimensional video data, in which the high-level features are obtained through the optimization approach. The proposed 3DDN decomposes the video frames into spatiotemporal features under a sparse constraint in an unsupervised way. In addition, it also can be regarded as a building block to develop deep architectures by stacking. The high-level representation of input sequential data can be used in multiple downstream machine-learning tasks, we evaluate the proposed 3DDN and its deep models in human action recognition. The experimental results from three datasets: 1) KTH data; 2) HMDB-51; and 3) UCF-101, demonstrate that the proposed 3DDN is an alternative approach to feedforward convolutional neural networks (CNNs), that attains comparable results. [ABSTRACT FROM AUTHOR]

Abstract Background The past decade has seen the emergence of rehabilitation treatments using virtual reality. One of the advantages in using this technology is the potential to create positive motivation, by means of engaging environments and tasks shaped in the form of serious games. The aim of this study is to determine the efficacy of immersive Virtual Environments and weaRable hAptic devices (VERA) for rehabilitation of upper limb in children with Cerebral Palsy (CP) and Developmental Dyspraxia (DD). Methods A two period cross-over design was adopted for determining the differences between the proposed therapy and a conventional treatment. Eight children were randomized into two groups: one group received the VERA treatment in the first period and the manual therapy in the second period, and viceversa for the other group. Children were assessed at the beginning and the end of each period through both the Nine Hole Peg Test (9-HPT, primary outcome) and Kinesiological Measurements obtained during the performing of similar tasks in a real setting scenario (secondary outcomes). Results All subjects, not depending from which group they come from, significantly improved in both the performance of the 9-HPT and in the parameters of the kinesiological measurements (movement error and smoothness). No statistically significant differences have been found between the two groups. Conclusions These findings suggest that immersive VE and wearable haptic devices is a viable alternative to conventional therapy for improving upper extremity function in children with neuromotor impairments. Trial registration ClinicalTrials, NCT03353623. Registered 27 November 2017-Retrospectively registered, https://clinicaltrials.gov/ct2/show/NCT03353623

Accurately measuring the lower extremities and L5/S1 moments is important since L5/S1 moments are the principal parameters that measure the risk of musculoskeletal diseases during lifting. In this study, protocol that predicts lower extremities and L5/S1 moments with an insole sensor was proposed to replace the prior methods that have spatial constraints. The protocol is hierarchically composed of a classification model and a regression model to predict joint moments. Additionally, a single LSTM model was developed to compare with proposed protocol. To optimize hyperparameters of the machine learning model and input feature, Bayesian optimization method was adopted. As a result, the proposed protocol showed a relative root mean square error (rRMSE) of 8.06~13.88% while the single LSTM showed 9.30~18.66% rRMSE. This protocol in this research is expected to be a starting point for developing a system for estimating the lower extremity and L5/S1 moment during lifting that can replace the complex prior method and adopted to workplace environments. This novel study has the potential to precisely design a feedback iterative control system of an exoskeleton for the appropriate generation of an actuator torque. [ABSTRACT FROM AUTHOR]

Walking is a central activity of daily life, and there is an increasing demand for objective measurement-based gait assessment. In contrast to stationary systems, wearable inertial measurement units (IMUs) have the potential to enable non-restrictive and accurate gait assessment in daily life. We propose a set of algorithms that uses the measurements of two foot-worn IMUs to determine major spatiotemporal gait parameters that are essential for clinical gait assessment: durations of five gait phases for each side as well as stride length, walking speed, and cadence. Compared to many existing methods, the proposed algorithms neither require magnetometers nor a precise mounting of the sensor or dedicated calibration movements. They are therefore suitable for unsupervised use by non-experts in indoor as well as outdoor environments. While previously proposed methods are rarely validated in pathological gait, we evaluate the accuracy of the proposed algorithms on a very broad dataset consisting of 215 trials and three different subject groups walking on a treadmill: healthy subjects (n = 39), walking at three different speeds, as well as orthopedic (n = 62) and neurological (n = 36) patients, walking at a self-selected speed. The results show a very strong correlation of all gait parameters (Pearson's r between 0.83 and 0.99, p < 0.01) between the IMU system and the reference system. The mean absolute difference (MAD) is 1.4 % for the gait phase durations, 1.7 cm for the stride length, 0.04 km/h for the walking speed, and 0.7 steps/min for the cadence. We show that the proposed methods achieve high accuracy not only for a large range of walking speeds but also in pathological gait as it occurs in orthopedic and neurological diseases. In contrast to all previous research, we present calibration-free methods for the estimation of gait phases and spatiotemporal parameters and validate them in a large number of patients with different pathologies. The proposed methods lay the foundation for ubiquitous unsupervised gait assessment in daily-life environments.

