BMC Medical Informatics and Decision Making最新文献

{"title":"Addressing label noise for electronic health records: insights from computer vision for tabular data.","authors":"Jenny Yang, Hagen Triendl, Andrew A S Soltan, Mangal Prakash, David A Clifton","doi":"10.1186/s12911-024-02581-5","DOIUrl":"https://doi.org/10.1186/s12911-024-02581-5","url":null,"abstract":"<p><p>The analysis of extensive electronic health records (EHR) datasets often calls for automated solutions, with machine learning (ML) techniques, including deep learning (DL), taking a lead role. One common task involves categorizing EHR data into predefined groups. However, the vulnerability of EHRs to noise and errors stemming from data collection processes, as well as potential human labeling errors, poses a significant risk. This risk is particularly prominent during the training of DL models, where the possibility of overfitting to noisy labels can have serious repercussions in healthcare. Despite the well-documented existence of label noise in EHR data, few studies have tackled this challenge within the EHR domain. Our work addresses this gap by adapting computer vision (CV) algorithms to mitigate the impact of label noise in DL models trained on EHR data. Notably, it remains uncertain whether CV methods, when applied to the EHR domain, will prove effective, given the substantial divergence between the two domains. We present empirical evidence demonstrating that these methods, whether used individually or in combination, can substantially enhance model performance when applied to EHR data, especially in the presence of noisy/incorrect labels. We validate our methods and underscore their practical utility in real-world EHR data, specifically in the context of COVID-19 diagnosis. Our study highlights the effectiveness of CV methods in the EHR domain, making a valuable contribution to the advancement of healthcare analytics and research.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11212446/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141466325","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":"Establishment of prediction model for mortality risk of pancreatic cancer: a retrospective study.","authors":"Raoof Nopour","doi":"10.1186/s12911-024-02590-4","DOIUrl":"https://doi.org/10.1186/s12911-024-02590-4","url":null,"abstract":"<p><strong>Background and aim: </strong>Pancreatic cancer possesses a high prevalence and mortality rate among other cancers. Despite the low survival rate of this cancer type, the early prediction of this disease has a crucial role in decreasing the mortality rate and improving the prognosis. So, this study.</p><p><strong>Materials and methods: </strong>In this retrospective study, we used 654 alive and dead PC cases to establish the prediction model for PC. The six chosen machine learning algorithms and prognostic factors were utilized to build the prediction models. The importance of the predictive factors was assessed using the relative importance of a high-performing algorithm.</p><p><strong>Results: </strong>The XG-Boost with AU-ROC of 0.933 (95% CI= [0.906-0.958]) and AU-ROC of 0.836 (95% CI= [0.789-0.865] in internal and external validation modes were considered as the best-performing model for predicting the mortality risk of PC. The factors, including tumor size, smoking, and chemotherapy, were considered the most influential for prediction.</p><p><strong>Conclusion: </strong>The XG-Boost gained more performance efficiency in predicting the mortality risk of PC patients, so this model can promote the clinical solutions that doctors can achieve in healthcare environments to decrease the mortality risk of these patients.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11210158/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141466349","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 and usability testing of an online support tool to identify models and frameworks to inform implementation.","authors":"Lisa Strifler, Christine Fahim, Michael P Hillmer, Jan M Barnsley, Sharon E Straus","doi":"10.1186/s12911-024-02580-6","DOIUrl":"https://doi.org/10.1186/s12911-024-02580-6","url":null,"abstract":"<p><strong>Background: </strong>Theories, models and frameworks (TMFs) are useful when implementing, evaluating and sustaining healthcare evidence-based interventions. Yet it can be challenging to identify an appropriate TMF for an implementation project. We developed and tested the usability of an online tool to help individuals who are doing or supporting implementation practice activities to identify appropriate models and/or frameworks to inform their work.