Comparison the power of machine learning methods against traditional statistical approaches in predicting gestational diabetes: a study protocol

Comparison the power of machine learning methods against traditional statistical approaches in predicting gestational diabetes: a study protocol | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Study protocol Comparison the power of machine learning methods against traditional statistical approaches in predicting gestational diabetes: a study protocol Vahid Mehrnoush, Fatemeh Darsareh This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7583974/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Gestational diabetes (GDM) is linked to numerous negative pregnancy results for both the mother and the infant. Accurate predictions of GDM trends are essential for public health. Machine learning models have become effective instruments for disease prediction, showing enhanced performance compared to conventional techniques by identifying complex temporal relationships and nonlinear trends within health data. Consequently, we aimed to evaluate the power of machine learning compared to conventional statistical techniques in predicting GDM. Methods The birth records of all expectant mothers who were admitted from January 2020 to January 2022 to one of the primary referral tertiary centers situated in Bandar Abbas, Iran, will be incorporated. We will employ 2 methods for analysis. In the initial stage, we will employ conventional analytical techniques. The chi-square test will evaluate the relationship between categorical variables and GDM. Bivariate logistic regression will be performed to analyze the risk factors for GDM, estimating crude odds ratios (cORs) along with their 95% confidence intervals (CIs). At the second level, we will employ machine learning techniques to forecast GDM. The input data will be utilized in eight machine learning models. To assess the diagnostic capability of each model, the area under the receiver operating characteristic curve (ROC AUC), accuracy, precision, sensitivity, specificity, and F1 score will be calculated. Discussion The findings of this research show the hospital's existing incidence rates for gestational diabetes. Recognizing the risk factors for gestational diabetes forms the foundation for strategizing preventive actions against it. Machine learning gestational diabetes prognostic models Background Gestational diabetes mellitus (GDM) is usually characterized as hyperglycemia that is identified or arises during pregnancy. The American College of Obstetricians and Gynecologists (ACOG) defines gestational diabetes mellitus (GDM) as varying degrees of carbohydrate intolerance that begins or is detected during pregnancy [ 1 ]. GDM occurs in 3 to 5 percent of pregnancies [ 2 ]. GDM is linked to numerous negative pregnancy results for both the mother and the infant. Maternal complications encompass a heightened likelihood of preeclampsia, gestational hypertension, cesarean delivery, and postpartum bleeding. Fetal and neonatal issues encompass macrosomia, being large for gestational age (LGA), premature birth, respiratory distress syndrome, neonatal hypoglycemia, and needing admission to the neonatal intensive care unit (NICU) [ 3 ]. After birth, GDM can lead to hypoglycemia in newborns, seizures, jaundice, and slow motor development [ 4 ]. Elevated blood sugar levels during pregnancy adversely affect maternal well-being, increasing the likelihood of hypertension, metabolic complications, and cardiovascular problems [ 5 ]. Moreover, women with gestational diabetes mellitus face a considerably greater risk of developing type 2 diabetes later on [ 6 ]. Accurate predictions of GDM trends are essential for public health strategy, timely intervention, and distribution of resources. The benefits of identifying GDM have been extensively recorded, with various studies indicating that early diagnosis of GDM reduces adverse pregnancy results [ 7 ]. Multiple elements can be utilized to predict GDM. Factors include maternal age, ethnicity, residency place, body mass index, comorbidities, family diabetes history, and previous pregnancy history. Moreover, initial pregnancy glucose levels (both fasting and after the oral glucose tolerance test), HbA1c, and insulin sensitivity indicators such as the triglyceride-glucose index (TyGIS) demonstrate potential for early prediction [ 8 ]. Recently, the application of biomarkers like pregnancy-associated plasma protein (PAPP-A) has been implemented to anticipate GDM [ 9 ]. Conventional statistical models, though computationally efficient, frequently fail to represent the intricate, multifaceted aspects of GDM progression. Conventional statistical models like linear regression often struggle to effectively capture nonlinear relationships, long-term temporal dependencies, and the intricate interactions among socioeconomic and environmental influences [ 10 ]. Consequently, stronger and more flexible predictive models are necessary to manage extensive, practical healthcare data that contains missing data and noisy data. Machine learning models have become effective instruments for disease prediction, showing enhanced performance compared to conventional techniques by identifying complex temporal relationships and nonlinear trends within health data [ 11 – 17 ]. Nevertheless, these models continue to encounter issues such as data sparsity, sensitivity to absent data, and significant computational expenses, which could impede their use in practical applications [ 18 ]. Consequently, we aimed to evaluate the power of machine learning compared to conventional statistical