Brain tumor diagnosis using image processing: A survey

A literature overview is described on cerebrum (brain) tumor diagnosis. The aim of this survey is to provide an outline for those who are new to the field of image processing, and also to provide a reference for those searching for literature in this applications. Tumor is because of an abnormal development of cells (tissues) inside the brain. Magnetic Resonance Imaging (MRI), Computer Tomography (CT) imaging techniques are used for early detection of abnormal changes in tumor tissues or cells. Its correct detetcion and identification at an early stage is the only way to get cure. Brain tumor tissues may become malignant (cancerous) if not diagnosed at right time. A recent couple of years various image processing algorithms have been proposed for correct and efficient computer aided diagnosis of cerebrum tumors. An algorithm effectively work on CT, MRI images. It is been observed that an automatic segmentation method using Convolutional Neural Network (CNN) with 3∗3 kernels provide deeper architecture and positive results against overfitting. Watershed segmentation algorithm removes the salt & pepper noise without disturbing edges. It is very easy for automatic and accurate calculation of tumor area. Sobel edge detection based improved edge detection algorithm provide superior performance over conventional segmentation algorithm. The Otsu segmentation method for brain tumor makes the diagnosis and treatment planning more easy and accurate. Morphological operators can be used in the detection of tissues in the scan image of tumor. The use of PCA in optimizing the features obtained from segmented region can give the very good results as compared to other methods. The intensity based and wavelet based features are very useful for classification of benign and malignant tumors. Artificial Neural Network (ANNs), Support Vector Machine (SVM) based decision support system are reviewed.