International Journal of Medical Informatics最新文献

{"title":"Utilization, challenges, and training needs of digital health technologies: Perspectives from healthcare professionals","authors":"Ruby Khan ,&nbsp;Sumbal Khan ,&nbsp;Hailah M. Almohaimeed ,&nbsp;Amany I. Almars ,&nbsp;Bakht Pari","doi":"10.1016/j.ijmedinf.2025.105833","DOIUrl":"10.1016/j.ijmedinf.2025.105833","url":null,"abstract":"<div><div><strong>Introduction:</strong> Digital health technology (DHTs), such as electronic health records (EHRs), mobile health apps, and remote monitoring systems, is revolutionizing contemporary healthcare by improving diagnosis, patient care, and operational efficiency. Notwithstanding these developments, infrastructure, technical assistance, and personnel training remain obstacles to the successful deployment of DHTs.</div><div><strong>Methods:</strong> 500 medical experts participated in a survey to evaluate the use, advantages, and challenges of DHTs. The frequency of DHT use, the perceived advantages, and the challenges—such as technical difficulties and a lack of training—were the main topics of the data gathered.</div><div><strong>Results:</strong> The most popular technology was mobile health apps (44.4%), followed by EHR systems and diagnostic tools (33.3%). Benefits reported included decreased administrative burden (50%) and increased diagnostic accuracy (46.2%). However, there significant obstacles were found, though: 63% of respondents said they had only received limited training, and 51.9% mentioned software bugs and network problems. Despite these obstacles, 63% of those surveyed reported increases in the effectiveness of healthcare delivery.</div><div><strong>Discussion:</strong> Our study finds a gap between the infrastructure needed for DHTs to be implemented successfully and their quick adoption. This study challenges the notion that adopting technology alone increases productivity by highlighting the importance of thorough technical assistance and staff training. These issues need to be resolved if DHTs are to be fully utilized for improved healthcare delivery and operational effectiveness.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"197 ","pages":"Article 105833"},"PeriodicalIF":3.7,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143403335","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":"The fading structural prominence of explanations in clinical studies","authors":"Daniele Roberto Giacobbe,&nbsp;Matteo Bassetti","doi":"10.1016/j.ijmedinf.2025.105835","DOIUrl":"10.1016/j.ijmedinf.2025.105835","url":null,"abstract":"","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"197 ","pages":"Article 105835"},"PeriodicalIF":3.7,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143402525","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 the importance of clinical and sociodemographic factors on self-rated health in midlife: A cross-sectional study using machine learning","authors":"Hisrael Passarelli-Araujo","doi":"10.1016/j.ijmedinf.2025.105834","DOIUrl":"10.1016/j.ijmedinf.2025.105834","url":null,"abstract":"<div><h3>Background</h3><div>Self-rated health (SRH) is influenced by various factors, including clinical and sociodemographic characteristics. However, in the context of Brazil, we still lack a clear understanding of the relative importance of these factors and how they differ between men and women in midlife. Given the significance of gender equity in health, it is crucial to explore these differences to meet the specific needs of each group.</div></div><div><h3>Objective</h3><div>This study examines the importance of clinical and sociodemographic factors of SRH among middle-aged Brazilian adults and analyzes how they vary between men and women.</div></div><div><h3>Methods</h3><div>A cross-sectional analysis was conducted using data from the 2019 National Health Survey (PNS) with a representative sample of 31,926 middle-aged adults (40–59 years) living in private households on Brazilian territory. Five machine learning models—Naive Bayes, SVM, Logistic Regression, Random Forests, and XGBoost—were employed to analyze the data.</div></div><div><h3>Results</h3><div>The analysis revealed gender-specific patterns in SRH predictors. For men, education was the most critical factor, followed by diagnoses of physical and mental illnesses. For women, SRH was primarily influenced by chronic disease diagnoses, low education, and health insurance coverage. Alcohol consumption was a stronger predictor of poor SRH for men than women, likely due to cultural norms that promote higher alcohol use among men.</div></div><div><h3>Conclusion</h3><div>This study provides insights into midlife health disparities, emphasizing gender-specific factors influencing SRH. Machine learning demonstrated its value in uncovering nuanced patterns in health data, offering a powerful tool for public health research and policy.