BMC Medical Informatics and Decision Making最新文献

{"title":"Expert-augmented machine learning for predicting extubation readiness in the pediatric intensive care unit.","authors":"Jean Digitale, Deborah Franzon, Jin Ge, Charles McCulloch, Mark J Pletcher, Efstathios D Gennatas","doi":"10.1186/s12911-025-03070-z","DOIUrl":"10.1186/s12911-025-03070-z","url":null,"abstract":"<p><strong>Background: </strong>Determining extubation readiness in pediatric intensive care units (PICU) is challenging. We used expert-augmented machine learning (EAML), a method that combines machine learning with human expert knowledge, to predict successful extubation.</p><p><strong>Methods: </strong>We extracted electronic health record data from patients in two PICUs. Data from patients in one unit was split into 80% training and 20% test, while patients in the other served as an external test set. EAML begins by training RuleFit, which converts gradient-boosted trees into decision rules. Then, expert clinicians were asked to assess the relative probability of successful extubation of the subgroup defined by each rule compared with the entire sample. The rules were ranked in order of increasing chance of successful extubation according to (1) the RuleFit model and (2) clinician assessment, and differences between the two rankings were calculated. The initial RuleFit model was then regularized based on these differences, producing the EAML model.</p><p><strong>Results: </strong>The RuleFit model selected 46 rules; we surveyed 25 clinician experts to provide feedback on them. All clinicians worked in a PICU setting and were from multiple disciplines; over half (56%) had > 5 years of PICU experience. As expected, the added regularization slightly lowered performance of EAML compared with RuleFit in the internal test set, although the difference was not statistically significant (RuleFit AUC = 0.817 vs. best-performing EAML model AUC = 0.814, difference = 0.003, 95% CI of difference = -0.009, 0.003). EAML had superior performance in the external test set (RuleFit AUC = 0.791 vs. best-performing EAML model AUC = 0.799, difference = 0.007, 95% CI of difference = 0.002, 0.013).</p><p><strong>Conclusions: </strong>When creating a model to predict successful extubation in PICU patients, incorporating expert knowledge directly into the model construction process via EAML produced a model more generalizable to an external test set.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"232"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220236/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing GPT and DeepL for terminology translation in the medical domain: A comparative study on the human phenotype ontology.","authors":"Richard Noll, Alexandra Berger, Dominik Kieu, Tobias Mueller, Ferdinand O Bohmann, Angelina Müller, Svea Holtz, Philipp Stoffers, Sebastian Hoehl, Oya Guengoeze, Jan-Niklas Eckardt, Holger Storf, Jannik Schaaf","doi":"10.1186/s12911-025-03075-8","DOIUrl":"10.1186/s12911-025-03075-8","url":null,"abstract":"<p><strong>Background: </strong>This paper presents a comparative study of two state-of-the-art language models, OpenAI's GPT and DeepL, in the context of terminology translation within the medical domain.</p><p><strong>Methods: </strong>This study was conducted on the human phenotype ontology (HPO), which is used in medical research and diagnosis. Medical experts assess the performance of both models on a set of 120 translated HPO terms and their 180 synonyms, employing a 4-point Likert scale (strongly agree = 1, agree = 2, disagree = 3, strongly disagree = 4). An independent reference translation from the HeTOP database was used to validate the quality of the translation.</p><p><strong>Results: </strong>The average Likert rating for the selected HPO terms was 1.29 for GPT-3.5 and 1.37 for DeepL. The quality of the translations was also found to be satisfactory for multi-word terms with greater ontological depth. The comparison with HeTOP revealed a high degree of similarity between the models' translations and the reference translations.</p><p><strong>Conclusions: </strong>Statistical analysis revealed no significant differences in the mean ratings between the two models, indicating their comparable performance in terms of translation quality. The study not only illustrates the potential of machine translation but also shows incomplete coverage of translated medical terminology. This underscores the relevance of this study for cross-lingual medical research. However, the evaluation methods need to be further refined, specific translation issues need to be addressed, and the sample size needs to be increased to allow for more generalizable conclusions.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"237"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220062/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144538993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mortality predicting models for patients with infective endocarditis: a machine learning approach.","authors":"Yang Zi-Yang, Wang Qi, Xingyan Liu, Haolin Li, Shouhong Wang, Danqing Yu, Xuebiao Wei","doi":"10.1186/s12911-025-03025-4","DOIUrl":"10.1186/s12911-025-03025-4","url":null,"abstract":"<p><strong>Background: </strong>Infective endocarditis (IE) is a fatal cardiovascular disease with varied clinical manifestations but rapid progression. A series of existing risk models helped identify IE patients with high risk, but the imperfect predictive performance and limited application called for better predictive systems.