Gestational diabetes mellitus is the term used to describe an elevated blood sugar condition that develops during pregnancy. It can occur at any stage of the pregnancy and cause issues during labor and after delivery for both the mother and the unborn child. Finding and testing for GDM risk factors can help improve the health of women and their kids by allowing interventions to take place. GDM raises the risk of long-term issues like obesity, poor glucose metabolism, and cardiovascular disease in both the mother and the kid. Machine learning (ML) techniques are increasingly being used to identify risk variables and facilitate GDM early detection. Multivariate logistic regression (LR) modeling is a well-known machine learning (ML) technique for predicting diabetes and its effects. Furthermore, a variety of methods, such as Random Forest (RF), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM), are employed to address GDM-related problems. These methods are being used in more studies to identify risk factors for GDM and create early disease prediction models. For machine learning models that forecast GDM, the pooled area under the receiver operating characteristic curve (AUROC) is obtained. Metrics for statistical diagnosis is obtained. ML algorithms are more attractive than traditional screening techniques for GDM prediction. We compare machine learning models with respect to both general and particular screening methods. It is evident from the results that machine learning outperforms alternative methods. The importance of precise evaluations and consistent diagnostic criteria must be emphasized.

digital health, gestational diabetes mellitus, screening, diagnosis, machine learning, prediction model, prognostic model.

Received: November 28, 2023; Accepted: March 6, 2024; Published: April 5, 2024

How to cite this article: Velu Chinnasamy Shanmugam, C. Vijayalakshmi, M. Mynarani and K. Pradeepa Veerakumari, A study utilizing advanced machine learning techniques to analyze gestational diabetes mellitus and its implementations, JP Journal of Biostatistics 24(2) (2024), 227-237. http://dx.doi.org/10.17654/0973514324015

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

References:
[1] X. Mao, X. Chen, C. Chen, H. Zhang and K. P. Law, Metabolomics in gestational diabetes, Clinica Chimica Acta 475 (2017), 116-127.
[2] M. L. Geurtsen, E. E. L. V. Soest, E. Voerman, E. A. P. Steegers, V. W. V. Jaddoe and R. Gaillard, High maternal early pregnancy blood glucose levels are associated with altered fetal growth and increased risk of adverse birth outcomes, Diabetologia 62(10) (2019), 1880-1890.
[3] C. E. Powe, Early pregnancy biochemical predictors of gestational diabetes mellitus, Current Diabetes Reports 17(2) (2017), 1-10.
[4] S. Shinar and H. Berger, Early diabetes screening in pregnancy, International Journal of Gynecology and Obstetrics 142(1) (2018), 1-8.
[5] D. D. Miller and E. W. Brown, Artificial intelligence in medical practice: the question to the answer? American Journal of Medicine 131(2) (2018), 129-133.
[6] A. Sumathi and S. Meganathan, Ensemble classifier technique to predict gestational diabetes mellitus (GDM), Computers Systems Science and Engineering 40(1) (2022), 313-325.
[7] Yunzhen Ye, Yu Xiong, Qiongjie Zhou, Jiangnan Wu, Xiaotian Li and Xirong Xiao, Comparison of machine learning methods and conventional logistic regressions for predicting gestational diabetes using routine clinical data: a retrospective cohort study, Journal of Diabetes Research 2020 (2020), 1-10.
[8] Y. Xiong, L. Lin, Y. Chen, S. Salerno, Y. Li, X. Zeng and H. Li, Prediction of gestational diabetes mellitus in the first 19 weeks of pregnancy using machine learning techniques, Journal of Maternal-Fetal and Neonatal Medicine 33(1) (2020), 1-8.
[9] Iswaria Gnanadass, Prediction of gestational diabetes by machine learning algorithms, IEEE Potentials 39(6) (2022), 32-37.