Kinect Camera Research Articles

Total body weight estimation by 3D camera systems: Potential high-tech solutions for emergency medicine applications? A scoping review.

Weight estimation is required in adult patients when weight-based medication must be administered during emergency care, as measuring weight is often not possible. Inaccurate estimations may lead to inaccurate drug dosing, which may cause patient harm. High-tech 3D camera systems driven by artificial intelligence might be the solution to this problem. The aim of this review was to describe and evaluate the published literature on 3D camera weight estimation methods. A systematic literature search was performed for articles that studied the use of 3D camera systems for weight estimation in adults. Data on the study characteristics, the quality of the studies, the 3D camera methods evaluated, and the accuracy of the systems were extracted and evaluated. A total of 14 studies were included, published from 2012 to 2024. Most studies used Microsoft Kinect cameras, with various analytical approaches to weight estimation. The 3D camera systems often achieved a P10 of 90% (90% of estimates within 10% of actual weight), with all systems exceeding a P10 of 78%. The studies highlighted a significant potential for 3D camera systems to be suitable for use in emergency care. The 3D camera systems offer a promising method for weight estimation in emergency settings, potentially improving drug dosing accuracy and patient safety. Weight estimates were satisfactorily accurate, often exceeding the reported accuracy of existing weight estimation methods. Importantly, 3D camera systems possess characteristics that could make them very appropriate for use during emergency care. Future research should focus on developing and validating this methodology in larger studies with true external and clinical validation.

Holo-Stroke: Assessing for Immersive Stroke Care Through Stroke Hologram Teleportation.

Background: Augmented reality enables the wearer to see both their physical environment and virtual objects. Holograms could allow 3D video of providers to be transmitted to distant sites, allowing patients to interact with virtual providers as if they are in the same physical space. Our aim was to determine if Tele-Stroke augmented with Holo-Stroke, compared with Tele-Stroke alone, could improve satisfaction and perception of immersion for the patient. Methods: Kinect cameras positioned at 90-degree intervals around the hub practitioner were used. Cameras streamed real-time optical video to a unity point-cloud program where the data were stitched together in a 360-degree view. The resultant hologram was positioned in 3D space and was visible through the head-mounted display by the patient. Radiology images were shared in Tele-Stroke and via hologram. Likert satisfaction questions were administered. Wilcoxon signed-rank testing was used. Results: Each of the 30 neurology clinic participants scored both Tele-Stroke and Holo-Stroke. Out of these, 29 patients completed the assessments (1 failure owing to computer reboot). Average age was 52 years, with 53.3% of the patients being female, 70.0% being White, and 13.3% being Hispanic. Likert scale score median "Overall" was 32 Tele-Stroke versus 48 Holo-Stroke (p < 0.00001), "Immersion" was 5 versus 10 (p < 0.00001), "Beneficial Technique" was 6 versus 10 (p < 0.00001), and "Ability to See Images" was 5 versus 10 (p < 0.00001). Discussion: Holo-Stroke 3D holographic Tele-Stroke exams resulted in feasibility, satisfaction, and high perception of immersion for the patient. Patients were enthusiastic for the more immersive, personal discussion with their provider and a robust way to experience radiology images. Though further assessments are needed, Holo-Stroke can help the provider "be there, not just see there!"

Research on Depression Recognition Based on University Students’ Facial Expressions and Actions with the Assistance of Artificial Intelligence

As artificial intelligence (AI) technology advances, its application in the field of psychology has witnessed significant advancements. In this paper, with the assistance of AI, 80 university students with depression and 80 university students with normal psychology were selected as the subjects. The facial expression feature data were extracted through OpenFace, and the action feature data were extracted based on a Kinect camera. Then, the convolutional neural network-long short-term memory (CNN-LSTM) and temporal convolutional neural network (TCN) approaches were designed for recognition. Finally, a weighted fusion recognition method was proposed. The results showed that compared with the support vector machine, back-propagation neural network, and other approaches, the CNN-LSTM and TCN methods showed better performance in the recognition of single feature data, and the accuracy reached 0.781 and 0.769, respectively. After weighted fusion, the accuracy reached the highest at 0.875. The results verify that the methods designed in this paper are effective in identifying depressive emotions through facial expressions and actions among university students and have the potential for practical application.

