Application of Artificial Intelligence models in Public Health for the study of the interaction between exposome, biological aging, and maternal-child health

My PhD thesis focuses on exploring how maternal and environmental exposures influence maternal and child health, with particular attention to early biological aging and adverse outcomes. From the very beginning, an individual’s exposure starts in utero, shaped by environmental factors and social determinants that interact with the individual’s epigenetic mechanisms. Within the framework of precision medicine, this project aims to characterize the exposome during the peri-conceptional period and develop models to assess its interaction with early biological aging and health outcomes in mother-child pairs. The primary objective of this research is to integrate epidemiological data with advanced Artificial Intelligence (AI) techniques and insights from biomolecular and translational research. This integration seeks to evaluate environmental risk profiles, behavioral factors, and genetic susceptibility markers while also monitoring molecular markers such as those related to aging and epigenetics. Specifically, the research aims to understand how the exposome influences the early biological aging of the fetus, explore causal relationships between maternal exposures and adverse pregnancy or neonatal outcomes, and leverage AI to estimate the risk of these outcomes. Furthermore, it aims to assess the potential effectiveness of public health strategies and identify the most effective AI techniques for analyzing causal relationships in the context of population health. Epidemiological data and biological samples were collected from two Sicilian birth cohorts. Traditional linear and logistic regression models were used for the analysis, complemented by innovative causal machine learning models to explore the causal links between maternal exposures– such as Body Mass Index (BMI) and Gestational Weight Gain (GWG)– and adverse pregnancy outcomes.The results of this research will contribute to a deeper understanding of how the exposome impacts maternal and child health, particularly regarding early biological aging and adverse pregnancy outcomes. By identifying the key risk factors and causal mechanisms, this work will provide valuable insights for developing more effective public health interventions and strategies. Moreover, the integration of epidemiological, genetic, and epigenetic data with AI techniques represents a significant step forward in advancing precision medicine and optimizing maternal and child health outcomes.