Health Informatics Journal最新文献

Objective: ChatGPT has been recognised as a potentially transformative tool in higher education by enhancing the teaching and learning process. Cross-sectional evaluations have acknowledged this potential. This study evaluates ChatGPT's performance in solving specific biostatistical problems, focusing on accuracy, stability, and reproducibility, and explores its potential as a reliable educational tool in medical education. Methods: The correlation analysis task from Statistics at Square One by Swinscow and Campbell was chosen for its foundational role in biostatistics. Between October 2023 and March 2024, and July 2024, GPT-3.5 and GPT-4 were tested for accuracy in 12 parameters. Results: A statistically significant change in correct response rates was established in repeated measurements in the period October 2023, March 2024, and July 2024 for GPT-3.5 (Q = 100.99, p < 0.001), GPT-4.0 (Q = 89.55, p < 0.001), respectively. The significant GPT-3.5 improvement was established between March 2024/July 2024 (p = 0.004), and between October 2023 and July 2024 (p = 0.008). The significant GPT-4.0 improvement was established between October 2023 and March 2024 (p = 0.004), and between October 2023 and July 2024 (p = 0.026). Conclusion: Over 9 months, GPT-4 demonstrated rapid and consistent improvements, achieving perfect accuracy by March 2024. Although this study documented ChatGPT's advancement within 9 months, ChatGPT should be positioned as a supplementary tool in higher education classrooms, in the presence of educators, to enhance the learning process.

Background: The widespread availability of smartphones has created new opportunities for engaging pregnant women and enhancing their self-management abilities to promote maternal and fetal health through mobile interventions. This study focuses on the design and development of a gamification-based mobile health (mHealth) application aimed at providing nutritional support to pregnant women.Methods: An iterative, user-centered design approach and Agile development method were employed to create the application. The developmental stages included identifying the application's features, the design and development process, and evaluation. End users assessed usability using the System Usability Scale (SUS), while experts evaluated quality using the Mobile Application Rating Scale (MARS).Results: Feedback from experts and end users categorized the application's functionalities into general, specific, and gamification-related functions. Pregnant women rated the application's usability as acceptable (68.25 ± 10.86), and experts rated its quality as acceptable (mean 3.89 out of 5, SD 0.25).Conclusions: The positive evaluation results support the use of this application as a tool for managing gestational nutrition and enhancing self-awareness. Future research should investigate its impact on the nutritional status of pregnant women and their infants.

Objective: Accurately predicting functional outcomes after acute ischemic stroke is essential for healthcare institutions to optimize staffing and resource allocation. Although text mining has been applied to build such models, most prior studies emphasize traditional machine learning, with limited comparison to deep learning methods. Methods: Clinical text notes were collected from a Taiwanese hospital to build the experimental dataset. Four textual feature representation techniques were evaluated: bag-of-words (BOW), term frequency-inverse document frequency (TF-IDF), embeddings from language models (ELMo), and bidirectional encoder representations from transformers (BERT). Correspondingly, four predictive models were tested: k-nearest neighbor (KNN), support vector machine (SVM), convolutional neural network (CNN), and long short-term memory (LSTM). Results: The best performance was obtained using BOW features with an SVM classifier. Feature fusion strategies, combining representations such as BOW + TF-IDF and BOW + BERT, also yielded strong performance. Notably, the BOW + TF-IDF combination with SVM achieved the lowest type I error, effectively minimizing the misclassification of patients with poor outcomes. Conclusion: Traditional machine learning methods outperformed deep learning models in this study. Among all combinations, BOW + TF-IDF features with SVM provided the most accurate predictions and lowest risk of false positives in stroke outcome prediction.

Objective: Accessing reliable medical information online in Germany is often hindered by misinformation and low health literacy. Tala-med, an ad-free search engine, was developed to provide curated, expert-reviewed content with filters for trustworthiness, recency, user-friendliness, and comprehensibility. This study re-engineered the original system to overcome technical limitations while maintaining result consistency. Methods: A modular architecture was designed using Elasticsearch, a fastText-based synonym system, and a subZero-powered admin interface. The system was evaluated using 214 unique queries to compare performance and result similarity with the legacy version. Results: The new implementation improved query processing speed while preserving result consistency. Synonym handling was enhanced using fastText, and system maintainability increased via a centralized database and modular backend. The administrative interface simplified data updates and configuration tasks. Conclusion: The re-engineered tala-med search engine maintains the original system's strengths while enabling greater scalability, flexibility, and future extensibility. The open-source platform offers a foundation for advancing domain-specific search systems and supports applications beyond the medical field.

Introduction: Non-communicable diseases are a global health concern endangering public health as well as social and economic progress worldwide. Although Ethiopia's healthcare delivery system has seen tremendous advancements in the use of digital health technology (DHT) for disease management, no major changes have been made. Thus, this study aims to assess barriers and facilitators of DHT intervention for chronic disease management in Ethiopia. Method: A systematic review of literatures was conducted following PRISMA guidelines and using the PICOS approach. Studies were identified through PubMed, Cochrane, HINARI search and other gray literature between March and April 2024. Data was extracted and organized using a standardized Excel sheet, and a descriptive thematic analysis was performed to categorize and summarize the findings and presented in tables and diagrams. Results and conclusion: This review included 12 articles that fulfilled inclusion criteria. The review revealed barriers to DHTs such as lack of technological understanding, negative attitudes, and limited access to necessary resources and facilitators like, perceived usefulness, positive attitudes towards DHTs, and good access to necessary technological tools. This review highlights the need for promotion of facilitators and addressing barriers with targeted strategies to improve the design, implementation, scaling, and sustainability of DHTs.

