Using Deep Learning Boosting with Survival Analysis for Breast Cancer Diagnosis
Mahdi Saber Raza, Abbas M. Ali, Chiman Haydar Salh · ZANCO Journal of Pure and Applied Sciences · 2025
Around the world, women are diagnosed with breast cancer, which is a deadly disease. Early detection is very important in helping improve the survival rate and advance the excellence of care for patients. With the help of deep learning models, medical image analysis can be performed with high accuracy and can even be used to automatically identify breast cancer. Several research has been undertaken on the application of survival analysis and deep learning models for the detection of breast cancer. This work introduces a methodology that can effectively and precisely diagnose breast cancer with reliability. This approach utilizes survival analysis and Deep learning, enhanced with statistical techniques such as the Cox regression model and Kaplan Meier Method. Additionally, a dataset of breast cancer magnetic resonance imaging (MRI) has been created and refined. The study findings demonstrated that the utilized model for the survival analysis of breast cancer exhibited excellent performance, achieving a remarkable accuracy rate of 98%. The improved Mask R-CNN with the scanner known as enabled the comparison of tumor size and breast size.