</p><p><strong>Methods: </strong>The single-centered, retrospective observational study applied four machine learning methods for predictive model construction: LASSO logistic regression, random forest (RF), support vector machine (SVM), and k-nearest neighbors (KNN). A 10-fold cross-validated area under the receiver operating characteristic curve (AUC-ROC) was used for performance evaluation.</p><p><strong>Results: </strong>A total of 1705 patients with IE were enrolled in the study, with 119 in-hospital deaths and 178 deaths after 6-month follow-up. RF achieved the highest AUC-ROCs for in-hospital and six-month mortality prediction (in-hospital: 0.83, 6-month: 0.85). RF was also applied to assess variable importance. The following variables were selected by RF as top important predictors for both in-hospital and six-month mortality prediction: total bilirubin, N-terminal pro-B-type natriuretic peptide, albumin, diastolic blood pressure, fasting blood glucose, uric acid, and age.</p><p><strong>Conclusions: </strong>A risk model with machine learning approach was integrated in purpose of prognosis prediction in IE patients, helping rapid risk stratification and in-time management clinically.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"229"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220579/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of an ensemble prediction model for acute graft-versus-host disease in allogeneic transplantation based on machine learning.","authors":"Lin Song, Xingwei Wu, Mengjia Xu, Ling Xue, Xun Yu, Zongqi Cheng, Chenrong Huang, Liyan Miao","doi":"10.1186/s12911-025-03059-8","DOIUrl":"10.1186/s12911-025-03059-8","url":null,"abstract":"<p><strong>Background: </strong>Acute graft-versus-host disease (aGVHD) is a major post-transplantation complication and one of the most significant causes of non-relapse-related death. However, the massive and complex clinical data make aGVHD difficult to predict. Machine learning (ML), a branch of artificial intelligence, has since been introduced in medicine due to its ability to process complex, high-dimensional variables quickly and capture nonlinear relationships. However, the effects of immunosuppressants exposure was not considered in previous ML models. Thus, the purpose of this study was to develop and optimize models by Cox regression and machine learning algorithms to predict the risk of aGVHD in which cyclosporin A exposure and common clinical factors were included as variables.</p><p><strong>Methods: </strong>The data was preprocessed in the first step, and was randomly allocated at an 8:2 ratio. Cox regression model was constructed on the training set. Meanwhile, correlation analysis and recursive feature elimination were used for feature screening before machine learning model development. Then fifteen algorithms were used to establish models, and an ensemble model was established through soft voting based on the top five performance algorithms. Area under curve (AUC) was the main metric used to evaluate the model performance in the validation set, while nomogram and SHAP were applied to interpret the variables.</p><p><strong>Result: </strong>A total of 479 patients and 47 variables were included in the study. The incidence of grade II-IV aGVHD was 33.61%. The AUC of Cox regression model in the validation set was 0.625. In contrast, the new ensemble model has a better prediction ability (AUC = 0.776, Accuracy = 0.729, Precision = 0.667, Recall = 0.375, F1-score = 0.480). Except for the variables which were identified by previous studies, some rarely reported risk factors were found, such as quinolone, blood urea nitrogen and alkaline phosphatase.</p><p><strong>Conclusions: </strong>In summary, a new ensemble model with promising accuracy was established to predict grade II-IV classic aGVHD in allo-HSCT patients. It will help identify high-risk patients at an early stage and thus reduce the incidence of aGVHD.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"234"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12219984/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A deep learning model for predicting systemic lupus erythematosus-associated epitopes.","authors":"Jiale He, Zixia Liu, Xiaopo Tang","doi":"10.1186/s12911-025-03056-x","DOIUrl":"10.1186/s12911-025-03056-x","url":null,"abstract":"<p><strong>Background: </strong>The accurate prediction of epitopes associated with Systemic Lupus Erythematosus (SLE) plays a vital role in advancing our understanding of autoimmune pathogenesis and in designing effective immunotherapeutics. Traditional bioinformatics methods often struggle to capture the intricate sequence patterns and high-dimensional signals characteristic of epitope data. Deep learning presents a compelling alternative, with its ability to perform automatic feature learning and model complex dependencies inherent in biological sequences. This study proposes a hybrid deep learning architecture that synergistically integrates handcrafted biochemical features with data-driven deep sequence modeling to improve the identification of SLE-associated epitopes.