Objective This study examined behavioural intention to use digital health-promotion tools for weight management among Taiwanese young adults using an extended Technology Acceptance Model incorporating subjective norm and weight-management self-efficacy. Method This cross-sectional online survey used convenience sampling through social media and included 650 participants aged 18–25 years. Participants completed measures of perceived usefulness (6 items), perceived ease of use (6 items), subjective norm (2 items), behavioural intention (2 items), and weight-management self-efficacy (4 items). Structural equation modelling was the primary analysis, while hierarchical regression quantified the incremental variance explained by weight-management self-efficacy. Results The structural model showed adequate fit and explained 64.2% of the variance in behavioural intention. Perceived ease of use, perceived usefulness, and subjective norm showed significant direct associations with behavioural intention. Weight-management self-efficacy showed a significant indirect association through perceived usefulness. However, bootstrap inference did not support its direct association and its incremental contribution was small (ΔR 2 = .004). Conclusion Designers should prioritise intuitive interfaces and clearly demonstrating how tool functions support weight-management goals. Healthcare professionals may recommend credible, goal-aligned tools, consider users’ digital preferences and weight-management self-efficacy, and help identify usability barriers. Future research should evaluate these strategies by measuring actual uptake and sustained use.
Background and aims Appendicitis is one of the most frequent conditions in emergency departments. Artificial intelligence (AI)-based language models represent a novel source of medical information; however, their reliability and readability across languages remain uncertain. This study aimed to evaluate the reliability and readability of AI-generated appendicitis information. Methods Seven standardised questions were formulated in Spanish and English. Two independent reviewers assessed response reliability using a validated tool. Spanish readability was measured using the Szigriszt-Pazos formula, while English readability was assessed via the Flesch Reading Ease (FRE) and Flesch–Kincaid scores. Results were compared across tools and languages. Results For Spanish reliability, Perplexity® produced the highest proportion of complete responses (71.43%), followed by ChatGPT® (57.14%) and Gemini® (28.57%). In English, Perplexity® and Gemini® both achieved 71.43%, followed by ChatGPT® (57.14%). Readability analysis via FRE formula in English revealed significant differences between the models ( p < 0.05). Perplexity® showed the best reliability but yielded text that was difficult to read in Spanish and very difficult in English. Conclusion Perplexity® emerged as the most reliable model in both languages, although readability remains a concern. There is a clear need for improvements to optimize the accuracy and accessibility of AI-generated medical information.
Telehealth was rapidly adopted during the COVID-19 pandemic as a means of widening access to care, yet the same technologies can reproduce or amplify the disparities they are meant to close. This editorial introduces the Equity and Equality in Telehealth Special Collection, synthesizing four articles that examine virtual care from the level of the individual clinical encounter to that of global policy and technological design: the co-design and implementation of an Indigenous patient experience tool, an international expert roundtable on telemedicine infrastructure and sustainability, a rapid review of ethical and equity considerations in telerehabilitation, and a quantitative comparison of patient-centered communication across telehealth and in-person visits. Together they show that equitable telehealth depends not on access alone but on community co-design, sustained infrastructure investment, honest measurement, and an enduring commitment to the relational core of care.
Purpose Intracranial hemorrhage (ICH), as a severe brain injury, requires immediate treatment and is usually being diagnosed from Computed Tomography (CT) scans in clinical settings. This task is considered a challenging procedure due to interpretation variability between experts, imaging modality limitations, time sensitivity, and heterogeneous presentations. Recent studies have utilized deep neural networks to overcome limitations such as overfitting and poor generalization across scanners, for ICH diagnosis which requires heavy processing (in terms of Graphical Processing Units (GPUs)) and accurate detection in a time-sensitive period. Approach This paper presents deep learning models with attention mechanisms for the purpose of detecting and classifying acute ICH in CT scans. Some pre-processing is needed before inserting the images into the models, which include data splitting, normalization, resizing and augmentation. The proposed approaches combine Channel and Spatial Attention methods with two Convolutional Neural Network (CNN) architectures to extract features and make initial predictions. Furthermore, PCA is used followed by a Bi-LSTM network to reduce the number of features for faster classification and finally, it is concatenated with the initial prediction to produce the final accurate result. This integration particularly benefits ICH classification since Bi-LSTM better understands temporal relations between CT neighboring slices, attention modules focus on hemorrhagic regions, and the PCA method reduces the feature space for less computation. Results Result of experiments show that the proposed methods outperform baselines on the RSNA dataset for six classes in most evaluation metrics, including an average accuracy and an average F1 score of 97.8% and 73.3%, respectively. The Grad-CAM technique is also applied to visualize the important parts of images in the detection process of the model. Conclusions According to the results, using attention modules together with pre-trained models can improve the accuracy of detection and classification for ICH. In addition, Grad-CAM results were approved by the medical team as a supporting analysis, especially in critical situations where the hemorrhaged region could not be identified by physicians which shows interpretation models can help the physicians for more reliability in the detection process.
Introduction Current preventive screening reminders use generic due-date notifications that lack personalization. Targeted reminders allow personalized messages and resource allocation, yet the accuracy of behavioral forecasting with cholesterol screening behaviors remains unconfirmed. Therefore, this study aims to examine the accuracy of machine learning in predicting adults’ future cholesterol screening behavior. Methods This study included 8,288 participants from the Health and Retirement Study. Never receiving cholesterol screening over 12 years of follow-up (2007-2018) was identified as the target outcome. Predictors were constructed based on the Social-Ecological Model. Four machine learning (ML) algorithms were compared: logistic regression, random forests, gradient boosting machine (GBM), and eXtreme Gradient Boosting (XGBoost). Model performance was evaluated by Recall, Precision, and false-negative rate (FNR). Results Among the four ML algorithms, XGBoost performed best, with higher Recall (average: 0.8850) and the lowest FNR (0.1311), but suboptimal precision (average: 0.6228). The top five predictors were past cholesterol screening behavior, diagnosis of hypertension, osteoarthritis or rheumatoid arthritis, number of private insurance plans, and gender. Conclusion ML might be used to predict individuals’ cholesterol screening behavior. Future research is needed to examine whether leveraging ML in targeted reminders might enhance cholesterol screening behaviors.