Background: Large language models (LLM) still face challenges in accurately extracting and summarizing medical information from EHR and EMR. The variability in EHR and EMR formats across institutions further complicates information integration. Moreover, doctors need to spend a lot of time reviewing patient information, which affects the efficiency and effectiveness of clinical decision-making. Objective: This study aims to develop a medical record summarization system that uses the innovative X-RAG technique with GPT-4o to extract medical information from EHR and EMR and convert them into structured FHIR format. The system ultimately generates a doctor-friendly report to improve the efficiency and effectiveness of clinical decision-making. Methods: We propose an innovative X-RAG, which adds page-based chunking, chunk filtering, and guided extraction prompting to the basic framework of RAG and combines it with GPT-4o to extract medical measurement data, diagnostic reports, and medication history records from EHR and EMR with high accuracy. Results: The system achieved 96.5% accuracy in medical data extraction and reduced approximately 40% of the time doctors spend reviewing patient information in clinical applications. Conclusion: The proposed system improves the efficiency and effectiveness of clinical decision-making and provides a valuable tool to optimize medical information management and clinical workflows.

The study had two primary goals: (1) to propose a methodological approach for introducing crowdsourced e-health services within healthcare institutions, and (2) to evaluate the readiness of citizens to adopt the proposed services. The proposed methodological approach addresses the essential infrastructural elements required for introducing crowdsourced e-health services, including their integration into institutional web portals and alignment with broader national digital health systems. By enabling structured citizen participation and facilitating dynamic data exchange among key stakeholders, the approach supports the modernization of healthcare service delivery. This research examined young citizens' readiness to use crowdsourced e-health services to assess the potential for adopting the proposed method. The findings indicate that perceived value is positively influenced by trust, while both perceived value and perceived behavioral control have a significant impact on the intention to contribute. This research introduces an original methodological approach tailored to support the implementation of crowdsourced e-health services within healthcare institutions. The proposed model stands out for its adaptability, as it combines communication, collaboration, crowdsourcing, and payment services within a unified structure. Its flexibility allows integration across different institutional levels, promoting citizen participation and enabling more transparent, efficient, and needs-driven healthcare delivery.

Objective: With the growing burden of type 2 diabetes and its associated healthcare costs, the factors influencing future expenditures, particularly among long-term care insurance (LTCI) users, must be identified. Few studies have addressed the prediction of multiple cost domains, including medical, LTC, and dental expenditures. This study predicted medical, dental, and LTC costs in the following year for patients with type 2 diabetes and identified key predictors based on health information from the previous year. Methods: We applied three machine learning models-random forest, boosted trees, and neural networks-to LTCI users' data in Japan and incorporated prior-year healthcare costs, service usage patterns, and diabetes status. Results: In the 2019 medical cost model, boosted trees showed the best performance for those aged 74 or younger (R² = 0.46, RMSE = 151,804 JPY). LTC costs were influenced by prior LTC spending (∼40%) and facility service use (30-50%), while dental costs were predicted by prior dental expenditures. Conclusions: Prior-year medical costs strongly influenced later medical expenditures, while LTC costs reflected prior LTC spending and facility use. These quantified relationships provide insights for healthcare cost optimization and support policymakers in designing preventive strategies and care systems for aging populations with chronic diseases.

Objective: While smart healthcare services have shown potential in improving healthcare efficiency and effectiveness, significant barriers remain for consumers' long-term engagement in such services. The study sought to propose and validate a theoretical framework to investigate the continuous use of smart healthcare services. Methods: The research model integrates commitment-trust theory with the information system success model, empirically validated through partial least squares structural equation modeling. Data were collected via a Chinese online survey platform, targeting 355 active users of smart health services. Results: The proposed model explained 61.4% of the variance in continuous usage intention. Affective commitment, trust, and satisfaction significantly affected continuous usage intention (p's < 0.01). Trust and satisfaction were found to significantly influence affective commitment (p's < 0.001). Satisfaction and perceived value were found to be significant determinants of trust (p's < 0.05). Perceived value also significantly influenced satisfaction (p < 0.001). The relationships were also moderated by age, gender, and AI literacy. Conclusion: This study represents rare attempts to explore continuous usage intention of smart healthcare services from the commitment-trust theory perspective. Practitioners should prioritize trust-building measures (e.g., transparent data usage policies) and personalized features (e.g., adaptive health recommendations) to enhance long-term engagement. Demographic characteristics should also be considered when designing such services.

BackgroundAI tools are becoming primary information sources for patients with chronic kidney disease (CKD). However, as AI sometimes generates factual or inaccurate information, the reliability of information must be assessed.MethodsThis study assessed the AI-generated responses to frequently asked questions on CKD. We entered Japanese prompts with top CKD-related keywords into ChatGPT, Copilot, and Gemini. The Quality Analysis of Medical Artificial Intelligence (QAMAI) tool was used to evaluate the reliability of the information.ResultsWe included 207 AI responses from 23 prompts. The AI tools generated reliable information, with a median QAMAI score of 23 (interquartile range: 7) out of 30. However, information accuracy and resource availability varied (median (IQR): ChatGPT versus Copilot versus Gemini = 18 (2) versus 25 (3) versus 24 (5), p < 0.01). Among AI tools, ChatGPT provided the least accurate information and did not provide any resources.ConclusionThe quality of AI responses on CKD was generally acceptable. While most information provided was reliable and comprehensive, some information lacked accuracy and references.