Classification of Benign and Malignant Colorectal Polyps using Pit Pattern Classification
6 Pages Posted: 23 Mar 2020
Date Written: February 21, 2020
Abstract
Colorectal cancer is the third largest cancer in the world and is second most common cause of cancer related death in the world. It has led to a large number of deaths because of not being treated on time. It originates in the form of polyps which are not cancerous. If treated on time, it can be cured otherwise it grows and converts into cancerous form and may lead to death. In this paper, we propose a deep learning technique to classify benign tumor from malignant tumor. Benign tumor is not dangerous, if removed on time as they don’t grow back after being removed whereas malignant cancer can grow back sometimes. In this paper, we propose a deep learning technique to classify benign tumor from malignant tumor. We trained our deep convolution model with a private dataset and obtained 87% accuracy. The results obtained shows that deep learning techniques can be used for timely detection of colorectal cancer so that it can be treated on time. Further, it can also be used by specialists in polyp segmentation.
Keywords: Colorectal Cancer, Deep Learning, Segmentation, Convolutional Neural Network
JEL Classification: O30
Suggested Citation: Suggested Citation