DEEPHEALTHNET: ADOLESCENT OBESITY PREDICTION SYSTEM BASEDON DEEP LEARNINGFRAMEWORK

Abstract

Adolescent obesity has emerged as a critical global health concern, leading to severe long-term complications such as cardiovascular diseases, diabetes, and psychological disorders. Early prediction and intervention are essential to mitigate these risks. This project, “DEEPHEALTHNET: Adolescent Obesity Prediction System Based on Deep Learning Framework,” proposes an intelligent and scalable solution that leverages advanced deep learning techniques to accurately predict obesity risk among adolescents. The system integrates multi-dimensional health data, including demographic information, lifestyle habits, dietary patterns, physical activity levels, and clinical parameters. A hybrid deep learning architecture is employed, combining models such as Artificial Neural Networks (ANNs) and Convolutional Neural Networks (CNNs) to capture both structured and complex feature relationships within the dataset. The framework utilizes data preprocessing, feature normalization, and dimensionality reduction techniques to enhance model performance and generalization. DEEPHEALTHNET is designed to provide real-time prediction through a user-friendly interface, enabling healthcare professionals, parents, and educators to assess obesity risk at an early stage. The model is trained and evaluated using performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC, ensuring high reliability and robustness. Additionally, the system incorporates explainable AI components to interpret predictions and highlight key contributing factors influencing obesity risk. This research contributes to the development of preventive healthcare systems by combining deep learning, predictive analytics, and health informatics. The proposed solution not only improves prediction accuracy but also supports data-driven decision-making for personalized intervention strategies, ultimately promoting healthier lifestyles among adolescents