Breast cancer classification and prediction methods by employing machine and deep learning approaches-A survey

Abstract

Breast carcinoma stands as one of the most perilous afflictions affecting females, lacking an effective treatment to date. Recent advancements in deep learning techniques, coupled with artificial intelligence (AI), have demonstrated promising results in breast cancer identification. This innovation facilitates early detection, consequently enhancing patient survival rates. Deep learning necessitates minimal human intervention for feature extraction, contrasting with traditional machine learning methods. The ML and DL techniques are practised and comparison of all these techniques were shown. Specifically, emphasis is placed on genomic and histopathologic imaging data. Various algorithms, including R, SVM, logistic regression, KNN, Naïve Bayes, CNN, and ANN, are thoroughly researched and valuated to gauge their efficacy. Furthermore, many screening protocols were deployed to identify and examine the datasets. Lastly, the paper explores the challenges encountered and the most possible directions to detect the breast cancer .Hence researchers and clinicians with a thorough understanding and insights into this deep learning domain.