#2072 Impact of the GoodRENal platform on lower-limbs muscle strength in a Spanish sample

Abstract Background and Aims People with chronic kidney disease on hemodialysis have low levels of physical activity and low lower-limbs muscle strength, partly due to the sedentary lifestyle, and a big number of hours sitting or in a recumbent position. The GoodRENal project aimed at creating an intradialysis health platform to improve health-related quality of life. The present study focused on the impact of the GoodRENal intradialysis platform on lower-limbs muscle strength. Method People in hemodialysis treatment from Spain were included. Participants were randomized into two groups, the GoodRENal or the usual care group. The participants used the GoodRENal platform for 12 weeks, 3 times a week. Games using non-immersive virtual reality were created aiming at educating on nutritional habits, increasing physical activity during the HD session, improving cognitive skills and increasing psychological well-being. All of them included dual tasks, so that the movement of the legs was used in every game. An Azure Kinect camera was used to capture the movement and all activities were dual, so that movement of lower limbs was promoted in every game. A handheld dynamometer was used to assess hip flexors and abductors muscle strength during the hemodialysis session. This procedure has proven to be reliable and save. An ANOVA mixed model (group and time factors) was used to analyze the results. Results 19 participants were included (Control n = 9; GoodRENal n = 10). Participants’ mean age was 73.4 years for the control and 68.3 years for the GoodRENal group, with a hemodialysis vintage, median 34 months (minimum 3- maximum 277) and 25 months (minimum 5- maximum 48) for the control and GoodRENal groups. Right hip flexor muscles changed in the control group over time: Baseline 124.7 (23.2) N; 4 weeks 121.3 (20.3) N; 12 weeks 119.0 (21.1) N. In the GoodRENal group, the data were: Baseline 116.7 (34.7) N; 4 weeks 124.7 (36.7) N; 12 weeks 142.7 (44.7) N. The ANOVA showed a significant time × group interaction (F = 11.347 p &amp;lt; 0.001) with a significant improvement in the GoodRENal group. Left hip abductor muscles changed in the control group over time: Baseline 124.5 (24.0) N; 4 weeks 123.0 (26.1) N; 12 weeks 122.0 (24.7) N. In the GoodRENal group, the data were: Baseline 111.1 (30.4) N; 4 weeks 119.2 (33.0) N; 12 weeks 138.5 (36.3) N. The ANOVA showed a significant time × group interaction (F = 15.218 p &amp;lt; 0.001) with a significant improvement in the GoodRENal group. Both left hip flexors (F = 19.055 p &amp;lt; 0.001) and abductors (F = 15.036 p &amp;lt; 0.001) showed a significant time per group interaction, with an improvement in the GoodRENal group. Conclusion The GoodRENal platform is a feasible strategy to increase the muscle strength of the hip flexors and abductors muscles. Future studies with bigger samples should confirm these results.

#747 The digital intradialytic health intervention GoodRENal: feasibility and participants’ opinion

