I.-Chiu Chang

BACKGROUND Comprehensive geriatric assessment (CGA) is performed intermittently, leaving long intervals during which a resident's status may change unobserved. Continuous, low-burden passive sensing could supplement CGA, but the field has tended to overstate predictive value using single-split designs and imbalance-sensitive metrics. The realistic, deployable value of such sensing must be established with rigorous validation. OBJECTIVE To evaluate (1) the feasibility of long-term mattress-based passive sensing in long-term care and (2) whether the resulting behavioral phenotype is clinically useful as a between-assessment monitoring and triage layer for key CGA domains — explicitly not as a replacement for, or direct predictor of, CGA scores. METHODS Retrospective analysis of 192 long-term care residents aged ≥65 years at baseline, continuously monitored with the WhizPad mattress sensor. We quantified data coverage, then built daily-resolution behavioral-phenotype features over 7/14/30-day windows preceding each CGA assessment. For a primary domain (pressure-injury risk, Braden Scale) and a secondary domain (activities of daily living, ADL, Barthel Index) we evaluated discrimination of the higher-need band using unseen-resident (subject-wise) cross-validation (GroupKFold) with resident-cluster bootstrap 95% confidence intervals, calibration (Brier score and reliability), and clinical utility (decision-curve net benefit versus reassess-all and reassess-none strategies). A validation-methodology analysis quantified how naive designs overstate performance. RESULTS Sensing was feasible at scale: median monitoring span 1320 days (~3.6 years), median 1196 resident-days with data, 91% daily coverage, and 226,239 total resident-days. Under unseen-resident validation the sensor phenotype discriminated the higher-need band with an AUROC of 0.654 (95% CI 0.595–0.706) for pressure-injury risk and 0.713 (0.664–0.758) for ADL. Decision-curve analysis showed positive net benefit over both default strategies at decision thresholds of 0.30–0.60 (pressure injury) and 0.20–0.60 (ADL). Feature importance was mechanistically coherent, dominated by in-bed posture/position and movement signals. The methodology analysis showed that an imbalanced three-class macro-F1 endpoint and same-resident temporal validation substantially overstated apparent value: a status-proxy AUROC of ~0.76 under same-resident temporal splitting fell to ~0.62 under unseen-resident validation. CONCLUSIONS Continuous mattress sensing is feasible at scale and provides a modest but genuinely useful low-burden monitoring/triage signal for selected CGA domains, adding net clinical benefit at realistic decision thresholds despite modest discrimination. It is best positioned as a surveillance aid between assessments, not a CGA substitute. We provide a validation protocol to curb the overoptimistic claims common in this field and define the design changes required to strengthen the signal.

Objective: Intradialytic hypotension (IDH) is a common and serious complication during hemodialysis (HD), associated with increased morbidity and mortality. This study aims to develop an artificial intelligence (AI)-driven early warning system to predict IDH events and recommend real-time dialysis parameter adjustments to enhance patient safety. Materials and Methods: A total of 9,634 HD sessions from 178 patients were collected. Advanced feature engineering techniques were applied, including domain-specific variable generation, Synthetic Minority Oversampling Technique (SMOTE), and Recursive Feature Elimination with Cross-Validation (RFECV). An AutoML pipeline tested 100 hyperparameter combinations per run to tailor hospital-specific models. This approach enables rapid development of customized models optimized for institutional variability. Results: Among the evaluated models, CatBoost and LightGBM with Extra Trees achieved the highest performance. CatBoost yielded the best results with a ROC-AUC of 0.934 and an F1-macro score of 0.508. Feature selection and SMOTEEEN resampling further enhanced performance, improving the ROC-AUC to 0.949. Conclusion: The implementation of AutoML for large-scale hyperparameter optimization enables the efficient development of hospital-specific AI models with minimal manual input. This approach supports scalable customization across diverse clinical settings. Our findings underscore the value of AI-powered prediction systems in facilitating early identification and management of intradialytic hypotension (IDH), thereby contributing to improved patient safety and hemodialysis outcomes.