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105834"},"PeriodicalIF":3.7,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143388314","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":"Serious Games for constipation management for people with intellectual disabilities: A scoping review and narrative synthesis","authors":"Serena Daniel ,&nbsp;Ruth Bishop ,&nbsp;Ellie Killner ,&nbsp;Alison Whight ,&nbsp;Sarah Lennard ,&nbsp;Stephen Howard ,&nbsp;Richard Laugharne ,&nbsp;Rohit Shankar","doi":"10.1016/j.ijmedinf.2025.105832","DOIUrl":"10.1016/j.ijmedinf.2025.105832","url":null,"abstract":"<div><h3>Introduction</h3><div>People with intellectual disability (PwID) are 2% of the UK population. Constipation and bowel movement (BM) problems (diarrhoea/faecal incontinence etc.) affects over a third of PwID and is a serious cause of morbidity and mortality. Pw ID rely heavily on outside support (family/professional carers/healthcare professionals), many of whom are ignorant to bowel related harms. There is significant stigma to discuss BM particularly constipation.</div><div>Serious Games (SG) are increasingly used for education of health needs. This review examines if game-based technology can assist improving knowledge and reducing stigma of BM problems particularly constipation.</div></div><div><h3>Objective</h3><div>To identify and gain evidence of SGs aimed at improving knowledge of BM management particularly constipation.</div></div><div><h3>Methods</h3><div>A systematic search of publications between 2010 and 2024 was conducted following the PRISMA ScR statement for scoping reviews. The search inclusion/exclusion criteria were designed and overseen by an information specialist. PUBMED, EMBASE and PsychINFO databases were searched. Extracted variables included SG title, co-production and expert involvement, target outcome, evaluation methodology, effectiveness, sustainability and game platform. Results were narratively synthesised.</div></div><div><h3>Results</h3><div>Of 2966 papers retrieved, three were selected for inclusion, none RCTs. All three included SGs aimed to teach BM management or recognition to healthcare workers/ professionals. Two studies evaluated game efficacy. No SGs were assessed after initial trials, none were implemented in clinical practice. Only one game successfully improved participant knowledge. All game creators consulted experts during game design, but none consulted patients. None discussed reducing stigma amongst their audience.</div></div><div><h3>Conclusion</h3><div>Only one of three SGs identified improved BM knowledge in healthcare workers/professionals and was not specific to PwID. There is potential to co-produce with PwID and their carers a SG to support BM problems particularly constipation to reduce stigma, improve outcomes and be a templar for other similarly vulnerable groups like those with dementia.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105832"},"PeriodicalIF":3.7,"publicationDate":"2025-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing cancer data collection: ICD-11 implementation in European oncology and cancer registries","authors":"Maciej Trojanowski ,&nbsp;Irmina Maria Michalek ,&nbsp;Anna Kubiak ,&nbsp;Łukasz Taraszkiewicz ,&nbsp;Witold Kycler","doi":"10.1016/j.ijmedinf.2025.105821","DOIUrl":"10.1016/j.ijmedinf.2025.105821","url":null,"abstract":"<div><h3>Background</h3><div>The International Classification of Diseases (ICD) is a cornerstone in medical data standardisation worldwide. The newly released ICD-11, with its updated chapter on neoplasms, is expected to revolutionise cancer data collection, particularly in European healthcare system. However, significant challenges to implementation remain, especially when transitioning from ICD-10, which has long underpinned cancer registration. This study aimed to assess the feasibility of implementing ICD-11 in European cancer registries and healthcare system by examining its coding structure for neoplasms, particularly its ability to replace ICD-10 in existing cancer data collection processes.</div></div><div><h3>Methods</h3><div>The analysis focused on ICD-11 Chapter 2 “Neoplasms”, including pre-coordination and post-coordination coding structures. A comprehensive mapping of ICD-10 to ICD-11 codes was performed for cancer site, morphology, grade, stage, laterality and prognostic factors. Additionally, tools for coding validation and their applications in registries and hospital IT systems were reviewed. Results were assessed for accuracy, consistency, and ease of implementation.</div></div><div><h3>Results</h3><div>ICD-11 allows for more detailed coding of neoplasms, especially through post-coordination, which enhances precision in capturing tumour subtypes and prognostic factors. Nevertheless, inconsistencies were observed in grading and staging systems, allowing for invalid combinations. Mapping between ICD-10 and ICD-11 also revealed gaps, particularly in representing complex cancer diagnoses.