Breast Cancer detection from Thermograms Using Feature Extraction and Machine Learning Techniques

Abstract

Presence of tumors in breasts have lead to possibility of occurrence of cancer in a global level. It’s diagnosis is one of the challenging tasks. Researchers have come across a technique named thermography, which overcomes the drawbacks of a conventional technique i.e., mammography. In thermography, the early diagnosis of the breast cancer is carried out by implementing an analytical infrared thermal imaging techniques. This work focuses on different features-based machine learning techniques namely Support Vector Machine (SVM), k-Nearest Neighbour (KNN), Random Forest (RF) and Decision Tree (DT) to classify the images and detect possibility of cancerous image. This study also discusses about the SIFT and SURF features extraction techniques and a critical analysis of performance of various machine learning techniques have been presented.