Study of the Genetic Basis of Childhood and Adolescent Obesity with Stress Through the Analysis of Multidimensional Data with Machine Learning and Artificial Intelligence Tools.

Childhood and adolescent obesity is a growing health concern with complex multifactorial origins, encompassing genetic, environmental, physiological, and psychosocial factors. In Greece, the prevalence of childhood obesity is among the highest in Europe, indicating an urgent need to understand its underlying mechanisms. Herein, we explore the genetic basis of obesity, focusing on both monogenic and polygenic factors, and how early life stressors contribute to obesity's onset and progression. Genetic predispositions, such as mutations in leptin-melanocortin pathways, and the role of epigenetic modifications influenced by environmental factors, are examined to understand obesity's complexity. Moreover, stress-related hormonal dysregulation impacts metabolic pathways, exacerbating weight gain and obesity-related complications. Through advanced algorithms like neural networks, decision trees, and clustering techniques, ML/AI approaches have demonstrated high accuracy in predicting obesity, identifying genetic markers, and analyzing interactions between genetic and lifestyle factors. These technologies hold promise for early detection, personalized interventions, and the development of targeted prevention strategies. The integration of ML/AI with genomic, epigenomic, and clinical data offers a comprehensive understanding of childhood obesity, paving the way for more effective management and treatment. This study contributes to a deeper understanding of the genetic and environmental factors in childhood obesity and highlights the potential of AI-driven approaches in addressing this critical public health challenge.