techniques in forecasting GDM. Objectives To assess the prevalence of GDM. To determine the risk factors associated with GDM. To predict GDM utilizing conventional statistical methods To predict GDM utilizing machine learning approach To compare the accuracy of machine learning models for predicting the risk of GDM with the conventional statistical methods Methods Study Design With consent from the Hormozgan University of Medical School (HUMS) Research Ethics Board (REB), we will conduct an analysis of the medical records of patients who delivered at one of the primary referral tertiary centers situated in Bandar Abbas, Iran, which has an annual birth rate of 4800-5000. All data will be gathered and reviewed without any identification. Study Population The birth records of all expectant mothers who were admitted to our research site from January 2020 to January 2022 will be incorporated. Outcome measures The dataset covers a duration of two years and includes clinical information obtained from electronic health records, featuring demographic and clinical attributes (nationality, age, residence, maternal education, attendance at childbirth classes, medical insurance, body mass index, multiple pregnancy, gestational age, parity, smoking status, substance abuse, alcohol use, a history of abortion, neonatal death, intrauterine fetal death, infertility, assisted reproductive technology (ART), chronic hypertension, cardiovascular disease, iron deficiency anemia, hemoglobinopathy, hepatitis B, HIV, COVID-19, hypothyroidism, systemic lupus erythematosus or antiphospholipid syndrome, previous gestational diabetes, overt diabetes, familial diabetes, COVID-19 vaccination status, corticosteroid treatment during the ongoing pregnancy, and fetal sex). All the variables are categorical. Data Collection A number of research assistants will be hired to collect the required data of all patients. Handling missing data We will create a plan to address missing values according to their quantity. If any specific feature column contains over 40% of its values missing, we will eliminate that column to ensure data integrity. Given that numerous variables are categorical, we will utilize mode imputation, substituting missing values with the most frequently occurring category. This method guarantees that our dataset is comprehensive and maintains the overall quality and consistency of the data. Data Analysis We will employ 2 methods for analysis. In the initial stage, we will employ conventional analytical techniques. The data will be analyzed using IBM's Statistical Package for the Social Sciences Statistics (SPSS), version 25 (IBM Corp, Armonk, NY). Categorical variables will be represented as percentages. The chi-square test will evaluate the relationship between categorical variables and GDM. Bivariate logistic regression will be performed to analyze the risk factors for GDM, estimating crude odds ratios (cORs) along with their 95% confidence intervals (CIs). Variables showing a P value <.05 from a chi-square test will be included in a bivariate logistic regression analysis to calculate the cORs. P<.05 will be regarded as statistically significant. At the second level, we will employ machine learning techniques to forecast GDM. Every machine learning task contains a unique set of essential features for achieving optimal classification accuracy. Identifying the optimal feature combination for training a classifier is essential as it enhances both system accuracy and computational efficiency. Due to the absence of a definitive "best" machine learning algorithm for medical predictions involving categorical variables, the ideal selection relies on the particular dataset, the characteristics of the categorical variables, and the intended goal (e.g., interpretability versus predictive accuracy). Nevertheless, numerous algorithms are commonly utilized and show excellent results in this area [19]. In our research, the input data will be utilized in eight machine learning models, comprising deep learning-feed forward, XGBoost classification, random forest classification, support vector machine (SVM), logistic regression, permutation feature classification with KNN, light gradient boosting, and decision tree classification. K-fold cross-validation will be utilized to conduct internal validation. The cases will be allocated randomly to either the "training set" (70%) or the "test set" (30%) utilizing a random number generator. The original dataset will be split into GDM group and non-GDM groups while maintaining constancy in the training and test sets. Utilizing the training set, we will adjust the parameters of the prediction models and assess their performance with the "test set." The mean performance will be determined by conducting these ten times. To assess the diagnostic capability of each model, 6 metrics will be calculated on the test set, including the area under the receiver operating characteristic curve (ROC AUC), accuracy, precision, sensitivity, specificity, and F1 score. Since ROC AUC is a commonly utilized measure to evaluate a machine learning model's capacity to forecast results, we will adopt it as the main performance metric. The metrics varied from 0 to 1, where values nearer to 1 signified a superior model. The mistake rate of every model will be examined as well. Table 1 contains all of the metrics that will be used to evaluate the performance of each machine learning [20]. Evaluation metrics are quantitative measures used to examine the performance of machine learning, providing insights into how well the model is working and assisting in the comparison of alternative models or algorithms. Python (version 3.7.0) will be used for analysis. Table 1. performance evaluation metrics of machine learning models. Metric Definition Accuracy The proportion of the total number of correct predictions that were correct. Positive Predictive Value (Precision) The proportion of positive cases that were correctly identified. Negative Predictive Value The proportion of negative cases that were correctly identified. Sensitivity (Recall) The proportion of actual positive cases which are correctly identified. Specificity The proportion of actual negative cases which are correctly identified. F1 Score The harmonic mean of precision and recall values for a classification problem. Our findings will be presented in line with the Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary Perspective. The Python programming language will be utilized to create the machine learning model, while Scikit-learn will be used for the implementation of the machine learning algorithm [19]. Practical Implications The results of this study are important from several perspectives. 1) Shows the hospital's present rates for GDM. 2) Determining the risk factors associated with GDM. 3) Given that patient safety and satisfaction are closely tied to pregnancy complications, this study can enhance both by tackling the risk factors associated with GDM. 4) The research results may guide the creation of a model to predict GDM in its initial phase. 5) This study is relevant in various medical domains, extending beyond just obstetrics and women's health. Strength and limitations The findings of this research show the hospital's existing incidence rates for gestational diabetes Recognizing the risk factors for gestational diabetes forms the foundation for strategizing preventive actions against it This study is relevant to various medical issues during pregnancy, not just gestational diabetes Due to the retrospective nature of the study, there are concerns regarding missing data, which we will address with an effective strategy for managing them. Declarations Ethics approval and consent to participate With authorization from the Hormozgan University of Medical School Research Ethics Board (IR.HUMS.REC.1402.239), we will conduct a review of the medical records of every patient who delivered at our facility. The Ethics and Research Committee of Hormozgan University of Medical Sciences exempted the necessity for informed consent for involvement due to the study's retrospective design. Statistical analysis was performed while maintaining patient confidentiality and complying with ethics committee guidelines. The results of this study will be published in suitable scientific journals. Consent for publication Not applicable. Availability of data and materials Data sharing is not relevant since no datasets have been generated or analyzed for this study at this time. Competing interests None declared. Funding Hormozgan University of Medical Sciences. Authors' contributions FD and VM are in charge of protocol design and manuscript conception. Acknowledgments We wish to convey our heartfelt appreciation to Hormozgan University of Medical Sciences in Bandar Abbas, Iran for their steadfast support. We anticipate that, with your important cooperation and direction, this research will be carried out effectively in the future. References ACOG Practice Bulletin No. Gestational Diabetes Mellitus. Obstet Gynecol. 2018;190(2):e49–64. 10.1097/AOG.0000000000002501 . Sartayeva A, Danyarova L, Begalina D, Nurgalieva Z, Baikadamova L, Adilova G. Gestational Diabetes: Prevelence and Risk for the Mother and Child (Review). Georgian Med News. 2022 Jul-Aug;(328–329):47–52. Darsareh F, Jahromi M, Ranjbar A, Shekari M, Mehrnoush V, Roozbeh N. Pregnancy and Childbirth Outcomes of Gestational Diabetes Mellitus: A Retrospective Cohort Study. Dis Diagn. 2024;13(4):151–6. 10.34172/ddj.1608 . Tehrani FR, Naz MSG, Bidhendi-Yarandi R, Behboudi-Gandevani S. Effect of Different Types of Diagnostic Criteria for Gestational Diabetes Mellitus on Adverse Neonatal Outcomes: A Systematic Review, Meta-Analysis, and Meta-Regression. Diabetes Metab J. 2022;46(4):605–19. 10.4093/dmj.2021.0178 . Epub 2022 Mar 8. Weir TL, Majumder M, Glastras SJ. A systematic review of the effects of maternal obesity on neonatal outcomes in women with gestational diabetes. Obes Rev. 2024;25(7):e13747. 10.1111/obr.13747 . Sartayeva A, Danyarova L, Begalina D, Nurgalieva Z, Baikadamova L, Adilova G. Gestational Diabetes: Prevelence and Risk for the Mother and Child (Review). Georgian Med News. 2022 Jul-Aug;(328–329):47–52. Quintanilla Rodriguez BS, Vadakekut ES, Mahdy H, Gestational D. 2024 Jul 14. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 Jan–. PMID: 31424780. Sert UY, Ozgu-Erdinc AS. Gestational Diabetes Mellitus Screening and Diagnosis. Adv Exp Med Biol. 2021;1307:231–55. 10.1007/5584_2020_512 . Caliskan R, Atis A, Aydin Y, Acar D, Kiyak H, Topbas F. PAPP-A concentrations change in patients with gestational diabetes. J Obstet Gynaecol. 2020;40(2):190–4. 10.1080/01443615.2019.1615041 . Fayaz L, Joseph R, Ankayarkanni B, Princemary S, Asha P, Student U. Multi-scale and context aware optimized glucose prediction using neural networks. In 2023 7th International Conference on Trends in Electronics and Informatics (ICOEI). 820–825 (2023). Boujarzadeh B, Ranjbar A, Banihashemi F, Mehrnoush V, Darsareh F, Saffari M. Machine learning approach to predict postpartum haemorrhage: a systematic review protocol. BMJ Open. 2023;13(1):e067661. 10.1136/bmjopen-2022-067661 . Banaei M, Roozbeh N, Darsareh F, Mehrnoush V, Farashah MSV, Montazeri F. Utilizing machine learning to predict the risk factors of episiotomy in parturient women. AJOG Glob Rep. 2024;5(1):100420. 