BMC Medical Informatics and Decision Making最新文献

{"title":"Machine learning model development and validation using SHAP: predicting 28-day mortality risk in pulmonary fibrosis patients.","authors":"Zijun Wu, Mingliang Li, Zhiliang Xu, Gang Liu","doi":"10.1186/s12911-025-03172-8","DOIUrl":"https://doi.org/10.1186/s12911-025-03172-8","url":null,"abstract":"<p><strong>Background: </strong>Early prediction of mortality risk within 28 days of admission is crucial for personalized treatment in patients with pulmonary fibrosis (PF). This study aims to develop a predictive model for 28-day mortality risk in PF patients using interpretable machine learning (ML) methods.</p><p><strong>Methods: </strong>Data from patients with pulmonary fibrosis were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The study endpoint was mortality within 28 days of admission. Feature selection was performed using logistic regression and LASSO algorithms. Six machine learning algorithms-decision tree, k-nearest neighbors (KNN), LightGBM, single-hidden-layer neural network, support vector machine (SVM), and extreme gradient boosting (XGBoost)-were employed to construct risk prediction models. Additionally, SHapley Additive exPlanations (SHAP) were utilized to interpret the predictive models.</p><p><strong>Results: </strong>Among the six evaluated machine learning models, the LightGBM model demonstrated robust predictive performance, with an area under the receiver operating characteristic curve (AUC) of 0.819. SHAP analysis revealed that length of ICU stay, respiratory rate, and white blood cell count were the three most important features for predicting 28-day mortality risk in PF patients, with ICU stay duration having the most significant impact.</p><p><strong>Conclusion: </strong>This study indicates that machine learning methods hold potential for early prediction of mortality risk within 28 days of admission in patients with pulmonary fibrosis. Moreover, SHAP analysis enhanced the interpretability of the LightGBM model, thereby guiding clinical decision-making.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"382"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Electronic medical record (EMR) pre-deployment assessment for adoption readiness, acceptability, and associated factors at Hiwot Fana comprehensive specialized university hospital.","authors":"Admas Abera, Tilahun Shiferaw, Fitsum Yeneneh, Zewdu Alemu, Abdi Amin, Ahmed Mohammed, Adisu Birhanu Weldesenbet, Fila Ahmed, Dereje Weldesilassie, Abebe Tolera","doi":"10.1186/s12911-025-03169-3","DOIUrl":"https://doi.org/10.1186/s12911-025-03169-3","url":null,"abstract":"","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"376"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The development of a decision aid to support treatment choice in pelvic organ prolapse: a Delphi study.","authors":"Larissa Esmeralda Drost, Marjan Stegeman, Janneke van Dijk, Romy E D Lamers, Regina The, Maria B E Gerritse, Arie Franx, M Caroline Vos","doi":"10.1186/s12911-025-03209-y","DOIUrl":"https://doi.org/10.1186/s12911-025-03209-y","url":null,"abstract":"","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"380"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Statistical and machine-learning assessment of attitudinal, knowledge, and perceptual factors on diabetes awareness in Kuwait.","authors":"Ahmad T Al-Sultan, Ahmad Alsaber, Jiazhu Pan, Anwaar Al Kandari, Balqees Alawadhi, Khalida Al-Kenane, Sarah Al-Shamali","doi":"10.1186/s12911-025-03212-3","DOIUrl":"https://doi.org/10.1186/s12911-025-03212-3","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Objectives: &lt;/strong&gt;The primary objective was to identify and analyze the factors that impact diabetes awareness and perception among diabetic and non-diabetic participants. The study also sought to assess the effectiveness of current health awareness programs and identify gaps in public knowledge about diabetes.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Diabetes poses a significant global health challenge, with increasing prevalence worldwide. Comprehending the behavioral and demographic factors leading to diabetes is important for personalized interventions and prevention strategies in Kuwait.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methodology: &lt;/strong&gt;This study was cross-sectional in nature and employed a quantitative approach. It involved distributing a structured questionnaire to a sample of N = 1268 participants in Kuwait, 391 of them were diabetic and 877 were non-diabetic. The sample was stratified based on age, gender, administrative division and nationality. The study employed machine learning and statistical analyses to examine the nature of the relationship between diabetes awareness and the demographic factors. The study executed a random forest approach before employing a logistic regression model to determine the most significant features influencing diabetes. This involved prioritizing variables based on their importance metrics like a mean dropout loss and mean decrease in accuracy, this ensures that the most important predictors are included in the logistic regression model, facilitating a more concentrated and comprehensible examination of the factors affecting diabetes.