Breast tumour detection using machine learning: review of selected methods from 2015 to 2021

Breast tumour is one of the leading causes of death among women worldwide. Researchers are working hard to develop early and improved detection tools for breast tumour. Several innovations lead to the decline in the mortality rate for this lethal illness, but breast amputation and early death diagnosis contributed the most to preventing disease transmission. Early detection of a breast tumour allows for the most effective treatment. By using several different techniques, imaging of breast cancer can be done and some of these techniques are X-ray, MRI, CT, Ultrasonography, and now Molecular Imaging. This paper examines similar works that use Mammography, X-ray, Ultrasound, Biopsy and Artificial Intelligence, highlighting their benefits and limitations, as well as open issues and research challenges. In the literature, a variety of machine learning, artificial neural networks, and deep learning models were employed to process thermographic or mammographic images of breast tumour, including, Support Vector Machine, Bayes Net, decision tree, K-Nearest Neighbors (KNN), Deconvolutional Neural Networks (DNN), Convolutional Neural Networks (CNN) and CAD system. This study also discusses the different datasets used for breast tumour detection.