UNASSIGNED: Exercise is pivotal for maintaining physical health in contemporary society. However, improper postures and movements during exercise can result in sports injuries, underscoring the significance of skeletal motion analysis. This research aims to leverage advanced technologies such as Transformer, Graph Neural Networks (GNNs), and Generative Adversarial Networks (GANs) to optimize sports training and mitigate the risk of injuries.
UNASSIGNED: The study begins by employing a Transformer network to model skeletal motion sequences, facilitating the capture of global correlation information. Subsequently, a Graph Neural Network is utilized to delve into local motion features, enabling a deeper understanding of joint relationships. To enhance the model\'s robustness and adaptability, a Generative Adversarial Network is introduced, utilizing adversarial training to generate more realistic and diverse motion sequences.
UNASSIGNED: In the experimental phase, skeletal motion datasets from various cohorts, including professional athletes and fitness enthusiasts, are utilized for validation. Comparative analysis against traditional methods demonstrates significant enhancements in specificity, accuracy, recall, and F1-score. Notably, specificity increases by ~5%, accuracy reaches around 90%, recall improves to around 91%, and the F1-score exceeds 89%.
UNASSIGNED: The proposed skeletal motion analysis method, leveraging Transformer and Graph Neural Networks, proves successful in optimizing exercise training and preventing injuries. By effectively amalgamating global and local information and integrating Generative Adversarial Networks, the method excels in capturing motion features and enhancing precision and adaptability. Future research endeavors will focus on further advancing this methodology to provide more robust technological support for healthy exercise practices.

UNASSIGNED: In this article, we comprehensively review the current situation and research on technology related to outdoor travel for blind and visually impaired people (BVIP), given the diverse types and incomplete functionality of navigation aids for the blind. This aims to provide a reference for related research in the fields of outdoor travel for BVIP and blind navigation.
UNASSIGNED: We compiled articles related to blind navigation, of which a total of 227 of them are included in the search criteria. One hundred and seventy-nine articles are selected from the initial set, from a technical point of view, to elaborate on five aspects of blind navigation: system equipment, data sources, guidance algorithms, optimization of related methods, and navigation maps.
UNASSIGNED: The wearable form of assistive devices for the blind has the most research, followed by the handheld type of aids. The RGB data class based on vision sensor is the most common source of navigation environment information data. Object detection based on picture data is also particularly rich among navigation algorithms and associated methods, indicating that computer vision technology has become an important study content in the field of blind navigation. However, research on navigation maps is relatively less.
UNASSIGNED: In the study and development of assistive equipment for BVIP, there will be an emphasis on prioritizing attributes, such as lightness, portability, and efficiency. In light of the upcoming driverless era, the research focus will be on the development of visual sensors and computer vision technologies that can aid in navigation for the blind.
The visual deficiency can easily help blind and visually impaired people (BVIP) to develop psychological disorders.There are few, if any, devices to meet the outdoor travel needs of BVIP in all aspects.There is no comprehensive summary and overview in the field of outdoor navigation for the blind.The selection of appropriate assistive devices can help BVIP better understand the information of their surroundings and make safer and more effective outdoor trips.

UNASSIGNED: To investigate the current situation of assistive device (AD) usage among seniors in Tibet, China, and explore its influencing factors.
UNASSIGNED: A mixed-methods design was used. Cognition, attitude, behavior, and preference toward ADs were assessed by a self-designed questionnaire. Additionally, participants responded to the open-ended questions. Anderson\'s behavior model analyzed the impacts of the prerequisite factors, enabling factors, and demand factors influencing the utilization of ADs by Tibetan seniors.
UNASSIGNED: Of the 211 Tibetan seniors, 149 (70.6%) Tibetan seniors expressed the willingness to utilize ADs. Binary Logistic regression analysis showed that one prerequisite factor: age; one enabling factor: care situation, and one demand factor: dysfunctional condition were factors influencing the utilization of ADs. Qualitative comments described: psychological, physical environment, and social support factors were the main influencing factors.
UNASSIGNED: This study presents the current situation to utilize ADs by Tibetan seniors on the Qinghai-Tibet Plateau, incorporates Anderson\'s behavioral model for quantitative analysis, and combines qualitative research to explore the facilitating and hindering factors, to provide reference and basis for the development of ADs for seniors and policy formulation. The sample size of this study is relatively small and limited to ethnic groups, and we plan to increase the sample size and include more ethnic groups in the future study.