10.1016/j.xagr.2024.100420 . Ranjbar A, Taeidi E, Mehrnoush V, Roozbeh N, Darsareh F. Machine learning models for predicting pre-eclampsia: a systematic review protocol. BMJ Open. 2023;13(9):e074705. 10.1136/bmjopen-2023-074705 . Taeidi E, Ranjbar A, Montazeri F, Mehrnoush V, Darsareh F. Machine Learning-Based Approach to Predict Intrauterine Growth Restriction. Cureus. 2023;15(7):e41448. 10.7759/cureus . Ranjbar A, Ghamsari S, Boujarzade B, Mehrnoush V, Darsareh F. Predicting risk of postpartum hemorrhage using machine learning approach: A systematic review. GOCM. 2023;3(3):170–4. doi.org/10.1016/j.gocm.2023.07.002 . Malakooti N, Mehrnoush V, Abdi F, Farashah MSV, Darsareh F. Development of a machine learning model to identify the predictors of the neonatal intensive care unit admission. Sci Rep. 2025;15(1):20914. 10.1038/s41598-025-06651-0 . Safarzadeh S, Ardabili NS, Farashah MV, Roozbeh N, Darsareh F. Predicting mother and newborn skin-to-skin contact using a machine learning approach. BMC Pregnancy Childbirth. 2025;25(1):182. 10.1186/s12884-025-07313-9 . Esmaeilyfard R, Bayati M. Enhancing AI-driven forecasting of diabetes burden: a comparative analysis of deep learning and statistical models. Sci Rep. 2025;15:29137. doi.org/10.1038/s41598-025-14599-4 . Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay É. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825–30. 10.1145/2786984.2786995 . Erickson BJ, Kitamura F. Magician's Corner: 9. Performance Metrics for Machine Learning Models. Radiol Artif Intell. 2021;3(3):e200126. 10.1148/ryai.2021200126 . Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7583974","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Study protocol","associatedPublications":[],"authors":[{"id":585634258,"identity":"724e500e-6316-4f7d-a471-06e852599bdf","order_by":0,"name":"Vahid Mehrnoush","email":"","orcid":"","institution":"Hormozgan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Vahid","middleName":"","lastName":"Mehrnoush","suffix":""},{"id":585634259,"identity":"acc8505b-ad97-4e30-9904-89822524506e","order_by":1,"name":"Fatemeh Darsareh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYBAC9gYgwQNhszF8YGBIIKiF5wCSFsYZJGth5iFKC/vpxA9vau7Z888+fOyxbZtdHj97A+OHjzl4tPDkbpacc6w4cca5tHTj3LbkYsmeA8ySM7fh1mLPkLtBmoctIYHhDI+ZdG4bc+KGGwlszLx4tPDwv938m+dfgr38Gf5v0pZt9URokcjdJs3blsC44QwPmzRj22FitLzdZjm3LyFx4xk2M8mec8cTZ/YcbMbrFx7+3M033nxLsJc7w/xM4kdZdWI/e/PBDx/xaEEFjGxgsoFY9SDwhxTFo2AUjIJRMFIAALTpURiTGy40AAAAAElFTkSuQmCC","orcid":"","institution":"Hormozgan University of Medical Sciences","correspondingAuthor":true,"prefix":"","firstName":"Fatemeh","middleName":"","lastName":"Darsareh","suffix":""}],"badges":[],"createdAt":"2025-09-10 14:38:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7583974/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7583974/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101971435,"identity":"af0b464e-24aa-4940-a195-5bed3f1ad2f1","added_by":"auto","created_at":"2026-02-05 14:43:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":425294,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7583974/v1/7e9c8a22-d29a-40c5-8c63-536b0a405f2a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Comparison the power of machine learning methods against traditional statistical approaches in predicting gestational diabetes: a study protocol","fulltext":[{"header":"Background","content":"\u003cp\u003eGestational diabetes mellitus (GDM) is usually characterized as hyperglycemia that is identified or arises during pregnancy. The American College of Obstetricians and Gynecologists (ACOG) defines gestational diabetes mellitus (GDM) as varying degrees of carbohydrate intolerance that begins or is detected during pregnancy [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. GDM occurs in 3 to 5 percent of pregnancies [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. GDM is linked to numerous negative pregnancy results for both the mother and the infant. Maternal complications encompass a heightened likelihood of preeclampsia, gestational hypertension, cesarean delivery, and postpartum bleeding. Fetal and neonatal issues encompass macrosomia, being large for gestational age (LGA), premature birth, respiratory distress syndrome, neonatal hypoglycemia, and needing admission to the neonatal intensive care unit (NICU) [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. After birth, GDM can lead to hypoglycemia in newborns, seizures, jaundice, and slow motor development [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Elevated blood sugar levels during pregnancy adversely affect maternal well-being, increasing the likelihood of hypertension, metabolic complications, and cardiovascular problems [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Moreover, women with gestational diabetes mellitus face a considerably greater risk of developing type 2 diabetes later on [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Accurate predictions of GDM trends are essential for public health strategy, timely intervention, and distribution of resources. The benefits of identifying GDM have been extensively recorded, with various studies indicating that early diagnosis of GDM reduces adverse pregnancy results [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMultiple elements can be utilized to predict GDM. Factors include maternal age, ethnicity, residency place, body mass index, comorbidities, family diabetes history, and previous pregnancy history. Moreover, initial pregnancy glucose levels (both fasting and after the oral glucose