AUTHOR=Ilari Ludovica , Piersanti Agnese , Göbl Christian , Burattini Laura , Kautzky-Willer Alexandra , Tura Andrea , Morettini Micaela 

TITLE=Unraveling the Factors Determining Development of Type 2 Diabetes in Women With a History of Gestational Diabetes Mellitus Through Machine-Learning Techniques

JOURNAL=Frontiers in Physiology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2022.789219

DOI=10.3389/fphys.2022.789219

ISSN=1664-042X

ABSTRACT=Gestational diabetes mellitus (GDM) is a type of diabetes that usually resolves at the end of the pregnancy but exposes to a higher risk to develop type 2 diabetes mellitus (T2DM). This study aimed to unravel the factors, among those that quantify specific metabolic processes, which determine progression to T2DM by using machine-learning techniques. Classification of women who did (labeled as PROG, n=19) vs. those who did not (labeled as NONPROG, n=59) progress to T2DM has been performed in Orange software through a data analysis procedure on a generated dataset including anthropometric data and a total of 34 features, extracted through mathematical modeling/methods procedures. Feature selection has been performed through decision tree algorithm and then Naïve Bayes and penalized (L2) logistic regression were used to evaluate the ability of the selected features to solve the classification problem. Performances have been evaluated in terms of area under the operating receiver characteristics (AUC), classification accuracy (CA), precision, sensitivity, specificity and F1. Feature selection provided 6 features and basing on them, classification performances were: AUC (0.795, 0.831, 0.884); CA (0.827, 0.813, 0.840); precision (0.830, 0.854, 0.834); sensitivity (0.827, 0.813, 0.840); specificity (0.700, 0.821, 0.662); F1 (0.828, 0.824, 0.836) for tree, Naïve Bayes, and penalized logistic regression, respectively. Fasting glucose, age, and body mass index together with features describing insulin action and secretion may predict the development of T2DM in previous GDM women.