Carleton University - School of Computer Science Honours Project
Fall 2019
HISTOPATHOLOGICAL BREAST CANCER CLASSIFICATION BY NEURAL NETWORK USING SUPER-IMAGE RESOLUTION
Vincent Wang
SCS Honours Project Image
ABSTRACT
Over the past couple of years, with the continuous development of deep learning, there has been a rise in the success of using deep learning for medical image analysis. Deep learning in the medical field is used to identify patterns, analysis images, classify and segment medical images. As most of the task is based on the images, the conventional neural network is now being popularly used for medical image processing and classification. This project uses two convolutional networks models, which are Super-Resolution Convolutional Neural Network (SRCNN) and a classifier model based on Convolutional Neural Network (CNN), to proceed the image pre-processing and classification with breast cancer histopathological images. The proposed idea of this project is to investigate the potential improvement of the prediction accuracy of SRCNN enhanced images.