Optimizing Breast Cancer Detection: Machine Learning for Pectoral Muscle Segmentation in Mammograms

For timely identification and treatment in the field of medical imaging, accurate breast cancer segmentation from mammograms is crucial. To improve the identification of breast cancer from mammography images, this paper propose an ensemble model by combining various various machine learning models. In the first step, a series of pre-processing steps such as image enhancement, normalization, noise removal and artifact removal techniques are applied to improve image quality. Then, a variety of machine learning algorithms, such as Support Vector Machines (SVM) with linear, polynomial, and Radial Basis Function (RBF) kernels, Random Forest, XGBoost, and K-Nearest Neighbors (KNN). Effectiveness of the proposed ensemble model in segmenting and identifying breast cancer is evaluated using MIAS benchmark dataset. The proposed model achieves an accuracy of 93.27%. The results show that ensemble model has a great deal of potential to improve mammography image segmentation and analysis, which could help radiologists diagnose breast cancer. The groundwork for further investigation into advanced modeling methods is laid by this study, which emphasizes the revolutionary power of machine learning in improving oncology outcomes for patients and diagnostic procedures.