Empirical Analysis on Breast Cancer Datasets with Machine Learning

Breast cancer is a global health problem that requires early detection and precise diagnosis for better patient outcomes. Recent advances in machine learning and medical imaging, along with large datasets, have transformed breast cancer research. This in-depth examination digs into these datasets and their various applications in detection and diagnosis. While existing research focus on breast cancer, a thorough examination of available datasets is still required. This review tries to close the gap by thoroughly investigating datasets such as WDBC, BreakHis, Mammography, Ultrasound, and Thermography. It goes beyond technical specifics to investigate their diverse applications, ranging from feature-based classification to image-based diagnostics. The major goal is to better understand how databases help with breast cancer detection and diagnosis. This study demonstrates that each dataset has a distinct and important role.