Breast cancer recurrence remains a critical concern in oncology, requiring accurate predictive models to enhance patient outcomes. This study presents a comparative analysis of machine learning and deep learning techniques for predicting breast cancer recurrence using the METABRIC dataset. The dataset consists of 33 clinical and genetic features from 2,500 patients, augmented to 4,000 samples through image processing. We employed 12 machine learning models, including Random Forest, Gradient Boosting, and AdaBoost, alongside a custom deep learning model. The dataset was split into 60% training, 20% validation, and 20% testing subsets. Performance was evaluated using F1-score, recall, precision, and ROC-AUC metrics. Among the machine learning models, Random Forest achieved the highest performance with an F1-score of 96.00%, recall of 96.30%, and precision of 96.23%. Our proposed deep learning model outperformed the machine learning models, achieving an F1-score of 98.10%, recall of 98.59%, and precision of 98.76%. These results demonstrate the efficacy of deep learning models for breast cancer recurrence prediction, suggesting their potential for improving clinical decision-making and patient management.