Brain Tumour Segmentation and Volume Estimation using Efficient Convolution Neural Network for MRI Images

A brain tumor is a type of blood clot in the form of cerebral disease. A brain tumor is a way to view and Magnetic Resonance Imaging (MRI) image detail is required. It is difficult to distinguish between normal tissue and tumor tissue from the brain. To should be an accurate analysis of brain tumours. Brain tumour segmentation is one of the most important and difficult tasks in the medical imaging field. Manual classification with human assistance can lead to incorrect predictions and diagnoses. Moreover, it is a difficult task when there is a large amount of data to help. Brain tumours are highly heterogeneous in appearance, and similarities between tumor and normal tissue make it difficult to extract tumour regions from images. In this paper, we proposed a method for extracting brain tumours from 2D magnetic resonance brain images (MRI) by a Fuzzy C-Means clustering algorithm with a traditional classifier and a convolutional neural network. Initially the pre-processing was carried out to apply the filter for image normalization with wiener filter and Robust active shape model was applied to predict the voluminous of data presence and segmentation was carried out through fuzzy c-means clustering and selected features was trained into optimize fuzzy c-means convolution neural network (FCM-CNN). This proposed system produces high detection accuracy in precision rate as well compared to other system.