BMC Medical Informatics and Decision Making最新文献

{"title":"Risk factors and nomogram model for short-term postoperative complications in patients with hirschsprung disease.","authors":"Aohua Song, Bobin Zhang, Wei Feng, Jinping Hou, Xiaohong Die, Yi Wang, Zhenhua Guo","doi":"10.1186/s12911-025-03053-0","DOIUrl":"10.1186/s12911-025-03053-0","url":null,"abstract":"","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"214"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12211763/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539050","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":"Implicit bias in ICU electronic health record data: measurement frequencies and missing data rates of clinical variables.","authors":"Junming Shi, Alan E Hubbard, Nicholas Fong, Romain Pirracchio","doi":"10.1186/s12911-025-03058-9","DOIUrl":"10.1186/s12911-025-03058-9","url":null,"abstract":"<p><strong>Background: </strong>Systematic disparities in data collection within electronic health records (EHRs), defined as non-random patterns in the measurement and recording of clinical variables across demographic groups, can be reflective of underlying implicit bias and may affect patient outcome. Identifying and mitigating these biases is critical for ensuring equitable healthcare. This study aims to develop an analytical framework for measurement patterns, defined as the combination of measurement frequency (how often variables are collected) and missing data rates (the frequency of missing recordings), evaluate the association between them and demographic factors, and assess their impact on in-hospital mortality prediction.</p><p><strong>Methods: </strong>We conducted a retrospective cohort study using the Medical Information Mart for Intensive Care III (MIMIC-III) database, which includes data on over 40,000 ICU patients from Beth Israel Deaconess Medical Center (2001-2012). Adult patients with ICU stays longer than 24 h were included. Measurement patterns, including missing data rates and measurement frequencies, were derived from EHR data and analyzed. Targeted Machine Learning (TML) methods were used to assess potential systematic disparities in measurement patterns across demographic factors (age, gender, race/ethnicity) while controlling for confounders such as other demographics and disease severity. The predictive power of measurement patterns on in-hospital mortality was evaluated.</p><p><strong>Results: </strong>Among 23,426 patients, significant demographic systematic disparities were observed in the first 24 h of ICU stays. Elderly patients (≥ 65 years) had more frequent temperature measurements compared to younger patients, while males had slightly fewer missing temperature measurements than females. Racial disparities were notable: White patients had more frequent blood pressure and oxygen saturation (SpO2) measurements compared to Black and Hispanic patients. Measurement patterns were associated with ICU mortality, with models based solely on these patterns achieving an area under the receiver operating characteristic curve (AUC) of 0.76 (95% CI: 0.74-0.77).</p><p><strong>Conclusions: </strong>This study underscores the significance of measurement patterns in ICU EHR data, which are associated with patient demographics and ICU mortality. Analyzing patterns of missing data and measurement frequencies provides valuable insights into patient monitoring practices and potential systemic disparities in healthcare delivery. Understanding these disparities is critical for improving the fairness of healthcare delivery and developing more accurate predictive models in critical care settings.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"241"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220764/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539034","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":"Exploring the possibilities and limitations of customized large language model to support and improve cervical cancer screening.","authors":"Viola Angyal, Ádám Bertalan, Péter Domján, Elek Dinya","doi":"10.1186/s12911-025-03088-3","DOIUrl":"10.1186/s12911-025-03088-3","url":null,"abstract":"<p><strong>Background: </strong>The rapid advancement of artificial intelligence, driven by Generative Pre-trained Transformers (GPT), has transformed natural language processing. Prompt engineering plays a key role in guiding model outputs effectively. Our primary objective was to explore the possibilities and limitations of a custom GPT, developed via prompt engineering, as a patient education tool, which delivers publicly available information through a user-friendly design that facilitates more effective access to cervical cancer screening knowledge.</p><p><strong>Method: </strong>The system was developed using the OpenAI GPT-4 model and Python programming language, with the interface built on Streamlit for cloud-based accessibility and testing. It initially presented questions to testers for preliminary assessment. For cervical cancer-related information, we referenced medical guidelines. Iterative testing optimized the prompts for quality and relevance; techniques like context provision, question chaining, and prompt-based constraints were used. Human-in-the-loop and two independent medical doctor evaluations were employed. Additionally, system performance metrics were measured.