Locomotion mode recognition provides the prosthesis control with the information on when to switch between different walking modes, whereas the gait phase detection indicates where we are in the gait cycle. But powered prostheses often implement a different control strategy for each locomotion mode to improve the functionality of the prosthesis. Existing studies employed several classical machine learning methods for locomotion mode recognition. However, these methods were less effective for data with complex decision boundaries and resulted in misclassifications of motion recognition. Deep learning-based methods potentially resolve these limitations as it is a special type of machine learning method with more sophistication. Therefore, this study evaluated three deep learning-based models for locomotion mode recognition, namely recurrent neural network (RNN), long short-term memory (LSTM) neural network, and convolutional neural network (CNN), and compared the recognition performance of deep learning models to the machine learning model with random forest classifier (RFC). The models are trained from data of one inertial measurement unit (IMU) placed on the lower shanks of four able-bodied subjects to perform four walking modes, including level ground walking (LW), standing (ST), and stair ascent/stair descent (SA/SD). The results indicated that CNN and LSTM models outperformed other models, and these models were promising for applying locomotion mode recognition in real-time for robotic prostheses.

OBJECTIVE: The study aimed to gain an insight into the utilisation, self-perceived needs, and attitudes towards and influencing factors of assistive device (AD) usage among community-dwelling older adults in China.
METHODS: This is a cross-sectional study.
METHODS: A total of 5790 elderly people from eight communities within three provinces in China were recruited by convenience sampling. Utilisation, needs and attitudes towards ADs were assessed by a questionnaire designed by the authors. Barthel activities of daily living scale was used to determine disability, whereas cognitive function was assessed with the Mini-Mental State Examination. The impact of participant characteristics, enabling factors and demand factors on the utilisation of ADs were assessed by univariate and multifactor analyses.
RESULTS: The prevalence of AD ownership among participants was 10.9% (n = 634), whereas the self-perceived need for ADs was 46.1% (n = 2670). Most participants had negative attitudes towards ADs, with only 37.6% (n = 2175) of participants believing that ADs were of significant help. Factors influencing the usage of ADs included participant characteristics (age, occupation, living area, education), enabling factors (economic situation, number of children) and demand factors (activities of daily living score, attitudes, self-perceived needs).
CONCLUSIONS: Although ADs for the elderly in China have become more affordable and accessible after a series of reforms, there remains a gap in AD services resulting in low AD utilisation, high self-perceived needs and misconceptions of ADs. Certain factors influencing the use of ADs are more significant than others. The findings from this study will be informative for healthcare providers and decision-makers when designing strategies to achieve universal elderly AD usage.

Even after reconstructive surgery, it is still difficult for patients with severe burns to achieve independent eating activity. In this project, we customized the forearm pronation\'s assistant tableware to assist in improvement with eating activities.
From January 2017 to December 2018, 28 patients with severe burns including the hands were recruited. For the patient\'s independent eating activities, we customized forearm pronation\'s tableware (forks and spoons). We compared modified Barthel index (MBI) and Visual analogue scale (VAS) of satisfaction under three conditions: no auxiliary tableware, ADL universal cuff, or forearm pronation tableware; to compare the duration and the weight of food spilled during lunch when the patients wore the ADL universal cuff or the forearm pronation\'s tableware. Differences in MBI (rank data) were tested by the Friedman test, differences in VAS (normal distribution) were tested with One-way ANOVA (Bonferroni), differences in the duration and the weight (normal distribution data) were tested by paired sample t test.
After wearing the forearm pronation\'s assistant tableware, MBI VAS both increased more than when the patients did not wear the auxiliary tableware (all p＜0.05). When the subjects wore forearm pronation tableware, the duration of lunch significantly decreased and the quality of eating activity significantly improved compared to the ADL universal cuff in eating activity (all p＜0.05).
After wearing the forearm pronation\'s assistant tableware, the patients with severe burns completely or almost completely accomplished independent eating, the duration was decreased, and during eating activity the quality and the satisfaction were improved.
Chinese Clinical trial registry, ChiCTR1800019963.

Fast and accurate gait phase detection is essential to achieve effective powered lower-limb prostheses and exoskeletons. As the versatility but also the complexity of these robotic devices increases, the research on how to make gait detection algorithms more performant and their sensing devices smaller and more wearable gains interest. A functional gait detection algorithm will improve the precision, stability, and safety of prostheses, and other rehabilitation devices. In the past years the state-of-the-art has advanced significantly in terms of sensors, signal processing, and gait detection algorithms. In this review, we investigate studies and developments in the field of gait event detection methods, more precisely applied to prosthetic devices. We compared advantages and limitations between all the proposed methods and extracted the relevant questions and recommendations about gait detection methods for future developments.