</div></div><div><h3>Conclusion</h3><div>While ICD-11 presents advanced coding options for cancer data, its complexity and the lack of validation mechanisms present challenges for immediate use in cancer registries. The transition from ICD-10 will necessitate extensive training, improved mapping tools, and the introduction of validation systems to ensure accurate data collection. Cancer registries are well-suited to support this transition, but further refinements to ICD-11 are essential before full adoption.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"197 ","pages":"Article 105821"},"PeriodicalIF":3.7,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143479763","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":"Multimodal convolutional neural networks for the prediction of acute kidney injury in the intensive care","authors":"R. van Slobbe ,&nbsp;D. Herrmannova ,&nbsp;D.J. Boeke ,&nbsp;E.S. Lima-Walton ,&nbsp;A. Abu-Hanna ,&nbsp;I. Vagliano","doi":"10.1016/j.ijmedinf.2025.105815","DOIUrl":"10.1016/j.ijmedinf.2025.105815","url":null,"abstract":"<div><div>Increased monitoring of health-related data for ICU patients holds great potential for the early prediction of medical outcomes. Research on whether the use of clinical notes and concepts from knowledge bases can improve the performance of prediction models is limited. We investigated the effects of combining clinical variables, clinical notes, and clinical concepts. We focus on the early prediction of Acute Kidney Injury (AKI) in the intensive care unit (ICU). AKI is a sudden reduction in kidney function measured by increased serum creatinine (SCr) or decreased urine output. AKI may occur in up to 30% of ICU stays. We developed three models based on convolutional neural networks using data from the Medical Information Mart for Intensive Care (MIMIC) database. The models used clinical variables, free-text notes, and concepts from the Elsevier H-Graph. Our models achieved good predictive performance (AUROC 0.73-0.90). These models were assessed both when using Scr and urine output as predictors and when omitting them. When Scr and urine output were used as predictors, models that included clinical notes and concepts together with clinical variables performed on par with models that only used clinical variables. When excluding SCr and urine output, predictive performance improved by combining multiple modalities. The models that used only clinical variables were externally validated on the eICU dataset and transported fairly to the new population (AUROC 0.68-0.77). Our in-depth comparison of modalities and text representations may further guide researchers and practitioners in applying multimodal models for predicting AKI and inspire them to investigate multimodality and contextualized embeddings for other tasks. Our models can support clinicians to promptly recognize and treat deteriorating AKI patients and may improve patient outcomes in the ICU.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105815"},"PeriodicalIF":3.7,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143226609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluate the effect of virtual nurse-guided discharge education app on disease knowledge and symptom response in patients following coronary events","authors":"Ling Zhang ,&nbsp;Robyn Gallagher ,&nbsp;Huiyun Du ,&nbsp;Tracey Barry ,&nbsp;Jon Foote ,&nbsp;Tiffany Ellis ,&nbsp;Aarti Gulyani ,&nbsp;Robyn A. Clark","doi":"10.1016/j.ijmedinf.2025.105818","DOIUrl":"10.1016/j.ijmedinf.2025.105818","url":null,"abstract":"<div><h3>Background</h3><div>Pre-discharge patient education promotes better self-care and secondary prevention following acute coronary syndrome (ACS). Traditional methods do not adapt well to staff and patient time limitations and varied health literacy levels. Self-administered digital methods using engagement strategies may address these issues.</div></div><div><h3>Objectives</h3><div>To evaluate whether a co-designed, self-administered, virtual nurse avatar-guided patient education app can improve ACS knowledge, beliefs, and medication adherence, and be acceptable for patients and nurses.</div></div><div><h3>Methods</h3><div>A prospective pre-post-test study was used with patients recruited during hospitalisation for ACS and their associated nursing staff. Patients, alongside usual care, were provided with the education app on a tablet at discharge to use immediately and over the following month. Data were collected immediately following use and one-month post on heart disease knowledge, ACS symptom response attitudes and beliefs and medication adherence. User satisfaction data was collected from both patients and nurses.