Although industrial shoulder exoskeletons have undergone rapid advancement, their acceptance by industrial workers is limited owing to the misalignment and interference between the exoskeletal frame and the wearer’s body and bulkiness of the frames. Several joint mechanisms have been developed to offset misalignments; however, none of the existing systems can simultaneously alleviate the interference and bulkiness problems. Furthermore, the reduction in the misalignments in terms of forces generated at the human–robot interface has not been experimentally verified. Therefore, in this study, design optimization was performed to address the various factors that limit the use of the existing industrial shoulder exoskeletons. Upper body motions were captured and converted into a target trajectory for the exoskeleton to follow. The optimal prismatic–revolute–revolute joint configuration was derived and used to manufacture a skeletal mock-up, which was used to perform experiments. The misalignments of the optimized configuration in the considered motions were 67% lower than those for the conventional joint configuration. Furthermore, the interaction forces were negligible (1.35 N), with a maximum reduction of 61.8% compared to those of conventional configurations.

Human motion analysis using inertial measurement units (IMUs) has recently been shown to provide accuracy similar to the gold standard, optical motion capture, but at lower costs and while being less restrictive and time-consuming. However, IMU-based motion analysis requires precise knowledge of the orientations in which the sensors are attached to the body segments. This knowledge is commonly obtained via time-consuming and error-prone anatomical calibration based on precisely defined poses or motions. In the present work, we propose a self-calibrating approach for magnetometer-free joint angle tracking that is suitable for joints with two degrees of freedom (DoF), such as the elbow, ankle, and metacarpophalangeal finger joints. The proposed methods exploit kinematic constraints in the angular rates and the relative orientations to simultaneously identify the joint axes and the heading offset. The experimental evaluation shows that the proposed methods are able to estimate plausible and consistent joint axes from just ten seconds of arbitrary elbow joint motion. Comparison with optical motion capture shows that the proposed methods yield joint angles with similar accuracy as a conventional IMU-based method while being much less restrictive. Therefore, the proposed methods improve the practical usability of IMU-based motion tracking in many clinical and biomedical applications.

Crouch gait is one of the most common gait abnormalities; it is usually caused by cerebral palsy. There are few works related to the modeling of crouch gait kinematics, crouch gait analysis, and visualization in both the workspace and joint space. In this work, we present a quaternion-based method to solve the forward kinematics of the position of the lower limbs during walking. For this purpose, we propose a modified eight-DoF human skeletal model. Using this model, we present a geometric method to calculate the gait inverse kinematics. Both methods are applied for gait analysis over normal, mild, and severe crouch gaits, respectively. A metric-based comparison of workspace and joint space for the three gaits for a gait cycle is conducted. In addition, gait visualization is performed using Autodesk Maya for the three anatomical planes. The obtained results allow us to determine the capabilities of the proposed methods to assess the performance of crouch gaits, using a normal pattern as a reference. Both forward and inverse kinematic methods could ultimately be applied in rehabilitation settings for the diagnosis and treatment of diseases derived from crouch gaits or other types of gait abnormalities.

In medical tasks such as human motion analysis, computer-aided auxiliary systems have become the preferred choice for human experts for their high efficiency. However, conventional approaches are typically based on user-defined features such as movement onset times, peak velocities, motion vectors, or frequency domain analyses. Such approaches entail careful data post-processing or specific domain knowledge to achieve a meaningful feature extraction. Besides, they are prone to noise and the manual-defined features could hardly be re-used for other analyses. In this paper, we proposed probabilistic movement primitives (ProMPs), a widely-used approach in robot skill learning, to model human motions. The benefit of ProMPs is that the features are directly learned from the data and ProMPs can capture important features describing the trajectory shape, which can easily be extended to other tasks. Distinct from previous research, where classification tasks are mostly investigated, we applied ProMPs together with a variant of Kullback-Leibler (KL) divergence to quantify the effect of different transcranial current stimulation methods on human motions. We presented an initial result with 10 participants. The results validate ProMPs as a robust and effective feature extractor for human motions.