Multi-Modal Hybrid Ensemble Framework for Breast Cancer Prognosis: Integrating Advanced Feature Engineering with Stacked Meta-Learning for Enhanced Clinical Decision Support

Breast cancer remains a significant global health issue and is one of the leading causes of cancer-related mortality among women. Although state-of-the-art deep learning models such as Dense Net, Inception Net, and NASNet have demonstrated notable success in detection tasks, their reliance on single-modal imaging data significantly limits prognostic accuracy and clinical applicability. To address this critical gap, this study proposes a novel multi-modal hybrid ensemble framework that leverages advanced feature engineering techniques in conjunction with stacked metalearning to integrate diverse clinical tabular data. By combining the strengths of multiple data modalities and robust machine learning strategies, the proposed approach aims to deliver more informed, reliable, and accurate breast cancer prognosis, ultimately supporting better patient outcomes and clinical decision-making.