Smart Cancer Diagnosis using Machine Learning Techniques

Abstract

Breast cancer (BC) is one of the most common cancers among women worldwide. According to world statistics, these are the majority of new cancers and deaths related to cancer, making them an important public health problem in today's society. Early diagnosis of breast cancer can significantly improve prognosis and survival as it promotes timely medical treatment of patients. Additional unnecessary treatments can be avoided by accurately classifying benign and malignant tumors. Therefore, the correct and correct diagnosis of breast cancer tumors and the classification into benign or malignant categories is an important field of research. Machine learning is emerging as a method of choice in the classification of breast cancer patterns and in the predictive model because of its advantages in detecting features from complex breast cancer data sets. In this project various techniques for extracting properties are used, such as: For example, the local binary pattern, scalar invariant property transformation (SIFT), and oriented FAST and rotated LETTER (ORB), GLCM, PFTAS. Thereafter, the machine learning techniques in breast cancer prognosis will be reviewed. The project provides a general description of the machine learning techniques, e.g. The support vector machine, the neural networks, the random structure and the decision tree as well as the first nearest neighbor. The primary data of the project comes from the breast cancer database BreakHis (BH).