Abstract Background and Aims People with chronic kidney disease on hemodialysis present many comorbidities, including malnutrition and frailty. These factors contribute to a sedentary lifestyle, cognitive impairment, and decreased emotional well-being. Intradialytic exercise with non-immersive virtual reality may improve health-related quality of life and functional capacity in these individuals. The GoodRENal project was developed to create a digital intradialytic health platform to improve health-related quality of life. This study aimed to test the feasibility of the GoodRENal intradialytic platform and to identify participants’ opinions about the use of the platform. Method People on hemodialysis treatment from Spain, Greece and Sweden were included. Non-immersive virtual reality games were created to educate on food intake aiming to better control potassium, phosphate, sodium and liquids. It also aimed at increasing physical activity during HD sessions, improving cognitive skills and enhancing psychological well-being. An Azure Kinect camera was used to capture the movement and all activities were designed as dual-tasks, so the movement of lower limbs was promoted in every game. During a trial session, patients were asked about their perception of the platform through a questionnaire designed to evaluate their opinion. It consisted of 10 general questions, 6 questions for the Physical Activity games, 4 for the Nutrition games, 6 for the Cognition games and 7 for the Psychology games. Every question was rated with a Likert scale. Answers are presented in three categories ‘agree’ (4 or 5 out of 5), ‘neither agree nor disagree’ (3 out of 5), and ‘disagree’ (1 or 2 out of 5). Results 27 participants were included in this study (Spain n = 9; Greece n = 10; Sweden n = 8). Participants’ mean age was 64.6 years, 9 women, with a hemodialysis vintage, median 27 months (minimum 4- maximum 240). The questions with a higher degree of agreement or disagreement are summarized. In terms of general feedback on the GoodRENal platform, more than 90% of respondents found the instructions easy to follow and that playing the games was a pleasant break during the hemodialysis session. Furthermore, they disagreed with the statement that the back was tiring. The questions related to the physical activity games with the highest rate of agreement related to the ideas that the program provides a good workout (80% agreement), the possibility to have more exercises (70% agreement) and that it was fun (60% agreement). Regarding the nutrition games, more than 70% agreed that they had gained new knowledge about what to eat and about 60% felt motivated to modify what they ate. For the cognition games, more than 80% maintained attention during the games, but only 35% felt motivated to engage in cognitively stimulating activities. Finally, regarding the games aimed at emotional well-being, 30% of the patients indicated that they felt good about the exercises. Conclusion The GoodRENal platform is a feasible strategy to increase physical activity during hemodialysis treatment, as well as to improve nutritional information and cognitive status of patients. As a general impression the platform was easy to follow and it was a nice break. Future studies should take into account participants’ opinion to improve the content of a digital intradialytic health platform.

Association between gait video information and general cardiovascular diseases: a prospective cross-sectional study.

Cardiovascular disease (CVD) may not be detected in time with conventional clinical approaches. Abnormal gait patterns have been associated with pathological conditions and can be monitored continuously by gait video. We aim to test the association between non-contact, video-based gait information and general CVD status. Individuals undergoing confirmatory CVD evaluation were included in a prospective, cross-sectional study. Gait videos were recorded with a Kinect camera. Gait features were extracted from gait videos to correlate with the composite and individual components of CVD, including coronary artery disease, peripheral artery disease, heart failure, and cerebrovascular events. The incremental value of incorporating gait information with traditional CVD clinical variables was also evaluated. Three hundred fifty-two participants were included in the final analysis [mean (standard deviation) age, 59.4 (9.8) years; 25.3% were female]. Compared with the baseline clinical variable model [area under the receiver operating curve (AUC) 0.717, (0.690-0.743)], the gait feature model demonstrated statistically better performance [AUC 0.753, (0.726-0.780)] in predicting the composite CVD, with further incremental value when incorporated with the clinical variables [AUC 0.764, (0.741-0.786)]. Notably, gait features exhibited varied association with different CVD component conditions, especially for peripheral artery disease [AUC 0.752, (0.728-0.775)] and heart failure [0.733, (0.707-0.758)]. Additional analyses also revealed association of gait information with CVD risk factors and the established CVD risk score. We demonstrated the association and predictive value of non-contact, video-based gait information for general CVD status. Further studies for gait video-based daily living CVD monitoring are promising.

