DIGITAL HEALTH最新文献

Introduction: Artificial intelligence (AI)-powered chatbots, such as ChatGPT-4 and DeepSeek, are increasingly utilized in providing medical information. However, their accuracy, comprehensiveness, and reliability, particularly in specialized fields such as colorectal cancer, remain under-evaluated. This study aimed to compare the performance of ChatGPT-4 and DeepSeek in responding to both community- and expert-oriented questions related to colorectal cancer.

Materials and methods: A total of 30 questions were formulated based on clinical experience, including 15 community-focused and 15 expert-oriented questions. On February 13, 2025, ChatGPT-4 (OpenAI, version 4.0) and DeepSeek-R1 (initial January 2025 release) were queried simultaneously in a single session. Responses were independently evaluated by four colorectal surgery experts for appropriateness (0-100), comprehensiveness (0-100), and reference provision (yes/no). Statistical analyses included Mann-Whitney U and chi-square tests, with significance set at p < 0.05.

Results: ChatGPT-4 and DeepSeek demonstrated comparable appropriateness scores (94.0 vs. 92.25, p > 0.05). In community-oriented questions, ChatGPT-4 showed significantly higher comprehensiveness (median 95.0, interquartile range (IQR) 92-98 vs. 90.0, interquartile range 85-94; p = 0.044). Neither chatbot provided scientific references. Inter-rater agreement ranged from good to moderate, with slightly higher consistency observed for DeepSeek (appropriateness ICC 0.83 vs. 0.81).

Discussion: Both chatbots exhibited distinct strengths and limitations. ChatGPT-4 demonstrated superior comprehensiveness in community-oriented responses, whereas DeepSeek provided slightly more consistent evaluations. The absence of scientific references represents a major limitation, restricting clinical applicability and reliability. Enhancing reference support and response consistency is essential before AI-powered chatbots can be safely integrated into colorectal cancer-related clinical decision-making.

Background: The US organ transplantation system is pursuing modernization of the allocation process through the integration of new technologies such as artificial intelligence (AI). However, the legal and ethical issues within the transplantation industry are still of concern.

Objective: We explore the opportunities and challenges for Organ Procurement Organizations (OPOs) to adopt AI. The US organ transplant system is a highly regulated industry yet open to innovation.

Methods: Ten structured interviews were conducted with OPO representatives using the Extended Technology, Organization, Environment (TOE) framework.

Results: Overall, we identified five core tensions in AI adoption: (1) misconceptions, (2) approach to training, (3) need for AI expertise, (4) impact of organization size, and (5) top-down versus bottom-up adoption viewpoints. First, some of the positive perceptions of AI, such as bias elimination, are related to misconceptions about what is technically possible. Second, some OPOs believed that using AI systems requires basic knowledge about the AI system, while others stated that AI should be intuitive and require no training. Third, they disagreed on whether it is necessary to add AI-experienced staff as part of an AI adoption strategy. Fourth, smaller OPOs may struggle to develop, maintain, and implement AI systems due to their limited resources, yet they are more nimble and able to pivot due to less bureaucracy. Fifth, there are competing visions for how AI should be adopted across OPOs nationwide, either top-down driven by regulatory requirements or bottom-up driven by performance expectations.

Conclusions: Ongoing work is needed to determine best practices for integrating AI in OPOs to support optimal organ use and expand transplant access for patients. The TOE framework highlights organization-level tensions that need to be addressed by the transplant sector for successful AI adoption and integration.

Objective: Understanding the factors that shape public support for tobacco control policies is essential for effective legislation. This study aims to examine how online and offline health information seeking behaviors (HISB) among chronic disease patients influence their support for tobacco control policies.

Methods: Using data from a national survey in China (N = 745), this study developed and empirically tested a parallel mediation model examining the direct associations between online and offline HISB and support for tobacco control policies, as well as the indirect paths through perceived social disapprove of smoking and negative smoking outcome expectancies.

Results: Results indicated that online HISB was positively associated with support for tobacco control policies, both directly and indirectly through increased perceived social disapproval of smoking and negative smoking outcome expectancies. In contrast, offline HISB showed no direct association with policy support and exhibited an indirect negative pathway through reduced negative smoking outcome expectancies.

Conclusions: Findings highlight the positive role of online HISB, and the potential negative role of offline HISB, in shaping support for tobacco control policies. We therefore recommend promoting online channels for health information seeking, especially among adults aged 40 to 70 and those from lower socioeconomic backgrounds, who rely more on offline media and face higher chronic disease risk. Online campaigns should emphasize the social unacceptability of smoking and negative smoking outcome expectancies. In parallel, stricter regulation of pro-tobacco content in offline media, especially subtle promotional exposure, is needed, along with increased frequency and depth of antitobacco coverage in traditional media.

