Breast Cancer with Deep Learning Using Feature Selection: A Systematic Literature Review

Abstract

Among the most hazardous illnesses for people is cancer, yet there is currently no long-term treatment available. One of the typical cancers is breast cancer. More than 276,000 new instances of invasive breast cancer and much more than 48,000 non-invasive instances were detected in India in 2022 alone, based on the National Breast Cancer Foundation. Considering that 64% of these instances are discovered early in the course of the illness, patients have a 99% probability of surviving. A systematic review was performed to understand various deep learning algorithms/models, classes and classification of diagnosis, accuracy rate based on those algorithms/models, databases, methods and performance evaluation parameters. Though various research articles from reputed journals have been reviewed, a final in-depth review was considered on 45 numbers of papers by eliminating the irrelevant research articles based on some filtration criteria. The results and discussions are provided which reveals the current trends and adoptions by the various researchers conducting their research on breast cancer diagnosis using deep learning feature selection technology.