The development of an automated assessment system for resistance training movement

ABSTRACT Portable data collection devices and machine learning (ML) have been combined in autonomous movement analysis models for resistance training (RT) movements. However, input features for these models were mostly extracted empirically and subsequent models demonstrated limited interpretability and generalisability to real-world settings. This study aimed to investigate the utility of interpretable and generalisable modelling techniques and several data-driven feature extraction (FE) methods. This was achieved by developing machine learning movement analysis models for the barbell back squat and deadlift using markerless motion capture. 61 participants performed submaximal and maximal repetitions of both RT movements. Movement data was collected using two Azure Kinect cameras. Joint and segment kinematic variables were calculated from the collected depth imaging, and input features were extracted using traditional, manual FE methods and novel data-driven techniques. Classifiers were developed for several predefined technical deviations for both movements. Many of the addressed technical deviations could be classified with good levels of accuracy (≥70%) while the remainder were poor (55%–60%). Additionally, data-driven FE techniques were comparable to previous, traditional FE methods. Interpretable and generalisable modelling techniques can be utilised to good effect for certain classification tasks while data-driven FE techniques did not provide a consistent advantage over traditional FE methods.

InterCap: Joint Markerless 3D Tracking of Humans and Objects in Interaction from Multi-view RGB-D Images

Humans constantly interact with objects to accomplish tasks. To understand such interactions, computers need to reconstruct these in 3D from images of whole bodies manipulating objects, e.g., for grasping, moving and using the latter. This involves key challenges, such as occlusion between the body and objects, motion blur, depth ambiguities, and the low image resolution of hands and graspable object parts. To make the problem tractable, the community has followed a divide-and-conquer approach, focusing either only on interacting hands, ignoring the body, or on interacting bodies, ignoring the hands. However, these are only parts of the problem. On the contrary, recent work focuses on the whole problem. The GRAB dataset addresses whole-body interaction with dexterous hands but captures motion via markers and lacks video, while the BEHAVE dataset captures video of body-object interaction but lacks hand detail. We address the limitations of prior work with InterCap, a novel method that reconstructs interacting whole-bodies and objects from multi-view RGB-D data, using the parametric whole-body SMPL-X model and known object meshes. To tackle the above challenges, InterCap uses two key observations: (i) Contact between the body and object can be used to improve the pose estimation of both. (ii) Consumer-level Azure Kinect cameras let us set up a simple and flexible multi-view RGB-D system for reducing occlusions, with spatially calibrated and temporally synchronized cameras. With our InterCap method we capture the InterCap dataset, which contains 10 subjects (5 males and 5 females) interacting with 10 daily objects of various sizes and affordances, including contact with the hands or feet. To this end, we introduce a new data-driven hand motion prior, as well as explore simple ways for automatic contact detection based on 2D and 3D cues. In total, InterCap has 223 RGB-D videos, resulting in 67,357 multi-view frames, each containing 6 RGB-D images, paired with pseudo ground-truth 3D body and object meshes. Our InterCap method and dataset fill an important gap in the literature and support many research directions. Data and code are available at https://intercap.is.tue.mpg.de.

Gait disorder classification based on effective feature selection and unsupervised methodology

In gait stability analysis, patients suffering from dysfunction problems are impacted by shifts in their dynamic balance. Monitoring the patients’ progress is important for allowing physicians and patients to observe the rehabilitation process accurately. In this study, we designed a new methodology for classifying gait disorders to quantify patients’ progress. The dataset in this study includes 84 measurements of 37 patients based on a physician’s opinion. In this study, the system, which includes a Kinect camera to observe and store the frames of patients walking down a hallway, a key-point detector to detect the skeletal key points, and an encoder transformer classifier network integrated with generator–discriminator networks (ET-GD), is designed to evaluate the classification of gait dysfunction. The detector extracts the skeletal key points of patients. After feature engineering, the selected high-level features are fed into the proposed neural network to analyse patient movement and perform the final evaluation of gait dysfunction. The proposed network is inspired by the 1D encoder transformer, which is integrated with two main networks: a network for classification and a network to generate fake output data similar to the input data. Furthermore, we used a discriminator structure to distinguish between the actual data (input) and fake data (generated data). Due to the multi-structural networks in the proposed method, multi-loss functions need to be optimised; this increases the accuracy of the encoder transformer classifier.

