Health Informatics Journal最新文献

This study describes the development and design of the Joint Transition Clinic, a collaborative solution designed to support adolescents with epilepsy (AWE) in their transition from pediatric to adult healthcare. Using a participatory design (PD) approach, the study involved AWEs, parents, nurses, and physicians in an iterative development and co-design process. Through a series of workshops, users suggested three ideas: "The Bridge," "The Knowledge Giraffe," and "No rules app." After considering all three ideas, the research team and participants agreed to proceed with the development of "The Bridge," integrating elements from "The Knowledge Giraffe." This process led to the creation of the Joint Transition Clinic, encompassing many of the AWEs' needs and wishes. The PD approach proved effective in creating an organizational-based intervention that addresses patient needs and supports self-management while ensuring AWEs and their parents had a voice in its development, leading to a solution ready for pilot testing.

ObjectiveTo develop a robust deep learning framework for automated multi-class retinal disease detection supporting clinical decision-making, addressing existing models' limitations in generalizability and accuracy.MethodsA novel ensemble model, ResEfficientNetB3, integrating EfficientNetB3 and ResNet50, was proposed. Two Kaggle datasets were used: Dataset 1 (4217 images, four classes) and Dataset 2 (8230 images, eight classes). Images were resized to 224 × 224 with augmentation (rotation ±20°, zoom 0.8-1.2, flipping, scaling). Models were trained using the Adam optimizer (learning rate = 1e-4, batch size = 20) for up to 50 epochs with early stopping and dropout (0.3-0.5). Performance was assessed via standard splits, five-fold cross-validation, and cross-dataset validation.ResultsResEfficientNetB3 achieved 99.0% accuracy on Dataset 1 and 98.2% on Dataset 2, outperforming EfficientNetB3 (94.0%) and ResNet50 (91.0%). Five-fold validation confirmed robustness (99.0% ± 0.2 and 98.2% ± 0.3), and cross-dataset validation showed strong transferability (94.5-95.8%).ConclusionResEfficientNetB3 effectively combines EfficientNetB3's scaling and ResNet50's residual learning, demonstrating superior accuracy, robustness, and generalization. The model offers a reliable, clinically applicable tool for automated retinal disease detection in real-world diagnostics.

Objective: Atrial fibrillation (AF) is the most common sustained arrhythmia in clinical practice and is associated with an elevated risk of stroke, heart failure, dementia, and mortality. As its clinical consequences are strongly influenced by modifiable risk factors, this study aims to design a patient journey for individuals undergoing AF treatment, with the goal of improving patient safety and healthcare delivery. Methods: An empirical study was conducted using an action design research approach. The research focused on identifying and implementing design principles to enhance digital health platforms and support AF management. Results: The study resulted in five key design principles: (i) comprehensive requests for medical interventions, (ii) visualization of patient trajectories, (iii) prioritization of waiting lists informed by real-time data, (iv) equality and inclusion throughout the patient journey, and (v) rapid access to and visualization of quality indicators. These principles collectively address current challenges in AF care by optimizing data use, strengthening patient involvement, and improving decision-making. Conclusion: We propose adjustments to the design of digital health platforms for AF management based on the identified principles. Such adaptations have the potential to enhance patient safety, improve healthcare delivery, and create more efficient, inclusive, and data-driven processes in AF management.

Objective. This cross-sectional study compared the color accuracy and image quality of intraoral photographs taken with DSLR cameras, smartphones, and smartphones with auxiliary lighting. Methods. Forty participants had five images captured per device, yielding 600 images. Image quality was evaluated using Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) via No-Reference Matrix (NRM), while experts provided qualitative evaluations. Results. Smartphones achieved the highest average NRM score (53.14 ± 28.2), followed by smartphones with auxiliary lighting (47.62 ± 1.9) and DSLR cameras (45.32 ± 2.2), with no significant difference between DSLR cameras and smartphones with auxiliary lighting (p = 0.34). Color accuracy (ΔE) was closest between DSLR cameras and smartphones with auxiliary lighting (4.89 ± 2.47), while other pairs showed higher differences. These two also showed comparable color consistency (p = 0.63 and 0.57 for a; b values), although smartphones produced brighter images (L = 50.56 ± 4.82 vs. 31.84 ± 4.82). Experts preferred DSLR images for caries diagnosis and presentation (96.7% preference), but found smartphone images with auxiliary lighting clinically acceptable. Conclusion. While DSLR cameras delivered superior image quality, smartphones with auxiliary lighting demonstrated comparable diagnostic performance as a practical, low-cost alternative in resource-limited settings. Further validation with newer devices is recommended.

Objective: Blood pressure (BP) is a vital factor for human health and survival, and its elevation and fluctuations can have dangerous consequences on an individual's well-being. Traditional BP measurement methods-including cuff-based devices and invasive arterial catheters-are unsuitable for continuous monitoring in daily life: cuffs are intermittent and uncomfortable, whereas arterial lines provide continuous data but are invasive and confined to clinical settings (e.g., ICUs/ORs). In response to this requirement, we propose a cuff-less, continuous, and noninvasive system for BP measurement using photoplethysmograph (PPG) signals and machine learning (ML) algorithms. Methods: In this investigation, we analyzed PPG signals acquired from a diverse cohort, with participants ranging in age from 21 to 86 years and including both healthy subjects and those with health conditions. The data underwent rigorous preprocessing and feature extraction procedures. To address computational efficiency and mitigate overfitting, we applied five distinct feature selection methods to refine the feature set. Subsequently, each method's selected features were independently trained and tested using five ML regression algorithms to estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP). Results: Our findings reveal that the ensemble-based extra trees (ET) algorithm, coupled with the SelectFromModel feature selection approach, surpassed competing algorithms in estimative performance. The ET algorithm achieved notably low root mean squared errors (RMSEs) of 5.21 for SBP and 2.65 for DBP, demonstrating its exceptional capability in the estimation of BP. Conclusion: The proposed approach demonstrates strong potential for accurate, non-invasive BP estimation. These findings have important implications for the development of wearable and mobile health technologies that support continuous, real-time BP monitoring for the prevention and management of hypertension and cardiovascular diseases.