</div></div><div><h3>Results</h3><div>Participants included nurses (n = 22) and patients (n = 22) who were diagnosed with ST-elevation myocardial infarction (STEMI) (73 %), aged mean 59.7 years and 40 % had not completed high school.</div><div>Patients’ heart disease knowledge improved from pre to one-month post-use (15.7 vs 17.0; p &lt; 0.001) and from immediately post to one-month post-use (16.3 vs 17.0; p = 0.003). Patients’ ACS symptom knowledge and response beliefs improved from pre- to immediate post-use (13.8 vs 15.5; p = 0.008; 23.8 vs 25.1; p = 0.038), and to one-month post-use (13.8 vs 17.0; p &lt; 0.001; 23.8 vs 25.7; p = 0.025), and ACS symptom response attitudes improved from pre- to one-month post-use (15.8 vs 17.0; p = 0.036).</div><div>Patients and nurses rated the app’s presentation, content, usability, and usefulness highly; 86% of nurses thought the app would help with discharge education.</div></div><div><h3>Conclusion</h3><div>A co-designed, self-administered, virtual nurse avatar-guided education app can improve heart disease knowledge, attitudes, and beliefs following ACS with high nurse and patient acceptability.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105818"},"PeriodicalIF":3.7,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143226558","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":"A comparative analysis of trauma-related mortality in South Korea using classification models","authors":"Yookyung Boo ,&nbsp;Youngjin Choi","doi":"10.1016/j.ijmedinf.2025.105805","DOIUrl":"10.1016/j.ijmedinf.2025.105805","url":null,"abstract":"<div><h3>Background</h3><div>Reducing mortality among severe trauma patients requires the establishment of an effective emergency transportation system and the rapid transfer of patients to appropriate medical facilities. Machine learning offers significant potential to enhance the efficiency and quality of these emergency medical services.</div></div><div><h3>Methods</h3><div>A retrospective secondary analysis was conducted using region-specific trauma survey data. The analysis focused on socio-economic characteristics, mechanisms of injury, injury severity, and variables indicating the effectiveness of the emergency medical system in optimizing machine learning algorithms for predicting severe patient transportation decisions.</div></div><div><h3>Results</h3><div>Among the 8,769 patients with severe trauma, 7.2 % died in the hospital, with an average age of 50.06 years. The average injury severity score was 8.44, and the average time from accident reporting to arrival at the emergency medical facility was 55.39 min. The trend showed that as the level of the emergency medical institution increased, the patient transport time increased, while the mortality rate decreased. Additionally, XGBoost showed the best performance in mortality classification using a dataset sampled with SMOTE-ENN. Although the difference was minimal, undersampling slightly outperformed oversampling in the classification of emergency patients.</div></div><div><h3>Conclusion</h3><div>The treatment of emergency patients is influenced not only by transport time but also by the resources and staff levels of specialized emergency medical centers, which in turn affect survival rates. Furthermore, given the superior performance of composite sampling methods in analyzing imbalanced datasets, the importance of considering such imbalanced datasets in the analysis is evident.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105805"},"PeriodicalIF":3.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143226606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effectiveness of digital health interventions for chronic conditions management in European primary care settings: Systematic review and meta-analysis","authors":"Elisa Ambrosi ,&nbsp;Elisabetta Mezzalira ,&nbsp;Federica Canzan ,&nbsp;Chiara Leardini ,&nbsp;Giovanni Vita ,&nbsp;Giulia Marini ,&nbsp;Jessica Longhini","doi":"10.1016/j.ijmedinf.2025.105820","DOIUrl":"10.1016/j.ijmedinf.2025.105820","url":null,"abstract":"<div><h3>Background</h3><div>The past decade has seen rapid digitalization of healthcare, significantly transforming healthcare delivery. However, the impact of these technologies remains unclear, with notable gaps in evidence regarding their effectiveness, especially in primary care settings.</div></div><div><h3>Objective</h3><div>This systematic review assesses the effectiveness of digital health interventions versus interventions without digital components implemented over the last 10 years in European primary care settings for managing chronic diseases.