</p><p><strong>Methods: </strong>We used methods guided by models and evidence on implementation science and user-centered design. Phases of tool development included applying findings from a scoping review of TMFs and interviews with 24 researchers/implementers on barriers and facilitators to identifying and selecting TMFs. Based on interview findings, we categorized the TMFs by aim, stage of implementation, and target level of change to inform the tool's algorithm. We then conducted interviews with 10 end-users to test the usability of the prototype tool and administered the System Usability Scale (SUS). Usability issues were addressed and incorporated into the tool.</p><p><strong>Results: </strong>We developed Find TMF, an online tool consisting of 3-4 questions about the user's implementation project. The tool's algorithm matches key characteristics of the user's project (aim, stage, target change level) with characteristics of different TMFs and presents a list of candidate models/frameworks. Ten individuals from Canada or Australia participated in usability testing (mean SUS score 84.5, standard deviation 11.4). Overall, participants found the tool to be simple, easy to use and visually appealing with a useful output of candidate models/frameworks to consider for an implementation project. Users wanted additional instruction and guidance on what to expect from the tool and how to use the information in the output table. Tool improvements included incorporating an overview figure outlining the tool steps and output, displaying the tool questions on a single page, and clarifying the available functions of the results page, including adding direct links to the glossary and to complementary tools.</p><p><strong>Conclusions: </strong>Find TMF is an easy-to-use online tool that may benefit individuals who support implementation practice activities by making the vast number of models and frameworks more accessible, while also supporting a consistent approach to identifying and selecting relevant TMFs.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11209996/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141466348","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":"Operationalizing and digitizing person-centered daily functioning: a case for functionomics.","authors":"Esther R C Janssen, Ilona M Punt, Johan van Soest, Yvonne F Heerkens, Hillegonda A Stallinga, Huib Ten Napel, Lodewijk W van Rhijn, Barend Mons, Andre Dekker, Paul C Willems, Nico L U van Meeteren","doi":"10.1186/s12911-024-02584-2","DOIUrl":"https://doi.org/10.1186/s12911-024-02584-2","url":null,"abstract":"<p><p>An ever-increasing amount of data on a person's daily functioning is being collected, which holds information to revolutionize person-centered healthcare. However, the full potential of data on daily functioning cannot yet be exploited as it is mostly stored in an unstructured and inaccessible manner. The integration of these data, and thereby expedited knowledge discovery, is possible by the introduction of functionomics as a complementary 'omics' initiative, embracing the advances in data science. Functionomics is the study of high-throughput data on a person's daily functioning, that can be operationalized with the International Classification of Functioning, Disability and Health (ICF).A prerequisite for making functionomics operational are the FAIR (Findable, Accessible, Interoperable, and Reusable) principles. This paper illustrates a step by step application of the FAIR principles for making functionomics data machine readable and accessible, under strictly certified conditions, in a practical example. Establishing more FAIR functionomics data repositories, analyzed using a federated data infrastructure, enables new knowledge generation to improve health and person-centered healthcare. Together, as one allied health and healthcare research community, we need to consider to take up the here proposed methods.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11212415/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141466350","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":"Patterns and factors associated with dental service utilization among insured people: a data mining approach.","authors":"Zahra Pouraskari, Reza Yazdani, Maryam Khademi, Hossein Hessari","doi":"10.1186/s12911-024-02572-6","DOIUrl":"10.1186/s12911-024-02572-6","url":null,"abstract":"<p><strong>Background: </strong>Insurance databases contain valuable information related to the use of dental services. This data is instrumental in decision-making processes, enhancing risk assessment, and predicting outcomes. The objective of this study was to identify patterns and factors influencing the utilization of dental services among complementary insured individuals, employing a data mining methodology.