Study on Machine Learning and Deep Learning Methods for Cancer Detection

Abstract

Cancer causes death of about million people every year. Cancer is the frequently recognized and is the major reason of death in men and women. Cancer is a group of diseases involving abnormal cell growth which will spread to other parts of the body. Colonography makes use of low dose radiation Computed tomography (CT) scanning to get an internal view of the cancer tumors making use of special x-ray machine to view tumors. Radiologists examine these images to find tumor like structure using computer tools. As CT Colonography image contain noise such as lungs, small intestine, instruments during image capturing. Cancer occurrence can be detected mainly using shape feature; eliminating shapes similar to tumor is challenging. Hence, to tackle above issues, image processing techniques are used by applying deep learning algorithm- Convolution Neural Network (CNN) and the results are compared with classical machine learning algorithm. The analysis is done with classical machine learning algorithms - Random Forest algorithm (RF) and k-nearest neighbour algorithm (KNN) by extracting texture feature- Local binary pattern (LBP) and shapefeature - Histogram oriented gradient (HOG) for comparison.