Abstract:
Obesity is a rapidly increasing global health concern linked to numerous chronic diseases and significant healthcare challenges, emphasizing the need for accurate early prediction and intervention strategies. This study aims to develop an effective predictive model for obesity risk using a hybrid deep learning approach that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) architectures. The proposed model utilizes health-related data, where CNN is employed for efficient feature extraction and LSTM captures temporal and sequential patterns within the dataset. The model performance was evaluated using standard metrics, including accuracy, precision, recall, F1-score, and ROC-AUC, along with cross-validation techniques to ensure robustness. The experimental findings indicate that the hybrid CNN-LSTM model achieves superior predictive performance compared to conventional machine learning models and individual deep learning approaches. The results demonstrate improved classification accuracy and reliability in identifying individuals at risk of obesity. This study highlights the potential of integrating advanced deep learning techniques into clinical decision support systems, enabling early diagnosis, effective risk stratification, and informed preventive healthcare interventions to reduce the burden of obesity.
