Kidney tumor segmentation: A review

The objective of this paper is to review widely used imaging modalities, mainly Magnetic Resonance images (MRI) along with Computed Tomography (CT), which are used to record kidney tumor. We have also studied four types of tumors which can be broadly classified as Clear cell type, Oncocytomas, Papillary tumor and Chromophobe tumor. We have reviewed the possibility of using Convolutional Neural Network (CNNs) for detecting and segmenting the malignant and benign tumors. In recent years, several authors have proposed multiple techniques to segment kidney and tumors in medical images acquired from various imaging systems. We have studied a number of articles to find out the possibility of how to identify tumor through segmentation approaches based on CNN architectures, as CNNs outperforms the other image processing techniques which can help to attain better diagnosis.