With a burgeoning market and a multitude of on-demand food delivery (OFD) platforms offering diverse options, comprehending the reasons that drive consumers to switch between platforms is paramount. The push-pull-mooring (PPM) theory provides a comprehensive framework for assessing why and how consumers navigate, guiding strategic decisions for service providers seeking to optimize their offerings and retain their customer base. This research employs the PPM theory to rigorously analyze how these elements influence consumers’ intentions to switch between OFD platforms in Taiwan. Findings from a comprehensive survey of 441 OFD users reveal that both pull and mooring factors exert a significant influence on consumers’ inclination to switch platforms, collectively explaining about 42% of the switching intention. Recognizing these critical factors empowers managers to make judicious decisions aimed at enhancing platform offerings and refining marketing strategies, ultimately fortifying customer retention and bolstering satisfaction levels.

This study focuses on the cooperative attitude and intention of retail stores in Taiwan to cooperate with the government’s related pandemic prevention measures. The study is based on the Theory of Planned Behavior (TPB). The study includes factors such as perceived risk of infection, job stress, pandemic prevention IT (information technology) convenience, pandemic prevention attitude, and pandemic prevention intention. Pandemic prevention attitude is used as a mediating variable to establish the research framework. This study collected research data through a questionnaire survey. A total of 457 valid questionnaires were collected through an electronic questionnaire platform. The findings showed that perceived risk of infection and pandemic prevention IT convenience had a positive and significant effect on pandemic prevention attitude (β = 0.567; β = 0.422) and pandemic prevention intention (β = 0.424; β = 0.296). Job stress has a significant negative effect on attitude (β = −0.173). In addition, job stress influenced intention through attitudes. Finally, perceived risk, job stress, and IT convenience had high explanatory power (R² = 0.706) on attitudes. Perceived risk, IT convenience, and attitude also had moderate explanatory power (R² = 0.588) on prevention intention. The study also suggests practical recommendations to improve and cooperate with pandemic prevention intention.

Family members of intensive care unit patients are often experience high anxiety and require more information about the patients. However, most Taiwanese healthcare institutions currently face manpower shortages due to the COVID-19 pandemic. Therefore, the task of providing additional services to meet family members’ needs and relieve their stress was deferred by some healthcare institutions. The self-service system, known to be effective and efficient in other industries, was recommended for use in the healthcare industry. This study aims to explore an intensive care unit self-service system (ICU-SSS) designed for the family members of ICU patients. This study investigates the feasibility of the system by following a mixed method approach, including qualitative interviews and a quantitative survey. Firstly, interviews with five family members and five ICU staff members of a case hospital were conducted to identify the need to develop an ICU-SSS for the family member. Secondly, a survey was completed by 30 family members to evaluate the system. The interview results reveal nine categories of family members’ needs and the survey results show that the ICU family members assigned acceptable scores to all the ICU-SSS functions, except the importance of “Logistical information”. Based on these findings, the scientific and practical implications are discussed.

Most long-term care facilities can offer residents’ with sufficiently material and physical care, but psychological support may not be always provided due to the tight financial budget or labor resources. Residents’ isolation and loneliness then become a big issue, especially for the residents. Social network systems (SNS) have been proved to be a more effective information transmission channel for thoughts, perspectives, and information sharing than traditional channels such as microblogging, e-mails, or telephones. This study conducted a quasi-experiment to identify factors that influence residents’ intention of using SNS and the impacts of SNS on them in a long-term care facility. The results showed that residents’ attached motivation of personal interacting is a significant factor that influences their intention to use the social network platform. Meanwhile, both the loneliness and depression scales of the participants were decreased significantly.

Background: Health information can be more easily transmitted and diffused through the Internet, but questionable online health information often misinforms patients. Physicians have a duty to inform patients how to achieve positive health outcomes. Many physicians often write blogs to provide patients with the right health information. However, most articles available on this subject only describe the blog phenomena without providing a theoretical background and an empirical analysis of doctors using blogs. Methods: This study based on social cognitive theory (SCT) explores the factors influencing physicians’ intention of continuously blogging. A total of 887 physician bloggers were invited to participate in an online survey and 128 valid responses were received. Results: The SCT was proven to be useful in explaining 36.8% of the variation in physicians’ continuous intention to blog. Conclusions: We provide references for platform developers with different strategies to motivate doctors to blog, and the implications and limitations of this study are discussed.