</p><p><strong>Methods: </strong>The framework comprises six interconnected components: (1) handcrafted feature extraction encoding biochemical and physicochemical attributes; (2) an embedding layer for dense sequence representation; (3) a Convolutional Neural Network (CNN) branch that captures local patterns from handcrafted features; (4) a Long Short-Term Memory branch for learning temporal dependencies in sequence data; (5) a scaled dot-product attention-based fusion module that integrates complementary information from both branches; and (6) a Multi-Layer Perceptron for final classification. Model evaluation employed metrics such as Accuracy, Precision, Recall, F1-score, and the area under the receiver operating characteristic curve (ROCAUC).</p><p><strong>Results: </strong>The hybrid model outperformed both baseline machine learning algorithms and ablated versions of itself. It achieved a ROCAUC of 0.9506 and an F1-score of 0.8333 on the SLE epitope prediction task. Notably, ablation studies revealed that the CNN component had the most substantial influence on performance, while the custom fusion mechanism yielded better integration of features than conventional strategies. These findings underscore the model's robustness and capacity to generalize across complex epitope prediction tasks.</p><p><strong>Conclusion: </strong>This work presents an interpretable, biologically informed deep learning approach for predicting SLE-associated epitopes. By merging domain-specific handcrafted features with dynamic deep learning representations, the model not only enhances predictive accuracy but also provides meaningful biological insights. The framework holds promise for broader applications in immunoinformatics and autoimmune disease research.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"230"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220259/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Health information system in primary health care units of the Central Zone, Tigray, Northern Ethiopia.","authors":"Letekirstos Gebreegziabher Gebretsadik, Abate Bekele Belachew, Gebregziabher Berihu Gebrekidan, Alemayohu Bayray, Akberet Lemlem, Lewtnesh Berihun Dangew, Haftom Temesgen Abebe","doi":"10.1186/s12911-025-03078-5","DOIUrl":"10.1186/s12911-025-03078-5","url":null,"abstract":"<p><strong>Background: </strong>Health information systems require the management of health information through health management information systems and research and knowledge management. In many low-income countries, including Ethiopia, poor data quality and limited use of health information remain major challenges in the health system. Reliable health data quality is essential for evidence-based decision-making and improving quality health service delivery. This study aimed to assess and explore the contextual factors of the quality and utilization of health information in primary health care units in the Central Zone, Tigray, Northern Ethiopia.</p><p><strong>Methods: </strong>A facility-based cross-sectional quantitative and qualitative study design was used. A total of seven primary health care units and four district health offices were selected. Data were collected via document review, structured questionnaires and in-depth interviews. A three-month document review was conducted to assess data accuracy via lot quality assurance sampling. Forty-eight health professionals, including Woreda Health Office heads, facility heads, health management information system focal persons, service providers and health extension workers, were interviewed for quantitative analysis. Additionally, 23 key informants with the same roles participated in the qualitative interviews. Descriptive statistics were computed, and thematic analysis was conducted for the qualitative data.</p><p><strong>Results: </strong>Four of the seven primary health care units have assigned health management information system personnel, and five of them have necessary equipment for health management information systems. The average lot quality assurance sampling of the primary health care units ranged from 35 to 60%, which falls below the national threshold of 90% data accuracy. In knowing and measuring the dimensions of data quality, the informants described this as a difficult task despite acknowledging its importance. Similarly, the culture of data use for decision making was limited.</p><p><strong>Conclusion: </strong>This study revealed that primary health care units in the Central Zone of Tigray face significant challenges in terms of data quality and utilization, primarily due to the limited capacity of service providers, unclear understanding of data quality dimensions and weak data use culture. The average lot quality assurance sampling accuracy rates are below the acceptable level, indicating issues in the data documentation and validation processes. Addressing these gaps through targeted capacity-building, including the integration of HMIS curricula at the university level and system-level improvements such as implementing computerized systems, ensuring accountability and allocating budgets, is needed to strengthen health information systems and enable evidence-based decision-making at all levels of the health system.