tolerance test), HbA1c, and insulin sensitivity indicators such as the triglyceride-glucose index (TyGIS) demonstrate potential for early prediction [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Recently, the application of biomarkers like pregnancy-associated plasma protein (PAPP-A) has been implemented to anticipate GDM [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Conventional statistical models, though computationally efficient, frequently fail to represent the intricate, multifaceted aspects of GDM progression. Conventional statistical models like linear regression often struggle to effectively capture nonlinear relationships, long-term temporal dependencies, and the intricate interactions among socioeconomic and environmental influences [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Consequently, stronger and more flexible predictive models are necessary to manage extensive, practical healthcare data that contains missing data and noisy data.\u003c/p\u003e \u003cp\u003eMachine learning models have become effective instruments for disease prediction, showing enhanced performance compared to conventional techniques by identifying complex temporal relationships and nonlinear trends within health data [\u003cspan additionalcitationids=\"CR12 CR13 CR14 CR15 CR16\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Nevertheless, these models continue to encounter issues such as data sparsity, sensitivity to absent data, and significant computational expenses, which could impede their use in practical applications [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Consequently, we aimed to evaluate the power of machine learning compared to conventional statistical techniques in forecasting GDM.\u003c/p\u003e \u003cp\u003e \u003cb\u003eObjectives\u003c/b\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eTo assess the prevalence of GDM.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eTo determine the risk factors associated with GDM.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eTo predict GDM utilizing conventional statistical methods\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eTo predict GDM utilizing machine learning approach\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eTo compare the accuracy of machine learning models for predicting the risk of GDM with the conventional statistical methods\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy Design\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWith consent from the Hormozgan University of Medical School (HUMS) Research Ethics Board (REB), we will conduct an analysis of the medical records of patients who delivered at one of the primary referral tertiary centers situated in Bandar Abbas, Iran, which has an annual birth rate of 4800-5000. All data will be gathered and reviewed without any identification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy Population\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe birth records of all expectant mothers who were admitted to our research site from January 2020 to January 2022 will be incorporated.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOutcome measures\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dataset covers a duration of two years and includes clinical information obtained from electronic health records, featuring demographic and clinical attributes (nationality, age, residence, maternal education, attendance at childbirth classes, medical insurance, body mass index, multiple pregnancy, gestational age, parity, smoking status, substance abuse, alcohol use, a history of abortion, neonatal death, intrauterine fetal death, infertility, assisted reproductive technology (ART), chronic hypertension, cardiovascular disease, iron deficiency anemia, hemoglobinopathy, hepatitis B, HIV, COVID-19, hypothyroidism, systemic lupus erythematosus or antiphospholipid syndrome, previous gestational diabetes, overt diabetes, familial diabetes, COVID-19 vaccination status, corticosteroid treatment during the ongoing pregnancy, and fetal sex). All the variables are categorical.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA number of research assistants will be hired to collect the required data of all patients.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHandling missing data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe will create a plan to address missing values according to their quantity. If any specific feature column contains over 40% of its values missing, we will eliminate that column to ensure data integrity. Given that numerous variables are categorical, we will utilize mode imputation, substituting missing values with the most frequently occurring category. This method guarantees that our dataset is comprehensive and maintains the overall quality and consistency of the data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe will employ 2 methods for analysis. In the initial stage, we will employ conventional analytical techniques. The data will be analyzed using IBM's Statistical Package for the Social Sciences Statistics (SPSS), version 25 (IBM Corp, Armonk, NY). Categorical variables will be represented as percentages. The chi-square test will evaluate the relationship between categorical variables and GDM. Bivariate logistic regression will be performed to analyze the risk factors for GDM, estimating crude odds ratios (cORs) along with their 95% confidence intervals (CIs). Variables showing a P value \u0026lt;.05 from a chi-square test will be included in a bivariate logistic regression analysis to calculate the cORs. P\u0026lt;.05 will be regarded as statistically significant.