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The output shown above describes the results for the logistics regression model indicating the different variables that are significant predictors for diabetes among the participants. From the odds ratio it was observed that age was a significant predictor and people above 60 years of age were 11.47 times more likely to have diabetes compared to the 18-30 age group. For those aged 46-60 the likelihood of having diabetes compared to the 18-30 age group was 5.79 times. Similarly, gender was a significant predictor and males were 2.27 times likely to have diabetes than females. Those who frequently interacted with medical staff were also at higher risk (odds of 1.41), likewise, individuals who had kidney complications were also at higher risk of getting diabetes (odds of 1.60). On the contrast, being overweight decreased the odds of getting diabetic (odds ratio of 0.55), likewise, having pregnancy related diabetes decreased the likelihood of being diabetic (odds ratio of 0.65). From these results, it can be seen that age, gender and certain health complications while interacting with the dependent variable need to be considered while assessing the risk of getting diabetes.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;The current study reveals that gender, age groups, kidney disorders and healthcare provider interactions among others, are significantly associat","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"379"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic fall risk prediction in hospitalized cancer patients: development and validation of a machine learning model using multidimensional clinical data to overcome over-sensitivity in traditional scales.","authors":"Huan Zhang, Yifei An, Min Song, Yingtao Meng","doi":"10.1186/s12911-025-03213-2","DOIUrl":"https://doi.org/10.1186/s12911-025-03213-2","url":null,"abstract":"","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"377"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ensemble techniques for predictive modeling of leishmanial activity via molecular fingerprints.","authors":"Saif Nalband, Pallavi Kiratkar, Maulik Gupta, Mansi Gambhir, Surabhi Sonam, Femi Robert, A Amalin Prince","doi":"10.1186/s12911-025-03041-4","DOIUrl":"https://doi.org/10.1186/s12911-025-03041-4","url":null,"abstract":"<p><strong>Background: </strong>Leishmaniasis, a neglected tropical disease caused by Leishmania protozoan parasites and transmitted by sandflies, poses a significant global health challenge, especially in resource-limited environments. The life cycle of the parasite includes crucial amastigote and promastigote stages, each contributing importantly to the infection process. The current therapies for leishmaniasis face limitations due to considerable side effects and the rise of drug-resistant strains, underscoring the pressing need for new, effective, and safe treatment options. Recent advancements in leishmaniasis vaccine development include live attenuated vaccines, recombinant vaccines, and the use of synthetic biology. These approaches aim to induce robust immune responses while ensuring safety. Controlled human infection studies are also being explored to accelerate vaccine development. However, a licensed vaccine remains elusive.</p><p><strong>Method: </strong>This study introduces a novel method for drug discovery targeting leishmaniasis, employing machine learning and cheminformatics to forecast the efficacy of compounds against Leishmania promastigotes. A detailed dataset consisting of 65,057 molecules sourced from the PubChem database is utilized, with the Alamar Blue-based assay applied to assess drug susceptibility. The data encoding relies on molecular fingerprints derived from Simplified Molecular Input Line Entry System (SMILES) notations. We employed three distinct fingerprint algorithms, Avalon, MACCS Key, and Pharmacophore, for the development of machine learning models. Various algorithms, including random forest, multilayer perceptron, gradient boosting, and decision tree, are utilized to create models that effectively classify molecules as either active or inactive based on their structural and chemical characteristics, which could significantly impact the drug discovery process for leishmaniasis.</p><p><strong>Results: </strong>We additionally introduced a model based on ensembles, achieving a peak accuracy of 83.65% and an area under the curve of 0.8367. This study offers significant promise in enhancing drug discovery efforts focused on tackling the global issue of leishmaniasis.