International Journal of Medical Informatics最新文献

{"title":"The actual performance of large language models in providing liver cirrhosis-related information: A comparative study","authors":"Yanqiu Li ,&nbsp;Zhuojun Li ,&nbsp;Jinze Li ,&nbsp;Long Liu ,&nbsp;Yao Liu ,&nbsp;Bingbing Zhu ,&nbsp;Ke shi ,&nbsp;Yu Lu ,&nbsp;Yongqi Li ,&nbsp;Xuanwei Zeng ,&nbsp;Ying Feng ,&nbsp;Xianbo Wang","doi":"10.1016/j.ijmedinf.2025.105961","DOIUrl":"10.1016/j.ijmedinf.2025.105961","url":null,"abstract":"<div><h3>Objective</h3><div>With the increasing prevalence of large language models (LLMs) in the medical field, patients are increasingly turning to advanced online resources for information related to liver cirrhosis due to its long-term management processes. Therefore, a comprehensive evaluation of real-world performance of LLMs in these specialized medical areas is necessary.</div></div><div><h3>Methods</h3><div>This study evaluates the performance of four mainstream LLMs (ChatGPT-4o, Claude-3.5 Sonnet, Gemini-1.5 Pro, and Llama-3.1) in answering 39 questions related to liver cirrhosis. The information quality, readability and accuracy were assessed using Ensuring Quality Information for Patients tool, Flesch-Kincaid metrics and consensus scoring. The simplification and their self-correction ability of LLMs were also assessed.</div></div><div><h3>Results</h3><div>Significant performance differences were observed among the models. Gemini scored highest in providing high-quality information. While the readability of all four LLMs was generally low, requiring a college-level reading comprehension ability, they exhibited strong capabilities in simplifying complex information. ChatGPT performed best in terms of accuracy, with a “Good” rating of 80%, higher than Claude (72%), Gemini (49%), and Llama (64%). All models received high scores for comprehensiveness. Each of the four LLMs demonstrated some degree of self-correction ability, improving the accuracy of initial answers with simple prompts. ChatGPT’s and Llama’s accuracy improved by 100%, Claude’s by 50% and Gemini’s by 67%.</div></div><div><h3>Conclusion</h3><div>LLMs demonstrate excellent performance in generating health information related to liver cirrhosis, yet they exhibit differences in answer quality, readability and accuracy. Future research should enhance their value in healthcare, ultimately achieving reliable, accessible and patient-centered medical information dissemination.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105961"},"PeriodicalIF":3.7,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143912439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SupportPrim CDSS: A clinical decision support system architecture based on microservices for non-specific musculoskeletal disorders","authors":"Amar Jaiswal ,&nbsp;Mohit Kumar ,&nbsp;Ingebrigt Meisingset","doi":"10.1016/j.ijmedinf.2025.105919","DOIUrl":"10.1016/j.ijmedinf.2025.105919","url":null,"abstract":"<div><h3>Background and Objective</h3><div>Non-specific musculoskeletal disorders (MSDs) pose significant challenges in primary care due to ambiguous symptoms and diverse etiologies. This research presents the SupportPrim clinical decision support system (CDSS), an innovative approach that combines case-based reasoning (CBR) with a scalable microservice framework, aiming to improve personalized treatment and clinical decision processes in MSD care.</div></div><div><h3>Methods</h3><div>The SupportPrim CDSS is engineered using a modular microservice architecture designed for scalability, reliability, and seamless clinical integration. Subjective patient-reported questionnaires and demographic data are processed through an optimized CBR engine that retrieves precedent cases to inform current clinical decisions. The system leverages rigorous evaluation through iterative experiments and a randomized controlled trial (RCT) in Norwegian primary care, thereby assessing its usability, clinical utility, and operational performance.</div></div><div><h3>Results</h3><div>The system demonstrates high reliability, characterized by negligible downtime and a mean case retrieval response time of 0.18 seconds. Clinicians reported favorable user interactions, emphasizing the system's ability to facilitate shared decision making and personalized care. While the SupportPrim study intentionally maintained a static casebase, the system possesses the ability to incorporate active learning to boost adaptability and precision. Extensive validation and verification from associated studies confirm considerable performance of both the CBR engine and the CDSS.</div></div><div><h3>Conclusion</h3><div>The SupportPrim CDSS effectively leverages CBR within a microservice-based framework to aid clinicians in delivering evidence-based, personalized patient care for patients with non-specific MSDs. Its robust design, coupled with comprehensive verification and validation across multiple associated studies, underscores its potential for broader healthcare applications and improved clinical decision support.