Comparison of CNN-based methods for yoga pose classification

Yoga is an exercise developed in ancient India. People perform yoga in order to have mental, physical, and spiritual benefits. While yoga helps build strength in the mind and body, incorrect postures might result in serious injuries. Therefore, yoga exercisers need either an expert or a platform to receive feedback on their performance. Since access to experts is not an option for everyone, a system to provide feedback on the yoga poses is required. To this end, commercial products such as smart yoga mats and smart pants are produced; Kinect cameras, sensors, and wearable devices are used. However, these solutions are either uncomfortable to wear or not affordable for everyone. Nonetheless, a system that employs computer vision techniques is a requirement. In this paper, we propose a deep-learning model for yoga pose classification, which is the first step of a quality assessment and personalized feedback system. We introduce a wavelet-based model that first takes wavelet transform of input images. The acquired subbands, i.e., approximation, horizontal, vertical, and diagonal coefficients of the wavelet transform are then fed into separate convolutional neural networks (CNN). The obtained probability results for each group are fused to predict the final yoga class. A publicly available dataset with 5 yoga poses is used. Since the number of images in the dataset is not enough for a deep learning model, we also perform data augmentation to increase the number of images. We compare our results to a CNN model and the three models that employ the subbands separately. Results obtained using the proposed model outperforms the accuracy output achieved with the compared models. While the regular CNN model has 61% and 50% accuracy for the training and test data, the proposed model achieves 91% and 80%, respectively.

Development of the RIKEN database for dynamic facial expressions with multiple angles

The development of facial expressions with sensing information is progressing in multidisciplinary fields, such as psychology, affective computing, and cognitive science. Previous facial datasets have not simultaneously dealt with multiple theoretical views of emotion, individualized context, or multi-angle/depth information. We developed a new facial database (RIKEN facial expression database) that includes multiple theoretical views of emotions and expressers’ individualized events with multi-angle and depth information. The RIKEN facial expression database contains recordings of 48 Japanese participants captured using ten Kinect cameras at 25 events. This study identified several valence-related facial patterns and found them consistent with previous research investigating the coherence between facial movements and internal states. This database represents an advancement in developing a new sensing system, conducting psychological experiments, and understanding the complexity of emotional events.

Markerless motion capture technology to quantify functional performance in dementias: a systematic review

AbstractBackgroundQuantifying functional performance objectively and unobtrusively could offer vital insights for clinical decision making on the level of support required by people living with dementia and other neurodegenerative conditions. Markerless motion capture (MMC) is more suitable for deployment in clinical and real‐world settings because single sensor MMC systems are less expensive and less complex compared to marker‐based systems. The aim of this study was to complete a systematic review of published literature on the use of MMC with full‐body tracking in people with dementias.MethodWe systematically searched for relevant articles using the search terms motion capture, motion analysis and movement analysis combined with dementia, mild cognitive impairment (MCI) and Parkinson’s disease (PD) which yielded a total of 1103 results. Inclusion criteria were MMC and full‐body tracking. Studies evaluating marker‐based motion capture, body‐worn sensors or Inertial Measurement Units, virtual reality, kinetics and interventions like DBS or exercises were excluded. 134 studies were included for full article screening out of which 17 eligible studies that met the selection criteria were included in the review.ResultMMC devices used in the reviewed studies included Microsoft Kinect (n = 15) and unspecified 2D motion camera (n = 2). Patient groups included in the studies were PD (n = 12), dementia (n = 4) and MCI (n = 1). Functional performance components evaluated were gait (n = 16), sit to stand (n = 2), turning (n = 1) and stepping in place (n = 1). About half the studies used advanced machine learning methods whereas the other half used descriptive statistical methods. The lack of standardisation in the measures used prevented us from comparing and synthesizing study results.ConclusionAlthough MMC has the potential to be used for quantifying functional performance in people with dementia, MCI and PD, there appears to be significant variation in the methods used in its evaluation. Although features of mobility were measured, none of the reviewed studies analysed activities of daily living tasks, a key component of functional performance. Standardising the process for extracting features and assessing accuracy would enable more effective validation of real‐world impact. Further research is required in people with neurodegenerative conditions for MMC to be routinely used in clinical and real‐world applications.

