A Deep CNN Approach Using Thermal Imagery for Breast Cancer Diagnosis

The aim of this research is to develop Deep Convolutional Neural Networks (DCNN) to diagnose the breast cancer using infrared thermal images. Four DCNN models (AlexNet, VGGNet, ResNet50 and GoogleNet) have been evaluated for cancer diagnosis using the Virtual Lab database and they have demonstrated their high performance in terms of medical image classification. The binary decision corresponds to two classes: malignant cancer and benign cancer. We have found that the considered DCNNs are providing very good accuracy in the breast cancer diagnosis using infrared thermography.