Efficacy of Robot-Assisted and Virtual Reality Interventions on Balance, Gait, and Daily Function in Patients With Stroke: A Systematic Review and Network Meta-analysis.

OBJECTIVE: This study aimed to evaluate the comparative effectiveness and ranking of robot-assisted training, virtual reality, and robot-assisted rehabilitation combined with virtual reality in improving balance, gait, and daily function in patients with stroke. DATA SOURCES: PubMed, EMBASE, the Cochrane Library, Physiotherapy Evidence Database, CINAHL, Web of Science, and ProQuest Dissertations and Theses abstracting and indexing databases were comprehensively searched to include randomized controlled trials published through August 31, 2022. STUDY SELECTION: Randomized controlled trials comparing robot-assisted training, virtual reality, robot-assisted rehabilitation combined with virtual reality, and conventional therapy to assess the effects on balance, gait, and daily function of patients with stroke. DATA EXTRACTION: The risk of bias was assessed using the Cochrane Risk of Bias tool and the methodological quality of the studies was assessed using the Physiotherapy Evidence Database scale. A network meta-analysis of random effects models was performed for direct and indirect effects. Data were analyzed using Stata SE 17.0 and R 4.2.1. DATA SYNTHESIS: A total of 52 randomized controlled trials involving 1,559 participants were included in this study. Based on the ranking probabilities, robot-assisted rehabilitation combined with virtual reality was most effective in improving balance (surface under the cumulative ranking curve [SUCRA]=82.0%; mean difference [MD]=4.10; 95% confidence interval [CI], 0.43 to 7.67). Virtual reality was most effective in improving velocity (SUCRA=97.8%; MD=-0.15; 95% CI, -0.24 to -0.06) and daily function (SUCRA=92.1%; MD=-7.85; 95% CI, -15.18 to -1.07). CONCLUSIONS: Compared to robot-assisted training and conventional therapy, robot-assisted training combined virtual reality was most likely the best intervention for balance, and virtual reality might be the most helpful in improving daily function for patients after stroke. Further studies are needed to clarify the specific efficacy of robot-assisted training combined with virtual reality and virtual reality on gait.

Development of an Intelligent Risk Hierarchical Management System on Pulmonary Aspiration for the Elderly.

For the elderly, the risks associated with pulmonary aspiration are significant, and can severely impact their quality of life. Therefore, the management of pulmonary aspiration risk is crucial in promoting healthy aging, an aspect often overlooked in self-health management for the elderly. To enhance self-management of pulmonary aspiration risk among the elderly, we have organized a multidisciplinary team to develop an intelligent risk hierarchical management system on pulmonary aspiration for the elderly. This system tailors the assessment with different questionnaires based on the elderly individual's environment. Additionally, it utilizes elderly-friendly formats such as pictures, videos and animations, making it easier for seniors to comprehend the occurrence and hazards of pulmonary aspiration. This enables them to proactively take preventive measures to manage the risk, thus achieving self-management of pulmonary aspiration risk.

Development of an Intelligent Health Education System Based on Large Language Model for Elderly Pulmonary Aspiration Prevention.

As the aging process accelerates, the incidence of chronic diseases in the elderly is rising. As a result, it is crucial to optimize health education for the elderly. Pulmonary aspiration and aspiration pneumonia are significant concerns endangering the health of the elderly. The health education paradigm now in use to prevent pulmonary aspiration in the elderly has numerous flaws, including a lack of home-based health education and the digital divide. Large language model (LLM), an example of artificial intelligence technology, is anticipated to bring a chance to address these issues and offer easily comprehensible health information for the prevention of pulmonary aspiration in the elderly. Our multidisciplinary research team fully understood the needs from the perspective of physicians, nurses and patients, built a knowledge graph (KG), and developed an intelligent Health EducAtion system based on LLM for the prevention of elderly Pulmonary Aspiration (iHEAL-ePA system).

ChatGPT Research in Healthcare is Increasing Dramatically: A Bibliometric Analysis Based on VOSviewer.

The ChatGPT (Chat Generative Pretrained Transformer), an AI-based chatbot, has garnered significant attention in healthcare. We methodically extracted two significant English databases (PubMed and Web of Science) to examine the current state of research and potential future directions for ChatGPT in the healthcare field. Following screening, 921 publications were included in the analysis, demonstrating a sharp rise in the previous year and a peak number of publications released in October 2023. Meanwhile, we obtained 88 items and categorized them into 10 clusters, of which artificial intelligence, ChatGPT, and humans ranked among the top three in terms of frequency.