</p><p><strong>Methods: </strong>A secondary data analysis was conducted using a dental insurance dataset from Iran in 2022. The Cross-Industry Standard Process for Data Mining (CRISP-DM) was employed as a data mining approach for knowledge extraction from the database. The utilization of dental services was the outcome of interest, and independent variables were chosen based on the available information in the insurance dataset. Dental services were categorized into nine groups: diagnostic, preventive, periodontal, restorative, endodontic, prosthetic, implant, extraction/surgical, and orthodontic procedures. The independent variables included age, gender, family size, insurance history, franchise, insurance limit, and policyholder. A multinomial logistic regression model was utilized to investigate the factors associated with dental care utilization. All analyses were conducted using RapidMiner Version 2020.</p><p><strong>Results: </strong>The analysis encompassed a total of 654,418 records, corresponding to 118,268 insured individuals. Predominantly, restorative treatments were the most utilized services, accounting for approximately 38% of all services, followed by diagnostic (18.35%) and endodontic (13.3%) care. Individuals aged between 36 and 60 years had the highest rate of utilization for any dental services. Additionally, families comprising three to four members, individuals with a one-year insurance history, people contracted with a 20% franchise, individuals with a high insurance limit, and insured individuals with a small policyholder, exhibited the highest rate of service usage compared to their counterparts. The regression model revealed that all independent variables were significantly associated with the use of dental services. However, the patterns of association varied among different service categories.</p><p><strong>Conclusions: </strong>Restorative treatments emerged as the most frequently used dental services among insured individuals, followed by diagnostic and endodontic procedures. The pattern of service utilization was influenced by the characteristics of the insured individuals and attributes related to their insurance.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11197210/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141445616","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 tree-based explainable AI model for early detection of Covid-19 using physiological data.","authors":"Manar Abu Talib, Yaman Afadar, Qassim Nasir, Ali Bou Nassif, Haytham Hijazi, Ahmad Hasasneh","doi":"10.1186/s12911-024-02576-2","DOIUrl":"10.1186/s12911-024-02576-2","url":null,"abstract":"<p><p>With the outbreak of COVID-19 in 2020, countries worldwide faced significant concerns and challenges. Various studies have emerged utilizing Artificial Intelligence (AI) and Data Science techniques for disease detection. Although COVID-19 cases have declined, there are still cases and deaths around the world. Therefore, early detection of COVID-19 before the onset of symptoms has become crucial in reducing its extensive impact. Fortunately, wearable devices such as smartwatches have proven to be valuable sources of physiological data, including Heart Rate (HR) and sleep quality, enabling the detection of inflammatory diseases. In this study, we utilize an already-existing dataset that includes individual step counts and heart rate data to predict the probability of COVID-19 infection before the onset of symptoms. We train three main model architectures: the Gradient Boosting classifier (GB), CatBoost trees, and TabNet classifier to analyze the physiological data and compare their respective performances. We also add an interpretability layer to our best-performing model, which clarifies prediction results and allows a detailed assessment of effectiveness. Moreover, we created a private dataset by gathering physiological data from Fitbit devices to guarantee reliability and avoid bias.The identical set of models was then applied to this private dataset using the same pre-trained models, and the results were documented. Using the CatBoost tree-based method, our best-performing model outperformed previous studies with an accuracy rate of 85% on the publicly available dataset. Furthermore, this identical pre-trained CatBoost model produced an accuracy of 81% when applied to the private dataset. You will find the source code in the link: https://github.com/OpenUAE-LAB/Covid-19-detection-using-Wearable-data.git .</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11194929/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141445185","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 a quantitative index system for evaluating the quality of electronic medical records in disease risk intelligent prediction.","authors":"Jiayin Zhou, Jie Hao, Mingkun Tang, Haixia Sun, Jiayang Wang, Jiao Li, Qing Qian","doi":"10.1186/s12911-024-02533-z","DOIUrl":"10.1186/s12911-024-02533-z","url":null,"abstract":"<p><strong>Objective: </strong>This study aimed to develop and validate a quantitative index system for evaluating the data quality of Electronic Medical Records (EMR) in disease risk prediction using Machine Learning (ML).