</p><p><strong>Conclusion: </strong>Furthermore, the proposed approach has the potential to serve as a framework for addressing other overlooked tropical diseases, offering a promising alternative to conventional drug discovery methods and their associated difficulties.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"378"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The role of four ERS-related key genes in sepsis induced organ dysfunction and the forewarning model construction.","authors":"Yuanqun Zhou, Xinming Xiang, Han She, Yue Wu, Xingnan Ouyang, Xiaowei Zhou, Shunxin Yang, Liangming Liu, Tao Li","doi":"10.1186/s12911-025-03216-z","DOIUrl":"https://doi.org/10.1186/s12911-025-03216-z","url":null,"abstract":"","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"381"},"PeriodicalIF":3.8,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145291280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GPT-4o and the quest for machine learning interpretability in ICU risk of death prediction.","authors":"Moein E Samadi, Kateryna Nikulina, Sebastian Johannes Fritsch, Andreas Schuppert","doi":"10.1186/s12911-025-03224-z","DOIUrl":"https://doi.org/10.1186/s12911-025-03224-z","url":null,"abstract":"<p><strong>Background: </strong>Clinical utilization of machine learning is hampered by the lack of interpretability inherent in most non-linear black box modeling approaches, reducing trust among clinicians and regulators. Advanced large language models offer a potential framework for integrating medical knowledge into these models, potentially enhancing their interpretability.</p><p><strong>Methods: </strong>A hybrid mechanistic/data-driven modeling framework is presented for developing an ICU risk of death prediction model for mechanically ventilated patients. In the mechanistic modeling part, GPT-4o is used to generate detailed medical feature descriptions, which are then aggregated into a comprehensive corpus and processed with TF-I DF vectorization. Fuzzy C-means clustering is subsequently applied to these vectorized features to identify significant mortality cause-specific feature clusters, and a physician reviewed the resulting clusters to validate their relevance to actionable insights for clinical decision support. In the data-driven part, the identified clusters inform the creation of XGBoost-based weak classifiers, whose outcomes are combined into a single XGBoost-based strong classifier through a hierarchically structured feed-forward network. This process results in a novel GPT hybrid model for ICU risk of death prediction.</p><p><strong>Results: </strong>This study enrolled 16,018 mechanically ventilated ICU patients, divided into derivation (12,758) and validation (3,260) cohorts, to develop and evaluate a GPT hybrid model for predicting in-ICU death. Leveraging GPT-4o, we implemented an automated process for clustering mortality cause-specific features, resulting in six feature clusters: Liver Failure, Infection, Renal Failure, Hypoxia, Cardiac Failure, and Mechanical Ventilation. This approach significantly improved upon previous manual methods, automating the reconstruction of structured hybrid models. While the GPT hybrid model showed similar predictive accuracy to a Global XGBoost model, it demonstrated superior interpretability and clinical relevance by incorporating a wider array of features and providing a hierarchical structure of feature importance aligned with medical knowledge.</p><p><strong>Conclusion: </strong>We introduce a novel approach to predicting in-ICU risk of death for mechanically ventilated patients using a GPT hybrid model. Our methodology demonstrates the potential of integrating large language models with traditional machine learning techniques to create interpretable and clinically relevant predictive models.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"373"},"PeriodicalIF":3.8,"publicationDate":"2025-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145285665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"BCECNN: an explainable deep ensemble architecture for accurate diagnosis of breast cancer.","authors":"Uçman Ergün, Tuğçe Çoban, İsmail Kayadibi","doi":"10.1186/s12911-025-03186-2","DOIUrl":"https://doi.org/10.1186/s12911-025-03186-2","url":null,"abstract":"<p><strong>Background: </strong>Breast cancer remains one of the leading causes of cancer-related deaths globally, affecting both women and men. This study aims to develop a novel deep learning (DL)-based architecture, the Breast Cancer Ensemble Convolutional Neural Network (BCECNN), to enhance the diagnostic accuracy and interpretability of breast cancer detection systems.