Mammogram breast cancer CAD systems for mass detection and classification: a review

• Published in: Multimedia tools and applications Vol. 81; no. 14; pp. 20043 - 20075
• Main Authors: Hassan, Nada M., Hamad, Safwat, Mahar, Khaled
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2022; Springer Nature B.V
• Subjects: Breast cancer; Classification; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Datasets; Deep learning; Image processing; Machine learning; Mammography; Multimedia Information Systems; Special Purpose and Application-Based Systems; Mass; Mammogram; Breast cancer; Detection; CAD system; Classification
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-022-12332-1

Although there is an improvement in breast cancer detection and classification (CAD) tools, there are still some challenges and limitations that need more investigation. The significant development in machine learning and image processing techniques in the last ten years affected hugely the development of breast cancer CAD systems especially with the existence of deep learning models. This survey presents in a structured way, the current deep learning-based CAD system to detect and classify masses in mammography, in addition to the conventional machine learning-based techniques. The survey presents the current publicly mammographic datasets, also provides a dataset-based quantitative comparison of the most recent techniques and the most used evaluation metrics for the breast cancer CAD systems. The survey provides a discussion of the current literature and emphasizes its pros and limitations. Furthermore, the survey highlights the challenges and limitations in the current breast cancer detection and classification techniques.