Movement Digital Biomarkers for Functional and Cognitive Decline in Dementia

AbstractBackgroundDigital biomarkers based on accurate tracking of motor behaviour can provide a cost‐effective, objective, and robust measure for disease progression in dementia, changes in care needs, and the effect of interventions. Most of our brain is involved in the planning and execution of movement. Hence, quantifying and monitoring movement can produce powerful indicators of both motor and cognitive decline. In this project, we develop an intelligent system that computes and tracks movement‐based digital biomarkers of dementia.MethodsThe project is established in a controlled and ecologically valid mock apartment – the Living Lab – situated in the UK DRI Care Research and Technology Centre. We collected patients’ performance in a set of physical tasks and activities of daily living (ADL) with a depth‐based motion capture system, along with neurocognitive measures using Addenbrooke’s Cognitive Examination III and Cognitron. We first conducted temporal and spatial synchronisations of data collected from six Microsoft Azure Kinect cameras to produce high‐precision three‐dimensional joint coordinates that provide dynamic postural representations. These were then used to compute digital kinematic markers. This was followed by an assessment of the efficacy of the biomarkers in characterising people affected by dementia. We examined the association between the biomarkers with motor and cognitive function and overall efficiency in performing ADL.ResultsOur preliminary analysis showed that patients and healthy controls can be distinguished by the distribution of extracted features (an example feature – spinal curvature – is shown in Figure 1) in structured tasks and free behaviour. The variance of the distribution was correlated with performance in clinical assessments.ConclusionOur analysis suggests the feasibility of digital biomarkers in characterising functional limitations and cognitive decline. Successfully validated biomarkers can be integrated into remote health monitoring systems at home, care‐homes, and clinics, with the potential to revolutionise prognostic prediction and functional assessment of dementia.

Remote Control Of A Serial Manipulator Using Depth Camera

In this study, a serial robot arm was manipulated in real time using a depth sensor. Xbox-Kinect sensor is used as depth sensor. Kinect data was transferred to an industrial computer and processed. A three-axis serial robot arm was developed, and experiments were carried out on this developed prototype in real time. The drive of the three-axis robot is provided with RC servo motors and these motors are controlled by Arduino Uno R3 board. To find the joint angles, the image obtained from the Kinect camera has evolved into a skeleton form through the image processing program developed in the ARDUINO-Processing environment. Vector elements are defined on the human limbs of which posture is to be calculated. The angles between the limbs were obtained by applying trigonometric operations to these vectors. The reference angles were sent to the Arduino Uno R3 board, and the servo motors that provide the movement of the robot were controlled according to these angle values and the movement of the system was ensured. As a result of the experiments, it has been observed that the robot arm can imitate the movements made instantly.