</p><p><strong>Materials and methods: </strong>The index system was developed in four steps: (1) a preliminary index system was outlined based on literature review; (2) we utilized the Delphi method to structure the indicators at all levels; (3) the weights of these indicators were determined using the Analytic Hierarchy Process (AHP) method; and (4) the developed index system was empirically validated using real-world EMR data in a ML-based disease risk prediction task.</p><p><strong>Results: </strong>The synthesis of review findings and the expert consultations led to the formulation of a three-level index system with four first-level, 11 second-level, and 33 third-level indicators. The weights of these indicators were obtained through the AHP method. Results from the empirical analysis illustrated a positive relationship between the scores assigned by the proposed index system and the predictive performances of the datasets.</p><p><strong>Discussion: </strong>The proposed index system for evaluating EMR data quality is grounded in extensive literature analysis and expert consultation. Moreover, the system's high reliability and suitability has been affirmed through empirical validation.</p><p><strong>Conclusion: </strong>The novel index system offers a robust framework for assessing the quality and suitability of EMR data in ML-based disease risk predictions. It can serve as a guide in building EMR databases, improving EMR data quality control, and generating reliable real-world evidence.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11194906/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141445186","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":"Factors associated with the local control of brain metastases: a systematic search and machine learning application.","authors":"Hemalatha Kanakarajan, Wouter De Baene, Karin Gehring, Daniëlle B P Eekers, Patrick Hanssens, Margriet Sitskoorn","doi":"10.1186/s12911-024-02579-z","DOIUrl":"10.1186/s12911-024-02579-z","url":null,"abstract":"<p><strong>Background: </strong>Enhancing Local Control (LC) of brain metastases is pivotal for improving overall survival, which makes the prediction of local treatment failure a crucial aspect of treatment planning. Understanding the factors that influence LC of brain metastases is imperative for optimizing treatment strategies and subsequently extending overall survival. Machine learning algorithms may help to identify factors that predict outcomes.</p><p><strong>Methods: </strong>This paper systematically reviews these factors associated with LC to select candidate predictor features for a practical application of predictive modeling. A systematic literature search was conducted to identify studies in which the LC of brain metastases is assessed for adult patients. EMBASE, PubMed, Web-of-Science, and the Cochrane Database were searched up to December 24, 2020. All studies investigating the LC of brain metastases as one of the endpoints were included, regardless of primary tumor type or treatment type. We first grouped studies based on primary tumor types resulting in lung, breast, and melanoma groups. Studies that did not focus on a specific primary cancer type were grouped based on treatment types resulting in surgery, SRT, and whole-brain radiotherapy groups. For each group, significant factors associated with LC were identified and discussed. As a second project, we assessed the practical importance of selected features in predicting LC after Stereotactic Radiotherapy (SRT) with a Random Forest machine learning model. Accuracy and Area Under the Curve (AUC) of the Random Forest model, trained with the list of factors that were found to be associated with LC for the SRT treatment group, were reported.</p><p><strong>Results: </strong>The systematic literature search identified 6270 unique records. After screening titles and abstracts, 410 full texts were considered, and ultimately 159 studies were included for review. Most of the studies focused on the LC of the brain metastases for a specific primary tumor type or after a specific treatment type. Higher SRT radiation dose was found to be associated with better LC in lung cancer, breast cancer, and melanoma groups. Also, a higher dose was associated with better LC in the SRT group, while higher tumor volume was associated with worse LC in this group. The Random Forest model predicted the LC of brain metastases with an accuracy of 80% and an AUC of 0.84.</p><p><strong>Conclusion: </strong>This paper thoroughly examines factors associated with LC in brain metastases and highlights the translational value of our findings for selecting variables to predict LC in a sample of patients who underwent SRT. The prediction model holds great promise for clinicians, offering a valuable tool to predict personalized treatment outcomes and foresee the impact of changes in treatment characteristics such as radiation dose.