Healthcare Informatics Research最新文献

Objectives: The objective of this study was to develop a communication application for rehabilitation patients and physicians based on the shared decision-making (SDM) model. Specifically, an app called REHAB NOTE was designed and implemented for patients undergoing rehabilitation for cancer and central nervous system (CNS) injuries. The REHAB NOTE application aims to facilitate smooth communication between patients and physicians, provide patient-centered medical services, and ultimately enhance rehabilitation treatment effectiveness.

Methods: The development of REHAB NOTE followed a structured approach for mobile app creation, including rigorous requirement analysis, architecture design, navigation design, and detailed page layout planning. This systematic process ensured the platform met the specific needs of both rehabilitation patients and healthcare providers.

Results: We developed an application-based platform service (REHAB NOTE) that enables rehabilitation patients to view doctors' notes after treatment, document their health status, and share this information with their physicians. The platform was specifically designed for cancer rehabilitation patients and CNS injury rehabilitation patients. It can also be utilized by patients undergoing occupational, physical, and speech therapies.

Conclusions: The REHAB NOTE application incorporates concepts from shared decision-making and OpenNotes and is anticipated to positively impact rehabilitation treatment outcomes. Future studies should verify the application's effectiveness. Additionally, modifications and enhancements will be necessary to ensure its applicability to a broader spectrum of rehabilitation patients.

Objectives: The prevalence of chronic kidney disease (CKD) continues to rise, making it one of the leading causes of death worldwide. Recent advances in medical and health research have progressed beyond traditional statistical methodologies, increasingly leveraging artificial intelligence to identify and predict factors influencing mortality. Further AI-based research is therefore essential to deepen understanding of the determinants of death among CKD patients.

Methods: This study used data from the Korea Disease Control and Prevention Agency's in-depth survey of patients discharged between 2016 and 2021. Least absolute shrinkage and selection operator (LASSO) regression, a machine learning technique, was applied to identify significant factors associated with death in CKD patients. These selected variables were then incorporated into a deep learning-based predictive model.

Results: Eight factors influencing death were identified, including length of hospital stay (coefficient = 0.023), emergency admission (0.016), age (0.013), severity-adjusted score (0.008), and regional differences (0.003). The developed deep learning model achieved a loss value of 0.1207 and an accuracy of 96.84%.

Conclusions: This study identified emergency visits and prolonged hospital stays as key predictors of death in CKD patients. To mitigate these risks, regular monitoring by nephrology specialists and timely initiation of renal replacement therapy are essential. Age also emerged as a critical determinant, emphasizing the importance of age-stratified clinical guidelines amid global aging trends. The high-performing, simplified predictive model based on general characteristics offers a valuable tool for rapid prognosis assessment in primary and secondary healthcare settings.

Objectives: This study aimed to identify association rules in patients with multiple chronic conditions, with a focus on patterns involving depression, a highly prevalent psychiatric disorder and a significant risk factor for suicide. Understanding comorbidity patterns in patients with depression is critical for targeting screening efforts, enabling early diagnosis, and improving chronic disease management.

Methods: Maryland Medicaid claims data from 2021 to 2022 were analyzed to examine the co-occurrence of depression with 62 other chronic conditions using association rule learning. Analyses were stratified by sex and age group to identify patterns specific to demographic subgroups. Thresholds for case numbers and confidence levels were applied to ensure that identified rules were both clinically meaningful and statistically robust.

Results: The study showed a marked increase in the number of association rules with advancing age, particularly among women compared to men. In total, 582 association rules were identified, providing important insights into comorbidity structures.

Conclusions: This study demonstrates the utility of association rule learning for detecting clinically relevant patterns of depression comorbidities, including variations by age and sex. The identified rules could inform clinical practice by improving targeted screening, facilitating early diagnosis, and guiding management strategies for patients with multiple chronic conditions.