Objective: Despite having high healthcare needs and being suitable candidates for digital health consultations, older adults (OAs) use these practices less commonly, and are more likely to encounter challenges, which reduces digital health equity. The objective of this study was to examine the experiences of OAs and healthcare staff (HS) using and participating in digital consultations.

Methods: A descriptive approach was employed, using both structured and semi-structured interviews with OAs (≥65 years) and HS who had completed a digital health consultation via video or asynchronous chat within the past week. Interview data were analysed with thematic analysis and descriptive statistics.

Results: The descriptive data (N = 33 OAs; 13 HS, each reporting on the same 33 consultations) indicated that both OAs and HS gave mainly positive feedback about their recent digital health consultations, while the qualitative data offered a more nuanced perspective. About half of the OAs would choose a digital consultation again for the same health issue, if given the choice. Practical benefits and professional benefits were highlighted, though benefits for care were also noted by both OAs and HS. Barriers were primarily technical, affecting the conversation in various ways. Barriers to care also arose, mainly relating to experiences of deteriorated contact with the HS.

Conclusion: Digital health consultations offer significant practical benefits for both OAs and HS, and sometimes also benefit care. However, persistent technical and relational barriers highlight the need for improved usability and adaptability. Addressing these challenges and supporting individual preferences are essential steps toward increasing accessibility and ensuring equitable access to digital consultations for OAs.

Objective: To evaluate the quality, reliability, and user engagement of endometriosis-related videos on TikTok and Bilibili, identifying variations by platform, uploader type, and content category to inform digital health strategies.

Methods: The top 100 videos per platform were retrieved using the Chinese keyword for "endometriosis." After excluding irrelevant or promotional content, 195 videos (99 TikTok, 96 Bilibili) were analyzed. Categorization included uploader type (professional individuals, nonprofessionals, institutions) and content (disease knowledge, treatment, Traditional Chinese Medicine, other). Quality was assessed via Global Quality Score (GQS), modified DISCERN (mDISCERN), JAMA benchmarks, and Video Information and Quality Index (VIQI). Engagement (likes, collections, comments, shares) and duration were recorded. Analyses used the Wilcoxon rank-sum, Kruskal-Wallis, Fisher's exact, and Spearman correlations.

Results: Professionals uploaded 83.6% of videos; disease knowledge dominated (64.1%). Bilibili videos were longer (median 281.5 vs. 64.67 s; P < .0001) with higher GQS (3.29 vs. 3.04; P = .0123), mDISCERN (3 vs. 2; P < .0001), and JAMA (1 vs. 0; P < .0001). TikTok excelled in engagement (e.g., likes 355 vs. 18.5; P < .0001). Professional sources scored higher (P < .001-.003). Treatment content was most engaging but shorter (P < .001). Engagement correlated internally (P > .7) but weakly with quality (P < .3).

Conclusions: Videos show moderate quality, with Bilibili emphasizing reliability and TikTok virality. Professional content is superior, but the popularity-quality disconnect highlights needs for verification and education to reduce misinformation.

Objectives: Socially assistive robots (SARs) have emerged as promising solutions for elderly care, offering companionship, cognitive stimulation, and therapeutic interventions. However, their evaluation presents unique challenges due to the multidisciplinary nature of these systems and difficulty assessing them in real situations. This study presents SHARA-WoZ, a multistakeholder framework for evaluating SARs through Wizard of Oz (WoZ) methods.

Methods: The framework implements a modular web-based infrastructure comprising three components: robot simulation interface, central server, and desktop wizard control application, supporting autonomous, semi-autonomous, and full-manual operational modes. Evaluation employed a between-subjects study with healthcare professionals and technical experts ( $n=10$ each) using the user experience questionnaire (UEQ), custom usability assessment of system components, and qualitative analysis of post-session interviews.

Results: The framework achieved above-average ratings across all UEQ dimensions. Healthcare professionals showed consistently higher ratings than technical experts, with largest gaps in Efficiency ( $\Delta$ =0.46) and Dependability ( $\Delta$ =0.47), indicating successful alignment with clinical workflows and therapeutic requirements. Qualitative analysis yielded 55 improvement suggestions across Interface Design, New Functionalities, Technical Performance, Integration Features, and Robot Behavior categories.