</div></div><div><h3>Methods</h3><div>Following Cochrane guidelines, we conducted a systematic review with <em>meta</em>-analysis. We searched multiple databases for randomized controlled trials. Inclusion criteria encompassed studies on digital health interventions for chronic disease management in primary care settings in Europe, evaluating outcomes such as hospitalizations, quality of life, and clinical measures. Data extraction and quality assessment were independently conducted by two authors, with discrepancies resolved by a third author. The certainty of the evidence was judged according to the Grading of Recommendations, Assessment, Development, and Evaluation approach.</div></div><div><h3>Results</h3><div>From 9829 records, 23 studies were included, with most studies conducted in the UK and Spain. The most investigated conditions were type 2 diabetes and hypertension. Interventions mainly focused on patient monitoring, self-care education, and digital communication tools. The risk of bias was low to moderate for most studies. Meta-analyses showed no significant differences between digital health interventions and usual care for hospitalizations, depressive symptoms, anxiety, HbA1c, diastolic blood pressure, weight, or quality of life, except for a small improvement in systolic blood pressure.</div></div><div><h3>Conclusion</h3><div>Digital health interventions have not yet demonstrated substantial benefits over traditional care for chronic disease management in European primary care. While some improvements were noted, particularly in systolic blood pressure, the impact remains limited. Further research is needed to enhance the effectiveness of digital health interventions, address current methodological limitations, and explore tailored approaches for both specific patient populations and multimorbid populations.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105820"},"PeriodicalIF":3.7,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning and machine learning in CT-based COPD diagnosis: Systematic review and meta-analysis","authors":"Qian Wu,&nbsp;Hui Guo,&nbsp;Ruihan Li,&nbsp;Jinhuan Han","doi":"10.1016/j.ijmedinf.2025.105812","DOIUrl":"10.1016/j.ijmedinf.2025.105812","url":null,"abstract":"<div><h3>Background</h3><div>With advancements in medical technology and science, chronic obstructive pulmonary disease (COPD), one of the world’s three major chronic diseases, has seen numerous remarkable outcomes when combined with artificial intelligence, particularly in disease diagnosis. However, the diagnostic performance of these AI models still lacks comprehensive evidence. Therefore, this study quantitatively analyzed the diagnostic performance of AI models in CT images of COPD patients, aiming to promote the development of related research in the future.</div></div><div><h3>Methods</h3><div>PubMed, Cochrane Library, Web of Science, and Embase were retrieved up to September 1, 2024. The QUADAS-2 evaluation tool was used to assess the quality of the included studies. Meta-analysis of the included researches was performed using Stata18, RevMan 5.4, and Meta-Disc 1.4 software to merge sensitivity, specificity and plot a summary receiver operating characteristic curve (SROC). Heterogeneity was assessed using the Q statistic, and sources of inter-study heterogeneity were explored through <em>meta</em>-regression analysis.</div></div><div><h3>Results</h3><div>Twenty-two of 3280 identified studies were eligible. Meta-analysis was performed on 15 of these studies, encompassing a total of 22,817 patients for which statistical metrics were reported or could be calculated. Seven studies were based on deep learning (DL) model, three on machine learning (ML) model, and five on DL model with multiple-instance learning (MIL) mechanisms. One study evaluated both DL and ML models. The <em>meta</em>-analysis results showed that the pooled sensitivity of all DL and ML models was 86 % (95 %CI 78–91 %), specificity was 87 % (95 %CI 83–91 %), and area under the curve was 93 % (95 %CI 90–95 %). Subgroup analyses revealed no significant difference in diagnostic sensitivity and specificity between DL and ML models (sensitivity 82 % (95 %CI 76–87 %), 93 % (95 %CI 85–97 %); specificity 87 % (95 %CI 79–91 %), 84 % (95 %CI 79–88 %), and the DL model with MIL (sensitivity 87 % (95 %CI 61–96 %); specificity 89 % (95 %CI 78–95 %) improved the performance of DL model, but this improvement was not statistically significant (p &gt; 0.05).</div></div><div><h3>Conclusion</h3><div>Both DL and ML models for diagnosing COPD using CT images exhibited high accuracy. There was no significant difference in diagnostic efficacy between the two types of AI models, and the addition of the MIL mechanism may enhance the performance of the DL model.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"196 ","pages":"Article 105812"},"PeriodicalIF":3.7,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143076329","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}