Mass and Microcalcification Detection in Mammograms: A Deep Learning Perspective on Breast Cancer Classification

Cancer is typified by the hysterical explosion of peculiar cells which violates standard regulations of cell division. If proliferation persists and disseminates, resulting in the creation of metastases, it may be lethal. Identifying and interpreting breast lesions in mammograms is a complicated task, even for experienced radiologists. Mammograms may exhibit subtle and overlapping structures, making detection and accurate classification of abnormalities such as masses, calcifications, and architectural distortions difficult. The variability in lesion appearance, dense breast tissue, and the need for precise differentiation between benign and malignant conditions add to the challenge. This review compares findings and classifying two primary forms of breast anomalies: mass and microcalcification generally observed in mammograms based on the latest developments in the auto-recognition and categorization of breast cancer in mammographs. The comprehensive review presented here offers a solid introduction to the field of breast cancer classification. The basic motivation behind this review article is to study the various DL-based breast cancer detection methods proposed in recent years for mass and calcification detection and classification. It addresses the challenges associated with mass, and microcalcification detection using deep learning techniques to progress diagnostic accurateness to guide future applications.