Image Processing Based Early Breast Cancer Detection in Mammography Images Using GRU and XGBoost Approach

One out of every eight women will be affected by breast cancer at some point in their lives, making it one of the most common malignancies in females. The most reliable method for finding breast cancer in its early stages is mammography. Due to the low visibility and weak contrast of mammographic photos, early identification of breast cancer is crucial for effective therapy. As a result of several computer-aided detection approaches, radiologists are now able to make more precise diagnoses. Operating Sequencing is crucial for several tasks, including morphology, preprocessing, segmentation, feature selection, and training models. The suggested method makes use of preprocessing technologies that can enhance mammography, remove noise, and identify pectoral muscles. Using image processing techniques, a digital image can be “segmented” into many layers. Acquiring a precise image of the mass is the goal of the morphological process. Geometric algorithms are employed by the feature selection technique. Training a GRU-XGBoost model requires careful consideration of which features to use. The proposed method outperforms the current gold standard algorithms, GRU and XGBoost, in every respect. The results showed a remarkable improvement, with an accuracy of 95.63%.