Objectives: The purpose of this study was to conduct an integrative review of existing research on care robots for community-dwelling older adults and to suggest directions for future research and technology development in this area.

Methods: We focused on robots, including care robots and socially assistive robots, that help older adults living in the community maintain independence at home. Three electronic academic databases (PubMed, CINAHL, and Cochrane) were searched for eligible research articles. The keywords used included elder*, older adult*, robot* care, care robot*, assist* robot*, service robot*, companion* robot*, socia* robot*, home-based, and community-based, among others.

Results: A total of 834 research articles were identified, and 40 were ultimately reviewed and analyzed. The studies were categorized into three groups: perceptions and needs related to care robots; cognitive support; and assistance with activities of daily living.

Conclusions: It is necessary to develop and implement care robots with diverse functions that can provide practical assistance for the independent daily living of older adults. This will require collaboration among government agencies, public institutions, academia, and private health enterprises. In addition, policies must be established to support the purchase and maintenance costs of care robots to ensure continued access for community-dwelling older adults.

Objectives: The integration of digital technology has greatly expanded public access to health and drug-related information through the Internet. However, the rapid proliferation of unverified content on websites targeting the general population raises serious concerns about health misinformation. This study aimed to evaluate the quality of Indonesian drug information websites accessible to the public and to design a verified, web-based drug information system.

Methods: A cross-sectional evaluation was conducted using the Quality Evaluation Scoring Tool (QUEST) to assess the quality of publicly available drug information websites in Indonesia. Development of the verified drug information platform followed the Rapid Application Development model, employing a prototyping approach.

Results: Among the 14 publicly accessible drug information websites evaluated, 5 (35.71%) were classified as low quality (QUEST score ≤9), 4 (21.42%) as moderate quality (score 10-18), and 5 (35.71%) as high quality (score >18). The drug information website developed by the Faculty of Pharmacy, Universitas Andalas, achieved a high-quality rating, with a QUEST score of 27 (96.43%), although it received the lowest subscore in the Complementarity domain. Higher QUEST scores indicate better information quality.

Conclusions: The findings show that nearly half of the websites providing drug information to the Indonesian public are of low quality. The website developed by the Faculty of Pharmacy, Universitas Andalas, demonstrated strong overall quality, but improvements in the Complementarity domain are recommended to further strengthen user engagement and support.

Objectives: Ontologies play a crucial role in healthcare systems due to the diversity of concepts, roles, users, and diagnostic and therapeutic methods. They facilitate the development of knowledge bases and the sharing and representation of information. With the integration of artificial intelligence (AI) into healthcare, ontologies can serve as complementary tools to enhance the quality of services.

Methods: This review study examines existing research on the application of ontologies in AI systems within the healthcare industry. By analyzing their applications, benefits, challenges, and limitations, the study seeks to provide a deeper understanding of their impact on advancing AI technologies and improving healthcare processes. In addition, the study offers recommendations for strengthening the development and use of ontologies in intelligent healthcare systems.

Results: The findings of this review indicate that ontologies enhance the accuracy of results and support medical decision-making by enabling the semantic exchange of diverse and heterogeneous data. They are essential for the development of decision support systems and for fostering intelligent interactions between patients and healthcare systems. Furthermore, ontologies contribute to healthcare decision-making by semantically analyzing the connections between diseases, geographic regions, and environmental factors.

Conclusions: The use of ontologies in healthcare improves data analysis, patient diagnosis, treatment, and decision-making. Ontologies enhance data inference and interoperability in AI systems through data modeling, concept relationship extraction, knowledge enrichment, and information sharing. Given the vast scope of the healthcare domain, the diversity of specialties and data, and the absence of a dedicated ontology development methodology specific to this field, there is a clear need for a tailored and robust methodology.

Objectives: All clinical trials face a significant bottleneck in identifying eligible participants, particularly due to the complexity of unstructured medical texts. Recent advances in natural language processing, especially the advent of transformer-based models, have shown promise in this domain. In this study, we evaluated the performance of a prompt-based large language model (LLM) for cohort selection from unstructured medical notes.