Conclusion: The findings demonstrate that modular, web-based WoZ infrastructures effectively bridge the gap between research prototypes and real-world deployment requirements. The study confirms the critical importance of multistakeholder evaluation approaches, where healthcare professionals provide clinical perspectives while technical experts contribute optimization insights, ensuring both technical excellence and clinical utility in assistive robotics development.

Objective: This study aimed to identify risk factors for the occurrence and growth of thyroid nodules (TNs) in euthyroid individuals, and to develop a machine learning-assisted predictive model (least absolute shrinkage and selection operator (LASSO) Cox regression) for preliminary individualized risk stratification.

Methods: A nine-year retrospective cohort of 1140 participants with normal thyroid function (6444 data points) was analyzed. First, LASSO regression-a classic regularization algorithm in machine learning-was applied to screen key predictors from 17 candidate variables. The selected variables were then incorporated into multivariable Cox proportional hazards models to construct a LASSO-Cox predictive model. A nomogram was developed, and model performance was assessed using the area under the curve (AUC), bootstrap-corrected concordance index, calibration plots, and decision curve analysis (DCA).

Results: Machine learning-based LASSO regression screened 10 variables for TN occurrence (age, sex, systolic blood pressure (SBP), diastolic blood pressure (DBP), waist circumference (WC), body mass index (BMI), fasting plasma glucose (FPG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), metabolic syndrome (MetS)) and 10 variables for TN enlargement (age, sex, SBP, BMI, hemoglobin A1c (HbA1c), TC, triglycerides, HDL-C, uric acid, MetS). Cox analysis further confirmed that sex, SBP, DBP, WC, BMI, FPG, HDL-C, and MetS were predictors of TN occurrence, while sex, HbA1c, HDL-C, and MetS were associated with TN enlargement. The LASSO-Cox nomogram achieved AUCs of 0.65 for occurrence and 0.64 for enlargement, with bootstrap-corrected C-indices of 0.62 and 0.64. Calibration showed good agreement between predicted and observed risks, and DCA demonstrated consistent net benefit across threshold probabilities of 0.05-0.6.

Conclusion: Sex, HDL-C, and MetS are key predictors of TN occurrence and enlargement. The machine learning-assisted LASSO-Cox nomogram, based on routine clinical indicators, shows moderate discrimination and good calibration. This preliminary tool may assist in population-level risk stratification; however, its modest discrimination limits its immediate clinical implementation.

This commentary explores the intersection of technology, ethics and care delivery in the context of ageing populations. It argues that while smart technologies offer promising solutions to support independence and wellbeing, their adoption is shaped by complex human, ecological and philosophical factors. Drawing on theoretical frameworks, such as the bioecological theory and the cyber-ecological model, this article highlights the importance of co-design, cultural sensitivity and ethical integration. Practical examples illustrate how adaptive artificial intelligence, smart homes and other digital tools can enhance care when thoughtfully implemented. The authors call for interdisciplinary collaboration, inclusive design and evidence-based policy to ensure that technological innovation enhances, rather than replaces, human connections in aged care delivery.

Background: Fibromyalgia is a chronic pain syndrome affecting 2-4% of the global population. Social media platforms host patient narratives and public discussions that can complement traditional evidence.

Objective: To characterize English-language YouTube discourse on fibromyalgia over 2015-2025, describing longitudinal and seasonal engagement patterns and sentiment dynamics, while outlining platform- and translation-related methodological considerations.

Methods: A retrospective cross-sectional analysis was conducted on YouTube videos referencing fibromyalgia (2015-2025) identified via English keywords. When non-English videos appeared, relevant comments were processed using a documented translation pipeline prior to analysis. Counts of videos, comments, and words were summarized; seasonal groups were compared. Sentiment classification (negative/neutral/positive) was performed using an automated model; searches were repeated in anonymized sessions to gauge recommendation-system effects, and a small in-study manual validation was conducted.

Results: A total of 941 videos and 30,896 comments (1,505,579 words) were analyzed. Engagement peaked in spring (12,125 comments) and summer (8899). Neutral sentiment predominated across seasons (62.1-66.3%), with relatively higher positive proportions in summer (25.3%). Seasonal differences in engagement were statistically significant in nonparametric comparisons. Manual validation indicated good overall performance of the sentiment model, with reduced accuracy on translated comments.

Conclusions: English-language YouTube discussions on fibromyalgia increased over the decade and exhibited seasonal patterns. Findings should be interpreted as platform-specific and noncausal, given recommendation-system influences, language constraints, and automated sentiment limitations. Nevertheless, social media analytics can inform patient-centered communication and hypotheses for future research.