</p><p><strong>Result: </strong>The web application was tested 115 times over a three-week period in 2024, with 87 female (76%) and 28 male (24%) participants. A total of 112 users completed the user experience questionnaire. Statistical analysis showed a significant association between age and perceived personalization (p = 0.047) and between gender and system customization (p = 0.037). Younger participants reported higher engagement, though not significantly. Females valued guidance on screening schedules and early detection, while males highlighted the usefulness of information regarding HPV vaccination and its role in preventing HPV-related cancers. Independent evaluations by medical doctors demonstrated consistent assessments of the system's responses in terms of accuracy, clarity, and usefulness.</p><p><strong>Discussion: </strong>While the system demonstrates potential to enhance public health awareness and promote preventive behaviors, encouraging individuals to seek information on cervical cancer screening and HPV vaccination, its conversational capabilities remain constrained by the inherent limitations of current language model technology.</p><p><strong>Conclusions: </strong>Although custom GPTs can not substitute a healthcare consultations, these tools can streamline workflows, expedite information access, and support personalized care. Further research should focus on conducting well-designed randomized controlled trials to establish definitive conclusions regarding its impact and reliability.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"242"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12220158/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539030","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":"The impact of a patient decision aid for patients with advanced laryngeal carcinoma - a multicenter study.","authors":"Anne N Heirman, Japke F Petersen, Abrahim Al-Mamgani, Simone E J Eerenstein, Bertram J de Kleijn, Frank Hoebers, Bernard M Tijink, Lisette van der Molen, Gyorgy B Halmos, Richard Dirven, Martijn M Stuiver, Michiel W M van den Brekel","doi":"10.1186/s12911-025-03080-x","DOIUrl":"10.1186/s12911-025-03080-x","url":null,"abstract":"<p><strong>Purpose: </strong>Patients with advanced larynx cancer face challenging treatment decisions. To address this, we developed and tested a patient decision aid (PDA), aiming to reduce decisional conflict (DC), and enhance knowledge and perceived shared decision-making (SDM).</p><p><strong>Methods: </strong>In this multicenter study (ClinicalTrials.gov ID: NCT03292341, 2016-2023), a pre/post study design was used. Participants, meeting the inclusion criteria of advanced larynx cancer without distant metastasis, completed questionnaires on knowledge, DC and SDM immediately after counseling (T1) and 6 months post-treatment (T2). The intervention arm utilized the PDA (see https://beslissamen.nl/pda_launch.html?pda=tools/pda_larynx_en/story.html ) before completing T1 questionnaires, while the usual care arm followed standard procedures. Between-group differences in outcomes were estimated using regression models with correction for case mix differences.</p><p><strong>Results: </strong>Total DC score was significantly lower in the intervention arm (n = 46) compared to the usual care arm (n = 45) (adjusted mean difference - 32, 95% CI: -37.4; -26.1, p < 0.001). The intervention group demonstrated significantly higher overall knowledge (mean 69% correct) than the control group (mean 47% correct)(adjusted mean difference 24, 95% CI 15.3; 33.1, p < 0.001). Almost all patients in usual care (44/45, 98%) experienced clinically significant DC (CSDC, DCS > 25), compared to 89% (41/46) in the intervention arm (adjusted OR 0.25, 95%CI 0.01; 1.9) p = 0.238). Perceived SDM was significant higher in the intervention arm (mean 78.16) compared to the usual care arm (mean 70.32); however, both groups exhibited high levels.</p><p><strong>Conclusion: </strong>The PDA for advanced laryngeal cancer effectively reduced decisional conflict, enhanced patients' knowledge and improved perceived SDM.</p><p><strong>Trial registration: </strong>ClinicalTrials.gov ID NCT03292341, 20,151,231.</p><p><strong>Level of evidence: 3: </strong></p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"217"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12210726/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144539051","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":"Can we ever have evidence-based decision making in orthopaedics? A qualitative evidence synthesis and conceptual framework.","authors":"Arabella Scantlebury, Katherine Jones, Joy Adamson, Melissa Harden, Catriona McDaid, Amy Grove","doi":"10.1186/s12911-025-03032-5","DOIUrl":"10.1186/s12911-025-03032-5","url":null,"abstract":"<p><strong>Background: </strong>The perception and use of scientific evidence in orthopaedic surgical decision-making is variable, and there is considerable variation in practice. A previous conceptual framework described eight different drivers of orthopaedic surgical decision-making: formal codified and managerial knowledge, medical socialisation, cultural, normative and political influence, training and formal education, experiential factors, and individual patient and surgeon factors. This Qualitative Evidence Synthesis (QES) aims to refine the conceptual framework to understand how these drivers of decision-making are applied to orthopaedic surgical work in a dynamic and fluid way.