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105919"},"PeriodicalIF":3.7,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143912438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identifying and prioritizing competencies for health informatics master’s graduates to support the health sector transformation program in Saudi Arabia","authors":"Bader Alshoumr ,&nbsp;Jwaher A. Almulhem ,&nbsp;Khulud Alkadi ,&nbsp;Lujain Aljarallah ,&nbsp;Lobna A. Aljuffali ,&nbsp;Raniah N. Aldekhyyel","doi":"10.1016/j.ijmedinf.2025.105944","DOIUrl":"10.1016/j.ijmedinf.2025.105944","url":null,"abstract":"<div><h3>Background</h3><div>The 2030 Health Sector Transformation Program (HSTP) in Saudi Arabia includes digital health as one of its objectives. To achieve this transformation a knowledgeable and skillful workforce is needed. No research identifies the specific health informatics competencies needed to support HSTP’s objectives. Our study aims to identify and prioritize key competencies for health informatics master’s graduates needed to support Saudi Arabia’s digital health transformation strategies for 2030.</div></div><div><h3>Methods</h3><div>A series of semi-structured interviews were conducted with 17 health informatics experts in Saudi Arabia, identified through LinkedIn, with over 10 years of experience working on large-scale national projects. Interviewees were conducted in August and September of 2024. Experts ranked competencies based on their priority and alignment with HSTP’s digital transformation goals. The competencies were drawn from of the International Medical Informatics Association (IMIA) international recommendations in biomedical and health informatics (BMHI) educational framework, which served as the gold standard.</div></div><div><h3>Results</h3><div>A total of 37 competencies were ranked as high priority, 38 as moderate priority, and 5 as neutral. The top high-priority competencies for health informatics master’s graduates included change management, patient safety, data and information analysis, system security, business alignment, ethics, security and privacy, and leadership. The management science and social and behavioral domains were identified as the most critical for developing health informatics curricula and professional training programs. Two new competencies were identified: innovation, and emerging digital health technologies.</div></div><div><h3>Conclusion</h3><div>The findings reflect the changes that the healthcare system in the country is experiencing specifically related to data and digitalization as identified by the 2030 HSTP. There needs to be more standardized educational programs focused on the competencies needed for the workforce to contribute to the digital health transformation plans. These findings can serve as a standard guide in revising or establishing BMHI educational programs.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105944"},"PeriodicalIF":3.7,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143906815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"“I don’t know”: An uncertainty-aware machine learning model for predicting patient disposition at emergency department triage","authors":"Abubakar Sadiq Bouda Abdulai ,&nbsp;Jean Storm ,&nbsp;Michael Ehrlich","doi":"10.1016/j.ijmedinf.2025.105957","DOIUrl":"10.1016/j.ijmedinf.2025.105957","url":null,"abstract":"<div><h3>Background</h3><div>Machine learning (ML) models are widely used for predicting patient disposition at emergency department (ED) triage. However, these models generate predictions regardless of the level of uncertainty, potentially leading to overconfident outputs that can compromise clinical decision-making.</div></div><div><h3>Objective</h3><div>To develop a conformal prediction model for ED triage that provides uncertainty-aware patient disposition predictions.</div></div><div><h3>Methods</h3><div>This retrospective study analyzed 560,486 adult ED visits (March 2014 – July 2017) from one academic and two community hospitals. An extreme gradient boosting (XGBoost) model was trained, validated, and conformalized to introduce a “Don’t know” prediction for high-uncertainty cases. The model was tested on a random sample of 56,000 ED cases.</div></div><div><h3>Results</h3><div>The standard XGBoost model achieved an AUC of 0.9307 (95% CI: 0.9285 – 0.9329), with sensitivity of 0.72 and specificity of 0.94. With conformal prediction at a lower confidence threshold of 60%, the model indicated “Don’t know” in 4.9% of cases while returning sensitivity and specificity values of 0.74 and 0.95, respectively. As confidence thresholds increased, the model returned more “Don’t know” predictions and fewer misclassifications. At 90% confidence, the model returned “Don’t know” in 34.5% of cases while returning sensitivity and specificity values of 0.88 and 0.99, respectively. This trade-off highlights a balance between model confidence and prediction accuracy<strong>.