Academy & Industry Research Collaboration Center (AIRCC)

Volume 10, Number 06, June 2020

Melanoma Detection in Histopathological Images using Deep Learning


Salah Alheejawi, Richard Berendt, Naresh Jha and Mrinal Mandal, University of Alberta, Canada


Histopathological images are widely used to diagnose diseases such as skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of abnormal cell nuclei would be very helpful for doctors to perform fast diagnosis. In this paper, we propose a technique, using deep learning algorithms, to first segment the cell nuclei in Hematoxylin and Eosin (H&E) stained images and detect the abnormal melanocytes on the histopathological images. The cell segmentation is done by using a novel Convolutional Neural Network (CNN) architecture. The segmented cells are then classified into melanoma and other nuclei using a Support Vector Machine classifier. Experimental results show that the CNN can segment the nuclei with more than 90% accuracy. The proposed technique has a low computational complexity.


Histopathological image analysis, Nuclei segmentation, Melanoma Detection, Deep learning.