Detection of Breast Cancer Thermograms based on Asymmetry Analysis using Texture features

Abstract

Breast Cancer is one of the most frequent diseases that occurs in every third woman in the world. Different measures of technology help in diagnosing the presence of the tumor in the breast. One of the most effective and early diagnosis methods is Thermography which records temperature values of the surface and creates an image called thermogram. In this paper, images are extracted from the temperature matrix, dataset available at DMR visual labs and the texture features based on different gray levels are extracted. The variation in temperature for the left and right breast is observed on the basis of asymmetry analysis. For classification, the performance of five models such as Support Vector Machine, Decision Tree, Random Forest, Bagging Classifier and Artificial Neural Network are critically assessed.