A computer vision approach to continuously monitor fatigue during resistance training

Monitoring fatigue during resistance training is essential to avoid injuries caused by overtraining. Fatigue can be comprehensively quantified by the external and internal load, where the external load is the work done by the athlete, and the internal load is the psychological and physiological response to the external load. This paper proposes a computer vision method to continuously monitor fatigue during resistance training by predicting external and internal parameters, namely the generated power and the rating of perceived exertion. We utilize the human pose estimation from two Microsoft Azure Kinect cameras to capture the movement of athletes while performing stationary exercises. Our method processes the obtained kinematic data, computes skeleton features to train traditional machine learning algorithms, and constructs feature maps to train convolutional neural network-based models to predict the load parameters. For evaluation, we recorded a dataset of 16 subjects who performed squat exercises on a Flywheel and rated their perceived exertion after each set. A measuring unit integrated into the Flywheel provided power readings for each repetition. The results show that our method achieves good results in predicting both parameters. Gradient Boosting Regression Trees best predicted perceived exertion with a mean absolute percentage error of 8.08% and a Spearman’s ρ=0.74. Multi-layer Perceptron performed best in predicting power with a mean absolute error of 23.13 Watts and ρ=0.79. Our findings show that our approach delivers promising external and internal load quantifications for fatigue, with great potential to provide external feedback to coaches or athletes.

Design and validation of depth camera-based static posture assessment system

Postural abnormalities have become a prevalent issue affecting individuals of all ages, resulting in a diminished quality of life. Easy-use and reliable posture assessment tools can aid in screening for and correcting posture deviation at an early stage. In this study, we present a depth camera-based static posture assessment system to screen for common postural anomalies such as uneven shoulders, pelvic tilt, bowlegs and knock-knees, forward head, scoliosis, and shoulder blade inclination. The system consists of an Azure Kinect camera, a laptop, and evaluation software. Our system accurately measures skeleton and posture indexes and shows favorable agreement with a golden standard optical infrared motion capture system. The findings indicate that the system is a low-cost posture assessment tool with high precision and accuracy, suitable for initial screening of postural abnormalities in individuals of all ages.

Fall Detection Approaches for Monitoring Elderly HealthCare Using Kinect Technology: A Survey

The severity of falls increases with age and reduced mobility. Falls are a frequent source of domestic accidents and accidental death on the part of fragile people. They produce anatomical injuries, reduce quality of life, cause dramatic psychological effects, and impose heavy financial burdens. A growing elderly population leads to a direct increase in health service costs, and indirectly to a deterioration of social life in the long term. Unsurprisingly, socioeconomic costs have triggered new scientific health research to detect falls in older people. One of the most appropriate solutions for monitoring the elderly and automatically detecting falls is computer vision. The Kinect camera plays a vital role in recognizing and detecting activities while ensuring seniors’ comfort, safety, and privacy preferences in the fall detection system. This research surveys several Kinect-based works in the literature that cover the approaches used in fall detection. In addition, we discuss the public fall benchmark based on Kinect technology. In general, the main objective of this survey is to provide a complete description of the modules making up the fall detectors and thereby guide researchers in developing fall approaches based on Kinect.

A Marker Free Visual-based Home Rehabilitation Framework

Adhesive capsulitis or more commonly known as frozen shoulder, is a familiar occurrence for adults aged above 40 caused by the inflammation of the connective tissues surrounding the shoulder joint. There are different severity of adhesive capsulitis but patients afflicted with frozen shoulder typically will experience stiffness, severe pain, and reduced range of motion (ROM) for the shoulder. No matter the course of treatment being non-steroidal anti-inflammatory drugs (NSAIDs) or steroid injections, which can help reduce the inflammation and reduce pain, in order to restore ROM for the afflicted shoulder joint, rehabilitation exercises need to be performed. Even without the current climate where medical workers are severely overworked, physical therapists are in short order especially for developing countries like Malaysia. A remedy for this situation would be to deploy home rehabilitation instead. This would be a way for patients to get proper rehabilitation exercises in between visits to the clinic to meet the physical therapist. This can also reduce the frequency of in-clinic visits while still allowing the patient to progress in the rehabilitation of their afflicted shoulder joint. Though home rehabilitation seems like a clear solution, it does come with its own set of challenges. How open will the patients themselves be to utilizing a home rehabilitation system? In light of that, this paper proposes a home rehabilitation framework focusing on a marker free visual-based implementation using the Microsoft Kinect camera. The framework will measure the impact of variables such as capability, motivation and opportunity on the adoption rate of the home rehabilitation. Cronbach’s Alpha tests were conducted to ascertain the reliability of the variables used in the framework.