</p><p><strong>Methods: </strong>The BCECNN architecture incorporates two ensemble learning (EL) structures: Triple Ensemble CNN (TECNN) and Quintuple Ensemble CNN (QECNN). These ensemble models integrate the predictions of multiple CNN architectures-AlexNet, VGG16, ResNet-18, EfficientNetB0, and XceptionNet-using a majority voting mechanism. These models were trained using transfer learning (TL) and evaluated on five distinct sub-datasets generated from the Artificial Intelligence Smart Solution Laboratory (AISSLab) dataset, which consists of 266 mammography images labeled and validated by radiologists. To improve transparency and interpretability, Explainable Artificial Intelligence (XAI) techniques, including Gradient-weighted Class Activation Mapping (Grad-CAM) and Local Interpretable Model-Agnostic Explanations (LIME), were applied. Additionally, explainability was assessed through clinical evaluation by an experienced radiologist.</p><p><strong>Results: </strong>Experimental results demonstrated that the TECNN model-comprising AlexNet, VGG16, and EfficientNetB0-achieved the highest accuracy of 98.75% on the AISSLab-v2 dataset. The integration of XAI methods substantially enhanced the interpretability of the model, enabling clinicians to better understand and validate the model's decision-making process. Clinical evaluation confirmed that the XAI outputs aligned well with expert assessments, underscoring the practical utility of the model in a diagnostic setting.</p><p><strong>Conclusion: </strong>The BCECNN model presents a promising solution for improving both the accuracy and interpretability of breast cancer diagnostic systems. Unlike many previous studies that rely on single architectures or large datasets, BCECNN leverages the strengths of an ensemble of CNN models and performs robustly even with limited data. It integrates advanced XAI techniques-such as Grad-CAM and LIME-to provide visual justifications for model decisions, enhancing clinical interpretability. Moreover, the model was validated using AISSLab dataset, designed to reflect real-world diagnostic challenges. This combination of EL, interpretability, and robust performance on small yet clinically relevant data positions BCECNN as a novel and reliable decision support tool for AI-assisted breast cancer diagnostics.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"374"},"PeriodicalIF":3.8,"publicationDate":"2025-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145285687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"I-ETL: an interoperability-aware health (meta)data pipeline to enable federated analyses.","authors":"Nelly Barret, Anna Bernasconi, Boris Bikbov, Pietro Pinoli","doi":"10.1186/s12911-025-03188-0","DOIUrl":"https://doi.org/10.1186/s12911-025-03188-0","url":null,"abstract":"<p><strong>Background: </strong>Clinicians are interested in better understanding complex diseases, such as cancer or rare diseases, so they need to produce and exchange data to mutualize sources and join forces. To do so and ensure privacy, a natural way consists in using a decentralized architecture and Federated Learning algorithms. This ensures that data stays in the organization in which it has been collected, but requires data to be collected in similar settings and similar models. In practice, this is often not the case because healthcare institutions work individually with different representations and raw data; they do not have means to normalize their data, and even less to do so across centers. For instance, clinicians have at hand phenotypic, clinical, imaging and genomic data (each individually collected) and want to better understand some diseases by analyzing them together. This example highlights the needs and challenges for a cooperative use of this wealth of information.</p><p><strong>Methods: </strong>We designed and implemented a framework, named I-ETL, for integrating highly heterogeneous healthcare datasets of hospitals in interoperable databases. Our proposal is twofold: ([Formula: see text]) we devise two general and extensible conceptual models for modeling both data and metadata and ([Formula: see text]) we propose an Extract-Transform-Load (ETL) pipeline ensuring and assessing interoperability from the start.</p><p><strong>Results: </strong>By conducting experiments on open-source datasets, we show that I-ETL succeeds in representing various health datasets in a unified way thanks to our two general conceptual models. Next, we demonstrate the importance of blending interoperability as a first-class citizen in integration pipelines, ensuring possible collaboration between different centers.</p><p><strong>Conclusion: </strong>As a framework, I-ETL contributes to integrate and improve interoperability between healthcare institutions. When used in a decentralized federated platform, it eases the federated analysis of the different hospital databases and helps clinicians to obtain insights and knowledge on medical conditions of interest.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"375"},"PeriodicalIF":3.8,"publicationDate":"2025-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145285628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}