</strong></div></div><div><h3>Conclusion</h3><div>Incorporating uncertainty-awareness in ML models improves reliability in ED triage. By acknowledging uncertainty, clinicians receive more interpretable insights, reducing the risk of overconfident predictions and enhancing patient safety.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105957"},"PeriodicalIF":3.7,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using longitudinal data and deep learning models to enhance resource allocation in home-based medical care","authors":"Ling Chen ,&nbsp;Ching-Po Lin ,&nbsp;Chi-Hua Chung ,&nbsp;Jason Jiunshiou Lee","doi":"10.1016/j.ijmedinf.2025.105953","DOIUrl":"10.1016/j.ijmedinf.2025.105953","url":null,"abstract":"<div><h3>Background</h3><div>The aging population is driving increased healthcare demands and costs, prompting the need for effective home healthcare programs. Accurate patient assessment is essential for optimizing resource allocation and tailoring services.</div></div><div><h3>Objective</h3><div>This retrospective study explores the application of artificial intelligence (AI) in predicting home medical care stages to enhance care delivery.</div></div><div><h3>Methods</h3><div>Data from Taipei City Hospital (2015–2021) included inpatient, outpatient, and home medical care records. Three deep learning (DL) models—Transformer encoder-based, long short-term memory (LSTM), and gated recurrent unit (GRU)—were compared with three baseline machine learning (ML) models. Models were trained on 3, 5, and 10 consecutive visits for binary and multiclass classification. Performance was evaluated using accuracy, precision, recall, and the area under the receiver operating characteristic curve (AUC).</div></div><div><h3>Results</h3><div>The study included 4,343 patients with a mean age of 85.04 ± 11.47 years. While models trained on 10 visits generally exhibited higher performance, data from 5 visits were sufficient for accurate predictions. With five visits, the LSTM model achieved the highest AUC (0.908) for distinguishing between the absence (S0) and presence (S1–S3) of home medical care. Meanwhile, the Transformer achieved the best AUC (0.86) for classifying S0–S3, with individual stage AUCs of 0.90, 0.82, 0.81, and 0.94 for S0, S1, S2, and S3, respectively.</div></div><div><h3>Conclusions</h3><div>AI deep learning models show strong potential for accurately predicting home medical care stages. The best-performing model could be a promising tool for healthcare professionals to optimize resource allocation in home medical care settings.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105953"},"PeriodicalIF":3.7,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Utilizing machine learning for predicting mortality in patients with heat-related illness who visited the emergency department","authors":"Wan-Yin Kuo ,&nbsp;Chien-Cheng Huang ,&nbsp;Chung-Feng Liu ,&nbsp;Mei-I Sung ,&nbsp;Chien-Chin Hsu ,&nbsp;Hung-Jung Lin ,&nbsp;Shih-Bin Su ,&nbsp;How-Ran Guo","doi":"10.1016/j.ijmedinf.2025.105951","DOIUrl":"10.1016/j.ijmedinf.2025.105951","url":null,"abstract":"<div><h3>Background</h3><div>In the context of climate change and global warming, heat-related illness (HRI) is anticipated to escalate and become a major concern. Patients with severe HRI primarily present to the emergency department (ED), but there are no prediction tools for mortality in HRI patients who visit ED. The objective of this study was to use machine learning approaches to establish prediction models for mortality in patients with HRI who visit ED.</div></div><div><h3>Methods</h3><div>We included all patients aged 20 and above with a final diagnosis of HRI who visited the EDs of three hospitals (Chi Mei Medical Center, Chi Mei Hospital Liouying, and Chi Mei Hospital Chiali) between January 2010 and October 2021. Patients who had transferred to other hospitals or had insufficient data were excluded. A total of 11 predictive feature variables were used in the algorithms. The primary outcome was in-hospital mortality or impending death discharge. We used machine learning algorithms including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), Light Gradient Boosting Machine (LightGBM), Multilayer Perceptron (MLP), and eXtreme Gradient Boosting (XGBoost) to establish prediction models for mortality in such patients. Accuracy, sensitivity, specificity, and area under curve (AUC) were used as indicators to evaluate the performance of prediction models.</div></div><div><h3>Results</h3><div>Out of the 820 HRI patients included in the analysis, 1.5% had mortality. All six prediction models had a high AUC, ranging from 0.825 to 0.991, and LightGBM which included peripheral oxygen saturation (SpO₂₎ and Glasgow Coma Scale (GCS) score on arrival as the two main features had the highest AUC. The accuracy, sensitivity, and specificity of LightGBM were 0.976, 1.000 and 0.975, respectively.