Methods: Medical records were annotated with Med- CAT using the Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT) ontology. For each trial eligibility criterion, we extracted sentences containing relevant annotated concepts through an ontology-driven summarization process. These summaries were then input into a prompt-based LLM (GPT-3.5-turbo), tasked with classifying eligibility criteria in a zero-shot setting. Model performance was assessed using the 2018 National Natural Language Processing Clinical Challenges (n2c2) dataset, which required the classification of 288 patients' medical records according to 13 eligibility criteria.

Results: The proposed prompt-based model achieved overall micro and macro F-measures of 0.9061 and 0.8060, respectively-among the highest scores reported for this dataset.

Conclusions: Our results demonstrate that integrating ontology-based extractive summarization with prompt-based LLMs can substantially improve eligibility classification. The summarization step enhanced model focus and interpretability, particularly for long or ambiguous narratives. This pipeline offers a scalable and adaptable framework for clinical trial automation and has the potential for real-world integration with electronic medical record matching systems.

Objectives: This study analyzed publications in Healthcare Informatics Research (HIR) to identify trends and shifts in research focus within both the journal and the broader Korean medical informatics landscape. By examining keywords across these papers, the study aimed to elucidate evolving priorities and innovations in the field over time.

Methods: Data from 958 papers published between 1995 and 2024 were extracted from the HIR journal's online archive. The analysis focused on English-language articles published since 2010 (n = 658) to examine publication trends using descriptive statistics. Keyword and Medical Subject Headings (MeSH) term analyses (term frequency-inverse document frequency, latent Dirichlet allocation, co-occurrence) were performed on a subset of articles with available abstracts (n = 632) to identify research themes and interrelationships. Inferential statistics, including chi-square and regression analysis, were applied to assess changes in research trends over time.

Results: Among 958 total papers identified (672 in English), analysis of 658 English articles published since 2010 revealed increasing publication trends, peaking between 2015 and 2018. Keyword and MeSH term analyses of 632 papers with abstracts highlighted persistent themes (e.g., health systems, electronic health records) alongside emerging topics (e.g., machine learning, telemedicine). Inferential analysis indicated no statistically significant changes in keyword distribution over time.

Conclusions: This study offers insights into the evolution of health informatics research in Korea, underscoring the role of HIR in documenting this progression. The findings reveal a balance between emerging technologies and foundational healthcare themes, demonstrating the field's adaptability and sustained relevance. Future research should extend the analysis to other journals and further consider ethical implications and global developments.

Objectives: This scoping review synthesizes literature on the integration of large-scale data analytics for cardiovascular disease (CVD) prediction, aiming to provide insights that support the adoption of predictive analytics for improved prevention and early detection in healthcare.

Methods: Searches were conducted in Medline (PubMed), EBSCO, Google Scholar, and Wiley Online Library. Medical Subject Headings (MeSH) search terms included: large-scale data, big data, cardiovascular diseases, prediction, machine-learning algorithms, artificial intelligence, and mortality. The search covered the period from 2020 to 2024.

Results: Of 262 retrieved articles, 16 were included. Three main themes were identified: large-scale data analysis techniques and machine-learning algorithms; applications of machine-learning algorithms and artificial intelligence in predicting cardiovascular diseases; and the role of integrating large-scale data in disease prediction to improve the quality of care.

Conclusions: While machine learning provides considerable opportunities for predicting CVD outcomes, limitations remain. Machine-learning approaches are not always the most appropriate option, particularly in basic research where causal relationships between variables may be more critical than optimized predictions. To ensure fair and effective healthcare outcomes, issues related to bias, data quality, ethical concerns, and practical implementation must be addressed. Overcoming these challenges will require interdisciplinary collaboration, methodological refinement, and further research.