</p><p><strong>Methods: </strong>A QES explored how different types of knowledge and evidence inform decision-making to explore why there is so much variation in orthopaedic surgical work. Nine databases were systematically searched from 2014 to 2023. Screening was undertaken independently by two researchers. Data extraction and quality assessment were undertaken by one researcher and accuracy checked by another. Findings were mapped to the conceptual framework and expanded through thematic synthesis.</p><p><strong>Results: </strong>Twenty-five studies were included. Our re-conceptualised framework of evidence-based orthopaedics portrays how surgeons undergo a constant process of medical brokering to make decisions. Routinely standardising, implementing and regulating surgical decision making presents a challenge when the decision-making process is in a constant state of flux. We found that surgeons constantly prioritise drivers of decision-making in a flexible and context-specific manner. We introduce the concept of socialisation in decision making, which describes \"the socialisation of factors affecting decision-making. Socialisation is additive to surgeon identity and organisational capacity, which as explanatory linchpins act to mediate our understanding of how and why surgical decision-making varies. Our conceptual framework allows us to rationalise why formal codified knowledge, typically endorsed through clinical guidelines, consistently plays a limited role in orthopaedic decision-making.</p><p><strong>Conclusions: </strong>We present a re-conceptualised framework for understanding what drives real world decision-making in orthopaedics. This framework highlights the dynamic and fluid way these drivers of decision-making are applied in orthopaedic surgical work. A shift in orthopaedics is required away from prioritising informal, experiential knowledge first to incorporating evidence-based sources of evidence as essential for decision-making. This paradigm shift, views decision-making as a complex intervention, that requires alternative approaches underpinned by multi-faceted, evidence-based implementation strategies to encourage evidence-based practice.</p><p><strong>Registration: </strong>PROSPERO CRD42022311442 CLINICAL TRIAL NUMBER: Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"216"},"PeriodicalIF":3.3,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12211141/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144538995","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":"The FAIR data point populator: collaborative FAIRification and population of FAIR data points.","authors":"Daphne Wijnbergen, Rajaram Kaliyaperumal, Kees Burger, Luiz Olavo Bonino da Silva Santos, Barend Mons, Marco Roos, Eleni Mina","doi":"10.1186/s12911-025-03022-7","DOIUrl":"10.1186/s12911-025-03022-7","url":null,"abstract":"<p><strong>Background: </strong>Use of the FAIR principles (Findable, Accessible, Interoperable and Reusable) allows the rapidly growing number of biomedical datasets to be optimally (re)used. An important aspect of the FAIR principles is metadata. The FAIR Data Point specifications and reference implementation have been designed as an example on how to publish metadata according to the FAIR principles. Metadata can be added to a FAIR Data Point with the FDP's web interface or through its API. However, these methods are either limited in scalability or only usable by users with a background in programming. We aim to provide a new tool for populating FDPs with metadata that addresses these limitations with the FAIR Data Point Populator.</p><p><strong>Results: </strong>The FAIR Data Point Populator consists of a GitHub workflow together with Excel templates that have tooltips, validation and documentation. The Excel templates are targeted towards non-technical users, and can be used collaboratively in online spreadsheet software. A more technical user then uses the GitHub workflow to read multiple entries in the Excel sheets, and transform it into machine readable metadata. This metadata is then automatically uploaded to a connected FAIR Data Point. We applied the FAIR Data Point Populator on the metadata of two datasets, and a patient registry. We were then able to run a query on the FAIR Data Point Index, in order to retrieve one of the datasets.</p><p><strong>Conclusion: </strong>The FAIR Data Point Populator addresses the limitations of the other metadata publication methods by allowing the bulk creation of metadata entries while remaining accessible for users without a background in programming. Additionally, it allows efficient collaboration. As a result of this, the barrier of entry for FAIRification is lower, which allows the creation of FAIR data by more people.