Objective: In 2019, Cyprus implemented on a country-wide basis the Electronic Health Record (EHR) system as part of its General Health System (GHS). This study aims to assess the efficiency levels of EHR users within the GHS. Methods: A cross-sectional study was conducted between October and December 2022 using an electronic self-reported questionnaire. A total number of 429 physicians, both general and outpatient, from various Cypriot provinces participated. Results: The study revealed a moderate level of EHR user efficiency. Several demographic and professional factors, including age, years of experience, computer literacy, EHR familiarity, training, and support, were positively correlated with perceived EHR efficiency. Conclusion: To achieve Cyprus' strategic eHealth goals within the broader European context, improvements in EHR implementation, user training, and support are crucial. Ensuring equal access for all healthcare professionals remains a key priority.

Objectives: This article aims to develop the ontology framework of smart healthcare system and identify the challenges to construct the smart healthcare system. The ontology framework provides both academics and practitioners a reference to understand and transform the healthcare system. Methods: The publications in the area of the smart healthcare system were extracted from WOS core collection database. Latent Dirichlet Allocation (LDA) was employed to find subjects of publications. Natural language processing (NLP) was used to extract entities from topics explored based on LDA. The developed ontology framework of the smart healthcare system was then presented in OWL format using Protégé software. The challenges in transforming towards the smart healthcare system were identified based on the developed ontology framework. Results: Fourteen challenges are identified through the ontology framework developed by NLP and LDA, including poor system interoperability, data security and data sharing, low adoption of data standards and data scalability, etc. These challenges provide a reference for future healthcare workers to deal with possible risks and difficulties. Conclusions: The ontology framework developed by NLP and LDA provides a unified description and structured knowledge in smart healthcare system, and provides valuable working methods and management basis for scholars and medical workers.

Objective: Factors influencing users' adoption of the home-based health monitoring system (HHMS) were examined by integrating embodied cognition with the unified theory of acceptance and use of technology (UTAUT) model. Methods: Data from 459 survey respondents were analyzed using partial least squares structural equation modeling (PLS-SEM). Results: The model explained 59.7% of the variance in behavioral intention to use the HHMS (typical range: 40%-60%). Perceived contextual adaptation, perceived sensorimotor feedback, and perceived body awareness significantly influenced behavioral intention. Perceived body awareness (i.e., an individual's ability to perceive and interpret bodily signals) was identified as a crucial factor affecting performance expectancy, effort expectancy, facilitating conditions, and social influence. Conclusions: The integration of embodied cognition with the UTAUT model contributes to theoretical advancements and demonstrates the importance of body awareness in users' adoption of the HHMS, providing practical guidance for the effective design of HHMS.

ObjectiveThis study aimed to assess the performance of Artificial Intelligence (AI) compared to human experts in healthcare policymaking.MethodsThis was a mixed-methods cross-sectional study conducted in Iran during the years 2024-2025, comparing, and analyzing the responses of multiple AI Large Language Models (LLMs) including Bing AI Copilot and Gemini and a sample of 15 human experts-using confusion matrix analysis. This analysis provided comprehensive data on the respondents' ability to answer context-specific questions regarding healthcare policy making, evaluated through multiple parameters including sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and overall accuracy.ResultsCopilot demonstrated a sensitivity of 0.867, specificity of 0, PPV of 0.722, NPV of 0, and accuracy of 0.65. In comparison, Gemini exhibited a sensitivity of 0.733, specificity of 0.4, PPV of 0.786, NPV of 0.333, and also an accuracy of 0.65. Additionally, the human experts' responses indicated a sensitivity of 0.5808, specificity of 0.2571, PPV of 0.7189, NPV of 0.1579, and an accuracy of 0.5050.ConclusionThe AI LLMs outperformed human experts in responding to the study questionnaire. The findings demonstrated the considerable potential of the LLMs in enhancing healthcare policy-making, particularly by serving as complementary tools and collaborators alongside humans.

Objective: This study aims to present the development of a mobile application incorporating accessibility, communication features, and repeated learning opportunities to support patients and caregivers in managing indwelling medical devices at home. Methods: The application development follows the Analysis, Design, Development, Implementation, Evaluation model. This protocol includes a literature review, application structure and prototype development, and pilot study design. The content is grounded in Bandura's self-efficacy theory and includes behavior change techniques to increase self-efficacy in patients and caregivers to manage indwelling medical devices at home. Results: The literature review in the analysis phase identified the need for a personalized interface, alarm function, and a community. The design and development phases produced a comprehensive feature list to guide the intervention protocol, along with the creation of a prototype. A pilot study will be conducted to evaluate the feasibility and potential effectiveness of the mobile application, as well as to refine it based on the feedback received. Conclusion: We expect that this application will reduce the burden on patients and caregivers providing home-based care, improve patient health, and reduce the waste of medical resources such as unnecessary hospitalizations.