Breast cancer early diagnosis based on hybrid strategy

Abstract

The frequent occurrence of breast cancer and its serious consequences have attracted worldwide attention in recent years. Problems such as low rate of accuracy and poor self-adaptability still exist in traditional diagnosis. In order to solve these problems, an AdaBoost-SVM classification algorithm, combined with the cluster boundary sampling preprocessing techniques (CBS-AdaBoost-SVM), is proposed in this paper for the early diagnosis of breast cancer. The algorithm uses machine learning method to diagnose the unknown image data. Moreover, not all of the characteristics play positive roles for classification. To address this issue the paper delete redundant features by using Rough set attribute reduction algorithm based on the genetic algorithm (GA). The effectiveness of the proposed methods are examined on DDSM by calculating its accuracy, confusion matrix, and receiver operating characteristic curves, which give important clues to the physicians for early diagnosis of breast cancer.