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 Suppl 1","pages":"211"},"PeriodicalIF":3.3,"publicationDate":"2025-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12150429/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144265303","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":"Predicting rapid kidney function decline in middle-aged and elderly Chinese adults using machine learning techniques.","authors":"Yang Li, Kun Zou, Yixuan Wang, Yucheng Zhang, Jingtao Zhong, Wu Zhou, Fang Tang, Lu Peng, Xusheng Liu, Lili Deng","doi":"10.1186/s12911-025-03043-2","DOIUrl":"10.1186/s12911-025-03043-2","url":null,"abstract":"<p><p>The rapid decline of kidney function in middle-aged and elderly people has become an increasingly serious public health problem. Machine learning (ML) technology has substantial potential to disease prediction. The present study use dataset from the Chinese Health and Retirement Longitudinal Study (CHARLS) and utilizes advanced Gradient Boosting algorithms to develop predictive models. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to identify the key predictors, and multivariate logistic regression was utilized to validate the independent predictive power of the variables. Furthermore, the study integrated SHapley Additive exPlanations (SHAP) to boost the interpretability of the model. The findings show that the Gradient Boosting Model demonstrated robust performance across both the training and test datasets. Specifically, it attained AUC values of 0.8 and 0.765 in the training and test sets, respectively, while achieving accuracy scores of 0.736 and 0.728 in these two datasets. LASSO regression identified key influencing factors, including estimated glomerular filtration rate (eGFR), age, hemoglobin (Hb), glucose, and systolic blood pressure (SBP). Multivariate linear regression further confirmed the independent associations between these variables and rapid kidney function deterioration (P < 0.05). This study developed a risk assessment model for rapid kidney function deterioration that is applicable to middle-aged and elderly populations in China.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"210"},"PeriodicalIF":3.3,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12144772/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144246499","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":"Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study.","authors":"Qiong Ma, Runqi Meng, Ruiting Li, Ling Dai, Fu Shen, Jie Yuan, Danqi Sun, Manman Li, Caixia Fu, Rong Li, Feng Feng, Yonggang Li, Tong Tong, Yajia Gu, Yiqun Sun, Dinggang Shen","doi":"10.1186/s12911-025-03050-3","DOIUrl":"10.1186/s12911-025-03050-3","url":null,"abstract":"<p><strong>Background: </strong>Prognostic prediction is crucial to guide individual treatment for patients with rectal cancer. We aimed to develop and validated a multitask deep learning model for predicting prognosis in rectal cancer patients.</p><p><strong>Methods: </strong>This retrospective study enrolled 321 rectal cancer patients (training set: 212; internal testing set: 53; external testing set: 56) who directly received total mesorectal excision from five hospitals between March 2014 to April 2021. A multitask deep learning model was developed to simultaneously predict recurrence/metastasis and disease-free survival (DFS). The model integrated clinicopathologic data and multiparametric magnetic resonance imaging (MRI) images including diffusion kurtosis imaging (DKI), without performing tumor segmentation. The receiver operating characteristic (ROC) curve and Harrell's concordance index (C-index) were used to evaluate the predictive performance of the proposed model.</p><p><strong>Results: </strong>The deep learning model achieved good discrimination capability of recurrence/metastasis, with area under the curve (AUC) values of 0.885, 0.846, and 0.797 in the training, internal testing and external testing sets, respectively. Furthermore, the model successfully predicted DFS in the training set (C-index: 0.812), internal testing set (C-index: 0.794), and external testing set (C-index: 0.733), and classified patients into significantly distinct high- and low-risk groups (p < 0.05).</p><p><strong>Conclusions: </strong>The multitask deep learning model, incorporating clinicopathologic data and multiparametric MRI, effectively predicted both recurrence/metastasis and survival for patients with rectal cancer. It has the potential to be an essential tool for risk stratification, and assist in making individualized treatment decisions.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"209"},"PeriodicalIF":3.3,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12143085/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144233265","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 systematic comparison of short-term and long-term mortality prediction in acute myocardial infarction using machine learning models.","authors":"Yawei Yang, Junjie Tang, Liping Ma, Feng Wu, Xiaoqing Guan","doi":"10.1186/s12911-025-03052-1","DOIUrl":"10.1186/s12911-025-03052-1","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background and objective: &lt;/strong&gt;The machine learning (ML) models for acute myocardial infarction (AMI) are considered to have better predictive ability for mortality compared to conventional risk scoring models. However, previous ML prediction models have mostly been short-term (1 year or less) models. Here, we established ML models for long-term prediction of AMI mortality (5 years or 10 years) and systematically compare the predictive capabilities of short-term models versus long-term models across varying survival time periods.