Breast cancer detection using bimodal image fusion: Thermography and mammography images

Abstract

Author(s): Prabira Kumar Sethy*, S. Shanthi, Komma Anitha, A. Geetha Devi and Preesat Biswas

Breast cancer is known as one of the major causes of mortality among women. Breast cancer can be treated with better patient outcomes and significantly lower costs if it is detected early. There are many modalities of images are available for breast cancer diagnosis. Image fusion is a technique that combines the information collected from multiple source images. In this paper, a bimodal image fusion technique is proposed, where the mammography images and thermography images of breast are considered. The deep features of both images are collected by the three pre-trained network like Alexnet, vgg16 and vgg19 individually. The extracted features are merge using concatenation technique and then fed to support vector machine classifier for classification to discriminate between sick and healthy. The vgg16 with SVM using thermal images and mammography images outperform the other two and resulted accuracy of 0.9808, sensitivity of 1, specificity of 0.9615, precision of 0.963, FPR of 0.0385, F1 Score of 0.981, MCC of 0.9623 and Kappa of 0.9615.

Share this article