</div></div><div><h3>Conclusion</h3><div>Machine learning-based prediction models are promising tools in accurately predicting mortality in HRI patients who present to the ED.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105951"},"PeriodicalIF":3.7,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring Multimorbidity Patterns in older hospitalized Norwegian patients using Network Analysis modularity","authors":"Mohsen Askar,&nbsp;Beate Hennie Garcia,&nbsp;Kristian Svendsen","doi":"10.1016/j.ijmedinf.2025.105954","DOIUrl":"10.1016/j.ijmedinf.2025.105954","url":null,"abstract":"<div><h3>Background</h3><div>Understanding Multimorbidity Patterns (MPs) is crucial for planning healthcare interventions, allocating resources, and improving patients’ outcomes.</div></div><div><h3>Objective</h3><div>We aim to demonstrate the use of Network Analysis (NA) to explore the MPs in hospitalized Norwegian older patients.</div></div><div><h3>Methods</h3><div>We utilized data from the Norwegian Patient Registry (NPR) of all admissions between 2017 and 2019. The study population included patients ≥ 65 years old with two or more different conditions. Multimorbidity was defined as the co-occurrence of two or more associated chronic conditions. Chronic conditions were identified using the Chronic Condition Indicator Refined (CCIR) list. The association between chronic conditions was determined by calculating Relative Risk (RR) and Phi-correlation to detect pairs of conditions that co-occur beyond chance. A multimorbidity network was created, and MPs were detected using Louvain method for community detection. We suggested a clinical interpretation for these MPs.</div></div><div><h3>Results</h3><div>A total of 539 chronic conditions were used to create a multimorbidity network revealing several MPs. These modules included patterns of vision and hearing disorders, cardiorenal syndrome, metabolic and cardiovascular disorders, respiratory disorders, endocrine and skin conditions, autoimmune and musculoskeletal disorders, as well as mental and behavioral disorders. Using NA centrality measures, we identified the most influential conditions in each module. An interactive network and sunburst graphs for each module are publicly available.</div></div><div><h3>Conclusion</h3><div>The study demonstrates the use of NA modularity detection in identifying MPs. The findings highlight the complex interaction of chronic conditions in the elderly and the potential of NA methodology in exploring these relationships.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105954"},"PeriodicalIF":3.7,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143882255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Relationship prediction between clinical subtypes and prognosis of critically ill patients with cirrhosis based on unsupervised learning methods: A study from two critical care databases","authors":"Shu Zhang ,&nbsp;Jie Li ,&nbsp;Ying Chen ,&nbsp;Shan Xu","doi":"10.1016/j.ijmedinf.2025.105952","DOIUrl":"10.1016/j.ijmedinf.2025.105952","url":null,"abstract":"<div><h3>Background</h3><div>Our objective was to identify distinct clinical subtypes among critically ill patients with cirrhosis and analyze the clinical features and prognosis of each subtype.</div></div><div><h3>Methods</h3><div>We extracted routine clinical data within 24 h of ICU admission from the MIMIC-IV database. To determine the number of clinical subtypes, we employed the “elbow method,” “cumulative distribution function (CDF) plot,” and “consensus matrix.” Consensus k-means, k-means, and SOM methods were used to identify different clinical subtypes of critically ill cirrhosis. We validated our findings using patients from the eICU database. The SHapley Additive exPlanations (SHAP) method was used to explore the features of each clinical subtype, and 28-day Kaplan-Meier curves were generated. Survival differences among the clinical subtypes were assessed using the log-rank test.</div></div><div><h3>Results</h3><div>Our study included 2,586 patients from the MIMIC-IV database and 1,670 patients from the eICU database. Based on the clinical routine variables, we identified three clinical subtypes among patients in the MIMIC-IV database. Subtype A (N = 1424, 55.07 %) was labeled the “common subtype” and exhibited the lowest mortality. Subtype B (N = 703, 27.18 %) was classified as the “hyperinflammatory response subtype” and had a relatively high mortality. Subtype C (N = 459, 17.75 %) was identified as the “liver dysfunction subtype” and had the highest mortality. These findings were consistent with the results obtained from both the internal validation set (MIMIC-IV database) and the external validation set (eICU database).</div></div><div><h3>Conclusions</h3><div>Our study presents a novel and clinically applicable approach for subtyping critically ill cirrhosis.