\u003c/p\u003e\n\u003cp\u003eAt the second level, we will employ machine learning techniques to forecast GDM. Every machine learning task contains a unique set of essential features for achieving optimal classification accuracy. Identifying the optimal feature combination for training a classifier is essential as it enhances both system accuracy and computational efficiency. Due to the absence of a definitive \"best\" machine learning algorithm for medical predictions involving categorical variables, the ideal selection relies on the particular dataset, the characteristics of the categorical variables, and the intended goal (e.g., interpretability versus predictive accuracy). Nevertheless, numerous algorithms are commonly utilized and show excellent results in this area [19]. In our research, the input data will be utilized in eight machine learning models, comprising deep learning-feed forward, XGBoost classification, random forest classification, support vector machine (SVM), logistic regression, permutation feature classification with KNN, light gradient boosting, and decision tree classification.\u003c/p\u003e\n\u003cp\u003eK-fold cross-validation will be utilized to conduct internal validation. The cases will be allocated randomly to either the \"training set\" (70%) or the \"test set\" (30%) utilizing a random number generator. The original dataset will be split into GDM group and non-GDM groups while maintaining constancy in the training and test sets. Utilizing the training set, we will adjust the parameters of the prediction models and assess their performance with the \"test set.\" The mean performance will be determined by conducting these ten times.\u003c/p\u003e\n\u003cp\u003eTo assess the diagnostic capability of each model, 6 metrics will be calculated on the test set, including the area under the receiver operating characteristic curve (ROC AUC), accuracy, precision, sensitivity, specificity, and F1 score. Since ROC AUC is a commonly utilized measure to evaluate a machine learning model's capacity to forecast results, we will adopt it as the main performance metric. The metrics varied from 0 to 1, where values nearer to 1 signified a superior model. The mistake rate of every model will be examined as well. Table 1 contains all of the metrics that will be used to evaluate the performance of each machine learning [20]. Evaluation metrics are quantitative measures used to examine the performance of machine learning, providing insights into how well the model is working and assisting in the comparison of alternative models or algorithms. Python (version 3.7.0) will be used for analysis.\u003c/p\u003e\n\u003cp\u003eTable 1. performance evaluation metrics of machine learning models.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"654\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eMetric\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eDefinition\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eAccuracy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eThe proportion of the total number of correct predictions that were correct.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003ePositive Predictive Value (Precision)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eThe proportion of positive cases that were correctly identified.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eNegative Predictive Value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eThe proportion of negative cases that were correctly identified.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eSensitivity (Recall)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eThe proportion of actual positive cases which are correctly identified.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eSpecificity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eThe proportion of actual negative cases which are correctly identified.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eF1 Score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp dir=\"RTL\"\u003eThe harmonic mean of precision and recall values for a classification problem.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eOur findings will be presented in line with the Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary Perspective. The Python programming language will be utilized to create the machine learning model, while Scikit-learn will be used for the implementation of the machine learning algorithm [19].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePractical Implications\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe results of this study are important from several perspectives.\u003c/p\u003e\n\u003cp\u003e1) Shows the hospital's present rates for GDM.\u003c/p\u003e\n\u003cp\u003e2) Determining the risk factors associated with GDM.\u003c/p\u003e\n\u003cp\u003e3) Given that patient safety and satisfaction are closely tied to pregnancy complications, this study can enhance both by tackling the risk factors associated with GDM.\u003c/p\u003e\n\u003cp\u003e4) The research results may guide the creation of a model to predict GDM in its initial phase.\u003c/p\u003e\n\u003cp\u003e5) This study is relevant in various medical domains, extending beyond just obstetrics and women's health.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStrength and limitations\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eThe findings of this research show the hospital's existing incidence rates for gestational diabetes\u003c/li\u003e\n \u003cli\u003eRecognizing the risk factors for gestational diabetes forms the foundation for strategizing preventive actions against it\u003c/li\u003e\n \u003cli\u003eThis study is relevant to various medical issues during pregnancy, not just gestational diabetes\u003c/li\u003e\n \u003cli\u003eDue to the retrospective nature of the study, there are concerns regarding missing data, which we will address with an effective strategy for managing them.