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"233"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220773/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the performance of artificial intelligence-based speech recognition for clinical documentation: a systematic review.","authors":"Joel Jia Wei Ng, Eugene Wang, Xinyan Zhou, Kevin Xiang Zhou, Charlene Xing Le Goh, Gabriel Zheng Ning Sim, Hiang Khoon Tan, Serene Si Ning Goh, Qin Xiang Ng","doi":"10.1186/s12911-025-03061-0","DOIUrl":"10.1186/s12911-025-03061-0","url":null,"abstract":"<p><strong>Background: </strong>Clinical documentation is vital for effective communication, legal accountability and the continuity of care in healthcare. Traditional documentation methods, such as manual transcription, are time-consuming, prone to errors and contribute to clinician burnout. AI-driven transcription systems utilizing automatic speech recognition (ASR) and natural language processing (NLP) aim to automate and enhance the accuracy and efficiency of clinical documentation. However, the performance of these systems varies significantly across clinical settings, necessitating a systematic review of the published studies.</p><p><strong>Methods: </strong>A comprehensive search of MEDLINE, Embase, and the Cochrane Library identified studies evaluating AI transcription tools in clinical settings, covering all records up to February 16, 2025. Inclusion criteria encompassed studies involving clinicians using AI-based transcription software, reporting outcomes such as accuracy (e.g., Word Error Rate), time efficiency and user satisfaction. Data were extracted systematically, and study quality was assessed using the QUADAS-2 tool. Due to heterogeneity in study designs and outcomes, a narrative synthesis was performed, with key findings and commonalities reported.</p><p><strong>Results: </strong>Twenty-nine studies met the inclusion criteria. Reported word error rates ranged widely, from 0.087 in controlled dictation settings to over 50% in conversational or multi-speaker scenarios. F1 scores spanned 0.416 to 0.856, reflecting variability in accuracy. Although some studies highlighted reductions in documentation time and improvements in note completeness, others noted increased editing burdens, inconsistent cost-effectiveness and persistent errors with specialized terminology or accented speech. Recent LLM-based approaches offered automated summarization features, yet often required human review to ensure clinical safety.</p><p><strong>Conclusions: </strong>AI-based transcription systems show potential to improve clinical documentation but face challenges in accuracy, adaptability and workflow integration. Refinements in domain-specific training, real-time error correction and interoperability with electronic health records are critical for their effective adoption in clinical practice. Future research should also focus on next-generation \"digital scribes\" incorporating LLM-driven summarization and repurposing of text.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"236"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220090/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multiclass skin lesion classification and localziation from dermoscopic images using a novel network-level fused deep architecture and explainable artificial intelligence.","authors":"Mehak Arshad, Muhammad Attique Khan, Nouf Abdullah Almujally, Areej Alasiry, Mehrez Marzougui, Yunyoung Nam","doi":"10.1186/s12911-025-03051-2","DOIUrl":"10.1186/s12911-025-03051-2","url":null,"abstract":"<p><strong>Background and objective: </strong>Early detection and classification of skin cancer are critical for improving patient outcomes. Dermoscopic image analysis using Computer-Aided Diagnostics (CAD) is a powerful tool to assist dermatologists in identifying and classifying skin lesions. Traditional machine learning models require extensive feature engineering, which is time-consuming and less effective in handling complex data like skin lesions. This study proposes a deep learning-based network-level fusion architecture that integrates multiple deep models to enhance the classification and localization of skin lesions in dermoscopic images. The goal is to address challenges like irregular lesion shapes, inter-class similarities, and class imbalances while providing explainability through artificial intelligence.</p><p><strong>Methods: </strong>A novel hybrid contrast enhancement technique was applied for pre-processing and dataset augmentation. Two deep learning models, a 5-block inverted residual network and a 6-block inverted bottleneck network, were designed and fused at the network level using a depth concatenation approach. The models were trained using Bayesian optimization for hyperparameter tuning. Feature extraction was performed with a global average pooling layer, and shallow neural networks were used for final classification. Explainable AI techniques, including LIME, were used to interpret model predictions and localize lesion regions. Experiments were conducted on two publicly available datasets, HAM10000 and ISIC2018, which were split into training and testing sets.</p><p><strong>Results: </strong>The proposed fused architecture achieved high classification accuracy, with results of 91.3% and 90.7% on the HAM10000 and ISIC2018 datasets, respectively. Sensitivity, precision, and F1-scores were significantly improved after data augmentation, with precision rates of up to 90.91%. The explainable AI component effectively localized lesion areas with high confidence, enhancing the model's interpretability.</p><p><strong>Conclusions: </strong>The network-level fusion architecture combined with explainable AI techniques significantly improved the classification and localization of skin lesions. The augmentation and contrast enhancement processes enhanced lesion visibility, while fusion of models optimized classification accuracy. This approach shows potential for implementation in CAD systems for skin cancer diagnosis, although future work is required to address the limitations of computational resource requirements and training time.