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105952"},"PeriodicalIF":3.7,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143903749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Readability, accuracy and appropriateness and quality of AI chatbot responses as a patient information source on root canal retreatment: A comparative assessment","authors":"Mine Büker,&nbsp;Gamze Mercan","doi":"10.1016/j.ijmedinf.2025.105948","DOIUrl":"10.1016/j.ijmedinf.2025.105948","url":null,"abstract":"<div><h3>Aim</h3><div>This study aimed to assess the readability, accuracy, appropriateness, and overall quality of responses generated by large language models (LLMs), including ChatGPT-3.5, Microsoft Copilot, and Gemini (Version 2.0 Flash), to frequently asked questions (FAQs) related to root canal retreatment.</div></div><div><h3>Methods</h3><div>Three LLM chatbots—ChatGPT-3.5, Microsoft Copilot, and Gemini (Version 2.0 Flash)—were assessed based on their responses to 10 patient FAQs. Readability was analyzed using seven indices, including Flesch reading ease score (FRES), Flesch-Kincaid grade level (FKGL), Simple Measure of Gobbledygook (SMOG), gunning FOG (GFOG), Linsear Write (LW), Coleman-Liau (CL), and automated readability index (ARI), and compared against the recommended sixth-grade reading level. Response quality was evaluated using the Global Quality Scale (GQS), while accuracy and appropriateness were rated on a five-point Likert scale by two independent reviewers. Statistical analyses were conducted using one-way ANOVA, Tukey or Games-Howell post-hoc tests for continuous variables. Spearman’s correlation test was used to assess associations between categorical variables.</div></div><div><h3>Results</h3><div>All chatbots generated responses exceeding the recommended readability level, making them suitable for readers at or above the 10th-grade level. No significant difference was found between ChatGPT-3.5 and Microsoft Copilot, while Gemini produced significantly more readable responses (p &lt; 0.05). Gemini demonstrated the highest proportion of accurate (80 %) and high-quality responses (80 %) compared to ChatGPT-3.5 and Microsoft Copilot.</div></div><div><h3>Conclusions</h3><div>None of the chatbots met the recommended readability standards for patient education materials. While Gemini demonstrated better readability, accuracy, and quality, all three models require further optimization to enhance accessibility and reliability in patient communication.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105948"},"PeriodicalIF":3.7,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143877235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing immunization in pediatric oncology: Development and evaluation of an automated scheduling tool","authors":"Dominik Wawrzuta ,&nbsp;Sylwia Giefert ,&nbsp;Justyna Klejdysz","doi":"10.1016/j.ijmedinf.2025.105950","DOIUrl":"10.1016/j.ijmedinf.2025.105950","url":null,"abstract":"<div><h3>Background</h3><div>Pediatric cancer patients often struggle to adhere to vaccination schedules, requiring personalized catch-up immunization after treatment. Currently, healthcare professionals create these schedules manually, which is time-consuming, error-prone, and limited in accessibility. We developed and validated an automated clinical decision support system to optimize this process through algorithmic schedule generation.</div></div><div><h3>Methods</h3><div>Following PRISMA guidelines, we systematically reviewed vaccination recommendations for pediatric cancer patients to develop comprehensive scheduling algorithms. The system implements multi-step optimization modeling incorporating age-dependent dosing, minimum intervals, and treatment-related timing rules for 27 vaccines. We validated the tool through a technical evaluation comparing algorithm-generated schedules against those created by experts for representative pediatric cancer cases and through clinical assessments by healthcare professionals in real-world settings.</div></div><div><h3>Results</h3><div>Technical validation revealed high agreement between algorithm-generated and expert-created schedules, with only minor acceptable variations. Clinical evaluation demonstrated strong user acceptance, with 70% (95% CI: 58.1%–81.9%) of professionals using the system for schedule validation, 61% (95% CI: 48.3%–73.7%) for schedule creation, and 47% (95% CI: 34.0%–60.0%) for vaccination history verification. Users reported median time savings of 20 min per schedule and rated the tool’s quality highly (median 4.5/5).</div></div><div><h3>Conclusions</h3><div>Our clinical decision support system demonstrates technical accuracy and clinical usefulness. It offers an effective solution for managing post-cancer treatment immunizations while alleviating healthcare professionals’ workloads, particularly in resource-limited settings.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"201 ","pages":"Article 105950"},"PeriodicalIF":3.7,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}