Comparative Study of Various Deep Convolutional Neural Networks in the Early Prediction of Cancer

In the recent decades, cancer has become a major cause of mortality worldwide. Predicting cancer cells and tumors at the early stages can be treated. Computer-aided diagnosis systems are used to analyze the MRI and CT scan images. However, it is inefficient to predict the disease as it works with high-level image features. It is important to extract the low-level feature of the image details in order to improve the prediction accuracy. Deep learning models are efficient in extracting the low-level image features. A convolutional neural network (CNN) is one of the popular deep learning architectures efficient in feature extraction. In this paper, the various types of CNN models are discussed. A comparative study of different CNN models along with segmentation and classification models are discussed. Finally, the prediction accuracy of CNN architectures with their dataset details are analyzed.