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWith authorization from the Hormozgan University of Medical School Research Ethics Board (IR.HUMS.REC.1402.239), we will conduct a review of the medical records of every patient who delivered at our facility. The Ethics and Research Committee of Hormozgan University of Medical Sciences exempted the necessity for informed consent for involvement due to the study's retrospective design. Statistical analysis was performed while maintaining patient confidentiality and complying with ethics committee guidelines. The results of this study will be published in suitable scientific journals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData sharing is not relevant since no datasets have been generated or analyzed for this study at this time.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone declared.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHormozgan University of Medical Sciences.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFD and VM are in charge of protocol design and manuscript conception.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe wish to convey our heartfelt appreciation to Hormozgan University of Medical Sciences in Bandar Abbas, Iran for their steadfast support. We anticipate that, with your important cooperation and direction, this research will be carried out effectively in the future.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eACOG Practice Bulletin No. Gestational Diabetes Mellitus. Obstet Gynecol. 2018;190(2):e49\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/AOG.0000000000002501\u003c/span\u003e\u003cspan address=\"10.1097/AOG.0000000000002501\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSartayeva A, Danyarova L, Begalina D, Nurgalieva Z, Baikadamova L, Adilova G. Gestational Diabetes: Prevelence and Risk for the Mother and Child (Review). Georgian Med News. 2022 Jul-Aug;(328\u0026ndash;329):47\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDarsareh F, Jahromi M, Ranjbar A, Shekari M, Mehrnoush V, Roozbeh N. Pregnancy and Childbirth Outcomes of Gestational Diabetes Mellitus: A Retrospective Cohort Study. Dis Diagn. 2024;13(4):151\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.34172/ddj.1608\u003c/span\u003e\u003cspan address=\"10.34172/ddj.1608\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTehrani FR, Naz MSG, Bidhendi-Yarandi R, Behboudi-Gandevani S. Effect of Different Types of Diagnostic Criteria for Gestational Diabetes Mellitus on Adverse Neonatal Outcomes: A Systematic Review, Meta-Analysis, and Meta-Regression. Diabetes Metab J. 2022;46(4):605\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4093/dmj.2021.0178\u003c/span\u003e\u003cspan address=\"10.4093/dmj.2021.0178\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Epub 2022 Mar 8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeir TL, Majumder M, Glastras SJ. A systematic review of the effects of maternal obesity on neonatal outcomes in women with gestational diabetes. Obes Rev. 2024;25(7):e13747. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/obr.13747\u003c/span\u003e\u003cspan address=\"10.1111/obr.13747\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSartayeva A, Danyarova L, Begalina D, Nurgalieva Z, Baikadamova L, Adilova G. Gestational Diabetes: Prevelence and Risk for the Mother and Child (Review). Georgian Med News. 2022 Jul-Aug;(328\u0026ndash;329):47\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuintanilla Rodriguez BS, Vadakekut ES, Mahdy H, Gestational D. 2024 Jul 14. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 Jan\u0026ndash;. PMID: 31424780.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSert UY, Ozgu-Erdinc AS. Gestational Diabetes Mellitus Screening and Diagnosis. Adv Exp Med Biol. 2021;1307:231\u0026ndash;55. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/5584_2020_512\u003c/span\u003e\u003cspan address=\"10.1007/5584_2020_512\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaliskan R, Atis A, Aydin Y, Acar D, Kiyak H, Topbas F. PAPP-A concentrations change in patients with gestational diabetes. J Obstet Gynaecol. 2020;40(2):190\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/01443615.2019.1615041\u003c/span\u003e\u003cspan address=\"10.1080/01443615.2019.1615041\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFayaz L, Joseph R, Ankayarkanni B, Princemary S, Asha P, Student U. Multi-scale and context aware optimized glucose prediction using neural networks. In 2023 7th International Conference on Trends in Electronics and Informatics (ICOEI). 820\u0026ndash;825 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoujarzadeh B, Ranjbar A, Banihashemi F, Mehrnoush V, Darsareh F, Saffari M. Machine learning approach to predict postpartum haemorrhage: a systematic review protocol. BMJ Open. 2023;13(1):e067661. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/bmjopen-2022-067661\u003c/span\u003e\u003cspan address=\"10.1136/bmjopen-2022-067661\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBanaei M, Roozbeh N, Darsareh F, Mehrnoush V, Farashah MSV, Montazeri F. Utilizing machine learning to predict the risk factors of episiotomy in parturient women. AJOG Glob Rep. 2024;5(1):100420. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.xagr.2024.100420\u003c/span\u003e\u003cspan address=\"10.1016/j.xagr.2024.100420\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRanjbar A, Taeidi E, Mehrnoush V, Roozbeh N, Darsareh F. Machine learning models for predicting pre-eclampsia: a systematic review protocol. BMJ Open. 2023;13(9):e074705. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/bmjopen-2023-074705\u003c/span\u003e\u003cspan address=\"10.1136/bmjopen-2023-074705\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaeidi E, Ranjbar A, Montazeri F, Mehrnoush V, Darsareh F. Machine Learning-Based Approach to Predict Intrauterine Growth Restriction. Cureus. 2023;15(7):e41448. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7759/cureus\u003c/span\u003e\u003cspan address=\"10.7759/cureus\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRanjbar A, Ghamsari S, Boujarzade B, Mehrnoush V, Darsareh F. Predicting risk of postpartum hemorrhage using machine learning approach: A systematic review. GOCM. 2023;3(3):170\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003edoi.org/10.1016/j.gocm.2023.07.002\u003c/span\u003e\u003cspan address=\"10.1016/j.gocm.2023.07.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalakooti N, Mehrnoush V, Abdi F, Farashah MSV, Darsareh F. Development of a machine learning model to identify the predictors of the neonatal intensive care unit admission. Sci Rep. 2025;15(1):20914. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-025-06651-0\u003c/span\u003e\u003cspan address=\"10.1038/s41598-025-06651-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSafarzadeh S, Ardabili NS, Farashah MV, Roozbeh N, Darsareh F. Predicting mother and newborn skin-to-skin contact using a machine learning approach. BMC Pregnancy Childbirth. 2025;25(1):182. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12884-025-07313-9\u003c/span\u003e\u003cspan address=\"10.1186/s12884-025-07313-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEsmaeilyfard R, Bayati M. Enhancing AI-driven forecasting of diabetes burden: a comparative analysis of deep learning and statistical models. Sci Rep. 2025;15:29137. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003edoi.org/10.1038/s41598-025-14599-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-025-14599-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay \u0026Eacute;. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1145/2786984.2786995\u003c/span\u003e\u003cspan address=\"10.1145/2786984.2786995\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErickson BJ, Kitamura F. Magician's Corner: 9. Performance Metrics for Machine Learning Models. Radiol Artif Intell. 2021;3(3):e200126. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1148/ryai.2021200126\u003c/span\u003e\u003cspan address=\"10.1148/ryai.2021200126\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Machine learning, gestational diabetes, prognostic models","lastPublishedDoi":"10.21203/rs.3.rs-7583974/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7583974/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground \u003c/strong\u003eGestational diabetes (GDM) is linked to numerous negative pregnancy results for both the mother and the infant.\u003cstrong\u003e \u003c/strong\u003eAccurate predictions of GDM trends are essential for public health. Machine learning models have become effective instruments for disease prediction, showing enhanced performance compared to conventional techniques by identifying complex temporal relationships and nonlinear trends within health data. Consequently, we aimed to evaluate the power of machine learning compared to conventional statistical techniques in predicting GDM.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods \u003c/strong\u003eThe birth records of all expectant mothers who were admitted from January 2020 to January 2022 to one of the primary referral tertiary centers situated in Bandar Abbas, Iran, will be incorporated. We will employ 2 methods for analysis. In the initial stage, we will employ conventional analytical techniques. The chi-square test will evaluate the relationship between categorical variables and GDM. Bivariate logistic regression will be performed to analyze the risk factors for GDM, estimating crude odds ratios (cORs) along with their 95% confidence intervals (CIs). At the second level, we will employ machine learning techniques to forecast GDM. The input data will be utilized in eight machine learning models. To assess the diagnostic capability of each model, the area under the receiver operating characteristic curve (ROC AUC), accuracy, precision, sensitivity, specificity, and F1 score will be calculated.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscussion\u003c/strong\u003e The findings of this research show the hospital's existing incidence rates for gestational diabetes. Recognizing the risk factors for gestational diabetes forms the foundation for strategizing preventive actions against it.\u003c/p\u003e","manuscriptTitle":"Comparison the power of machine learning methods against traditional statistical approaches in predicting gestational diabetes: a study protocol","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-05 14:42:19","doi":"10.21203/rs.3.rs-7583974/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"53a98089-d3f8-45f2-bc51-076e7058022b","owner":[],"postedDate":"February 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-05T14:42:19+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-05 14:42:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7583974","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7583974","identity":"rs-7583974","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}