</p><p><strong>Clinical trail number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"215"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12211947/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Framework for key benchmarking indicators in hospital information system.","authors":"Asghar Ehteshami, Ahamad-Reza Raeisi, Maedeh Rashedi, Yasin Sabet Kouhanjani","doi":"10.1186/s12911-025-03038-z","DOIUrl":"10.1186/s12911-025-03038-z","url":null,"abstract":"<p><strong>Introduction: </strong>Developing key benchmarking indicators (KBIs) for Hospital Information Systems (HIS) has the potential to enhance operational efficiency and effectiveness, resulting in the successful achievement of hospital goals. This study endeavored to develop a comprehensive framework of KBIs for HIS to address the need for structured evaluation in healthcare systems.</p><p><strong>Methods: </strong>This qualitative study was conducted at Ahvaz Jundishapour University of Medical Sciences' academic hospitals. The study comprised two distinct phases. In the initial phase, data collection involved purposive sampling with a snowball technique, consisting of 14 semi-structured interviews with health information technology, medical informatics, and health information management experts. During the second phase, benchmarking indicators were extracted through conventional content analysis and prioritized using two rounds of the Delphi technique.</p><p><strong>Results: </strong>The study identified three main themes and eight sub-themes, including Technical (software, hardware, architecture and user interface, and outputs), Human and Organizational (vendors, IT support, and HIS workflows), and Financial (costs). A total of 76 KBIs were developed and prioritized, with key indicators such as user-friendliness and response time scoring 100% in importance, while others like cost-effectiveness and interoperability quality scored 96.4% and 98.8%, respectively.</p><p><strong>Conclusion: </strong>The proposed framework incorporates essential benchmarking indicators intended to enhance HIS efficiency, effectiveness, and successful attainment of its objectives. These indicators may serve as a comprehensive tool for Health Information Technology managers to benchmark their HISs. Practical implications include the potential for hospital administrators to identify gaps in system performance, optimize resource allocation, and improve user satisfaction by applying the KBIs. The framework may also guide decision-making processes, supporting alignment with organizational goals and promoting the long-term sustainability of HIS implementations.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"213"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12210575/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 image reconstruction and adaptive statistical iterative reconstruction on coronary artery calcium scoring in high risk population for coronary heart disease.","authors":"Lijuan Zhu, Xiaomeng Shi, Lusong Tang, Haruhiko Machida, Lili Yang, Meixiang Ma, Ruoshui Ha, Yun Shen, Fang Wang, Dazhi Chen","doi":"10.1186/s12911-025-03049-w","DOIUrl":"10.1186/s12911-025-03049-w","url":null,"abstract":"<p><strong>Objective: </strong>Deep learning image reconstruction (DLIR) technology effectively improves the image quality while maintaining spatial resolution. The impact of DLIR on the quantification of coronary artery calcium (CAC) is still unclear. The purpose of this study was to investigate the effect of DLIR on the quantification of coronary calcium in high-risk populations.</p><p><strong>Methods: </strong>A retrospective study was conducted on patients who underwent coronary artery CT angiography (CCTA) at our hospital(China) from February 2022 to September 2022. Raw data were reconstructed with filtered back projection (FBP) reconstruction, 40% and 80% level adaptive statistical iterative reconstruction-veo (ASiR-V 40%, ASiR-V 80%) and low, medium and high-level deep learning algorithm (DLIR-L, DLIR-M, and DLIR-H). Calculate and compare the signal-to-noise and contrast-to-noise ratio, volumetric score, mass scores, and Agaston score of 6 sets of images.</p><p><strong>Results: </strong>There were 178 patients, female (107), mean age (62.43 ± 9.26), and mean BMI (25.33 ± 3.18) kg/m². Compared with FBP, the image noise of ASiR-V and DLIR was significantly reduced (P < 0.001). There was no significant difference in Agaston score, volumetric score, and mass scores among the six reconstruction algorithms (all P > 0.05). Bland-Altman diagram indicated that the Agatston scores of the five reconstruction algorithms showed good agreement with FBP, with DLIR-L(AUC, 110.08; 95% CI: 26.48, 432.92;)and ASIR-V40% (AUC,110.96; 95% CI: 26.23, 431.34;) having the highest consistency with FBP.</p><p><strong>Conclusion: </strong>Compared with FBP, DLIR and ASiR-V improve CT image quality to varying degrees while having no impact on Agatston score-based risk stratification.</p><p><strong>Clinical relevance statement: </strong>CACS is a powerful tool for cardiovascular risk stratification, and DLIR can improve image quality without affecting CACS, making it widely applicable in clinical practice.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"212"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12210892/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}