SMART AGRICULTURAL MONITORING SOLUTION FOR CHILLI LEAF DISEASES USING A LOW-COST KINECT CAMERA AND AN IMPROVED CNN ALGORITHM

Chilli is extensively grown all over the globe and is particularly important as a food. One of the most difficult issues confronting chilli cultivation is the requirement for accurate identification of leaf diseases. Leaf diseases have a negative impact on chilli production quality, resulting in significant losses for farmers. Numerous Machine Learning (ML) and Convolution Neural Network (CNN) models have been developed for classifying chilli leaf diseases under uniform background and uncomplicated leaf conditions, with an average classification accuracy achieved. However, a diseased leaf usually grows alongside a cluster of other leaves, making it difficult to classify the disease. It will be easier for farmers if there is a reliable model that can classify a chilli leaf disease in a cluster of leaves. The aim of this study was to propose a model for classifying chilli leaf disease from both a uniform background and a complex cluster of leaves. Images of diseased chilli leaves are acquired using a low-cost Kinect camera, which include discoloration, grey spots, and leaf curling. The different types of chilli leaf disease are then classified using an improved ShuffleNet CNN model. With a classification accuracy of 99.82%, the proposed model outperformed the other existing models.

Student and Faculty Perspectives on the Usefulness and Usability of a Digital Health Educational Tool to Teach Standardized Assessment of Persons After Stroke: Mixed Methods Study.

The VSTEP Examination Suite is a collection of evidence-based standardized assessments for persons after stroke. It was developed by an interdisciplinary team in collaboration with clinician users. It consists of 5 standardized assessments: 2 performance-based tests using the Kinect camera (Microsoft Corp) to collect kinematics (5-Time Sit-to-Stand and 4-Square Test); 2 additional performance-based tests (10-Meter Walk Test and 6-Minute Walk Test); and 1 patient-reported outcome measure, the Activities-Specific Balance Confidence Scale. This study aimed to describe the development of the VSTEP Examination Suite and its evaluation as an educational tool by physical therapy students and faculty to determine its usefulness and usability. A total of 6 students from a Doctor of Physical Therapy program in the United States and 6 faculty members who teach standardized assessments in different physical therapy programs from the United States and Israel were recruited by convenience sampling to participate in the study. They interacted with the system using a talk-aloud procedure either in pairs or individually. The transcripts of the sessions were coded deductively (by 3 investigators) with a priori categories of usability and usefulness, and comments were labeled as negative or positive. The frequencies of the deductive themes of usefulness and usability were tested for differences between faculty and students using a Wilcoxon rank sum test. A second round of inductive coding was performed by 3 investigators guided by theories of technology adoption, clinical reasoning, and education. The faculty members' and students' positive useful comments ranged from 83% (10/12) to 100%. There were no significant differences in usefulness comments between students and faculty. Regarding usability, faculty and students had the lowest frequency of positive comments for the 10-Meter Walk Test (5/10, 50%). Students also reported a high frequency of negative comments on the 4-Square Test (9/21, 43%). Students had a statistically significantly higher number of negative usability comments compared with faculty (W=5.7; P=.02), specifically for the 5-Time Sit-to-Stand (W=5.3; P=.02). Themes emerged related to variable knowledge about the standardized tests, value as a teaching and learning tool, technology being consistent with clinical reasoning in addition to ensuring reliability, expert-to-novice clinical reasoning (students), and usability. The VSTEP Examination Suite was found to be useful by both faculty and students. Reasons for perceived usefulness had some overlap, but there were also differences based on role and experience. Usability testing revealed opportunities for technology refinement. The development of the technology by interdisciplinary teams and testing with multiple types of users may increase adoption.