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;An observational retrospective study was conducted to analyse mortality prediction in patients with varying survival times. A total of 4,173 patients were enrolled from two different hospitals in China. The dataset was allocated into three groups and an external test set based on their survival duration: the 1-year group (n = 3,626), the 5-year group (n = 2,102), the 10-year group (n = 721), and the external test set (n = 545). A comprehensive set of 53 variables was collected and utilized for model development. Mortality prediction was analysed using oversampling and feature selection methods coupled with machine learning algorithms. SHapley Additive exPlanations (SHAP) values were utilized to quantify the feature importance of AMI risk. The best-performing models from each group were selected for a systematic comparison of predictive accuracy using the external test set with follow-up exceeding 10 years but with varying survival times.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;For the 1-year model, the RF model achieved the best performance on the test dataset, with an F1 score of 97.81% using only oversampling without feature selection. Conversely, in the case of the 5-years, the combination of LASSO and RF yielded the best performance, achieving F1 scores of 91.35% with both feature selection and oversampling. The best model of 10-years group was SVM with only oversampling without feature selection, yielding an F1 score of 80.7%. Age, BNP, and the Killip classification of AMI were consistently identified as robust predictors across all three groups. This underscores aging as a critical AMI risk factor contributing to mortality. However, despite the model's success, when examining the actual survival times of the 545 patients, of which 64% survived beyond 5 years and 37% beyond 10 years, the 1-year model failed to distinguish between these patients, predicting all as low risk. This highlights the limitation of short-term models, indicating their inability to accurately predict actual long-term survival times despite being commonly used in AMI mortality prediction.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;The study identifies Age, BNP, and Killip classification as consistent predictors of AMI mortality across all groups, with Age being the most significant factor. CBC parameters and renal biomarkers were pivotal in short-term models, while therapeutic interventions gained prominence over","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"208"},"PeriodicalIF":3.3,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12143097/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144233264","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 model applied to real-time delineation of colorectal polyps.","authors":"Moana Gelu-Simeon, Adel Mamou, Georgette Saint-Georges, Marceline Alexis, Marie Sautereau, Yassine Mamou, Jimmy Simeon","doi":"10.1186/s12911-025-03047-y","DOIUrl":"10.1186/s12911-025-03047-y","url":null,"abstract":"<p><strong>Background: </strong>Deep learning models have shown considerable potential to improve diagnostic accuracy across medical fields. Although YOLACT has demonstrated real-time detection and segmentation in non-medical datasets, its application in medical settings remains underexplored. This study evaluated the performance of a YOLACT-derived Real-time Polyp Delineation Model (RTPoDeMo) for real-time use on prospectively recorded colonoscopy videos.</p><p><strong>Methods: </strong>Twelve combinations of architectures, including Mask-RCNN, YOLACT, and YOLACT++, paired with backbones such as ResNet50, ResNet101, and DarkNet53, were tested on 2,188 colonoscopy images with three image resolution sizes. Dataset preparation involved pre-processing and segmentation annotation, with optimized image augmentation.</p><p><strong>Results: </strong>RTPoDeMo, using YOLACT-ResNet50, achieved 72.3 mAP and 32.8 FPS for real-time instance segmentation based on COCO annotations. The model performed with a per-image accuracy of 99.59% (95% CI: [99.45 - 99.71%]), sensitivity of 90.63% (95% CI: [78.95 - 93.64%]), specificity of 99.95% (95% CI: [99.93 - 99.97%]) and a F1-score of 0.94 (95% CI: [0.87-0.98]). In validation, out of 36 polyps detected by experts, RTPoDeMo missed only one polyp, compared to six missed by senior endoscopists. The model demonstrated good agreement with experts, reflected by a Cohen's Kappa coefficient of 0.72 (95% CI: [0.54-1.00], p < 0.0001).</p><p><strong>Conclusions: </strong>Our model provides new perspectives in the adaptation of YOLACT to the real-time delineation of colorectal polyps. In the future, it could improve the characterization of polyps to be resected during colonoscopy.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"206"},"PeriodicalIF":3.3,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12135501/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144224352","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}