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11191176/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141436322","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":"Using a smartphone-based self-management platform to study sex differences in Parkinson's disease: multicenter, cross-sectional pilot study.","authors":"Zhiheng Xu, Lirong Jin, Weijie Chen, Tianyu Hu, Shiyu Li, Xiaoniu Liang, Xixi Han, Yi Chen, Yilin Tang, Jian Wang, Danhong Wu","doi":"10.1186/s12911-024-02569-1","DOIUrl":"10.1186/s12911-024-02569-1","url":null,"abstract":"<p><strong>Background: </strong>Patient-reported outcome (PRO) is a distinct and indispensable dimension of clinical characteristics and recent advances have made remote PRO measurement possible. Sex difference in PRO of Parkinson's disease (PD) is hardly extensively researched.</p><p><strong>Methods: </strong>A smartphone-based self-management platform, offering remote PRO measurement for PD patients, has been developed. A total of 1828 PD patients, including 1001 male patients and 827 female patients, were enrolled and completed their PRO submission through this platform.</p><p><strong>Results: </strong>Sex differences in PROs have been identified. The female group had a significantly lower height, weight, and body mass index (BMI) than the male group (P < 0.001). For motor symptoms, a higher proportion of patients reporting dyskinesia was observed in the female group. For non-motor symptoms, there is a higher percentage (P < 0.001) as well as severity (P = 0.016) of depression in the female group. More male patients reported hyposmia, lisp, drooling, dysuria, frequent urination, hypersexuality, impotence, daytime sleepiness, and apathy than females (P < 0.05). In contrast, more female patients reported headache, palpation, body pain, anorexia, nausea, urinal incontinence, anxiety, insomnia (P < 0.05) than males.</p><p><strong>Conclusions: </strong>We provide evidence for sex differences in PD through the data collected from our platform. These results highlighted the importance of gender in clinical decision-making, and also support the feasibility of remote PRO measurement through a smartphone-based self-management platform in patients with PD.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11191196/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141436323","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 and evaluation of machine learning models for predicting large-for-gestational-age newborns in women exposed to radiation prior to pregnancy.","authors":"Xi Bai, Zhibo Zhou, Zeyan Zheng, Yansheng Li, Kejia Liu, Yuanjun Zheng, Hongbo Yang, Huijuan Zhu, Shi Chen, Hui Pan","doi":"10.1186/s12911-024-02556-6","DOIUrl":"10.1186/s12911-024-02556-6","url":null,"abstract":"<p><strong>Introduction: </strong>The correlation between radiation exposure before pregnancy and abnormal birth weight has been previously proven. However, for large-for-gestational-age (LGA) babies in women exposed to radiation before becoming pregnant, there is no prediction model yet.</p><p><strong>Material and methods: </strong>The data were collected from the National Free Preconception Health Examination Project in China. A sum of 455 neonates (42 SGA births and 423 non-LGA births) were included. A training set (n = 319) and a test set (n = 136) were created from the dataset at random. To develop prediction models for LGA neonates, conventional logistic regression (LR) method and six machine learning methods were used in this study. Recursive feature elimination approach was performed by choosing 10 features which made a big contribution to the prediction models. And the Shapley Additive Explanation model was applied to interpret the most important characteristics that affected forecast outputs.</p><p><strong>Results: </strong>The random forest (RF) model had the highest average area under the receiver-operating-characteristic curve (AUC) for predicting LGA in the test set (0.843, 95% confidence interval [CI]: 0.714-0.974). Except for the logistic regression model (AUC: 0.603, 95%CI: 0.440-0.767), other models' AUCs displayed well. Thereinto, the RF algorithm's final prediction model using 10 characteristics achieved an average AUC of 0.821 (95% CI: 0.693-0.949).</p><p><strong>Conclusion: </strong>The prediction model based on machine learning might be a promising tool for the prenatal prediction of LGA births in women with radiation exposure before pregnancy.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11188254/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141431472","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}