Convolutional Neural Network with Segmentation in Brain Tumour Diagnosis: An extensive review

Abstract

Convolutional Neural Network have been researched for diagnosis of Brain tumour. However, few techniques have been used in the real world because of various factors. The aim of this work is to introduced The Brain MRI Data, Segmentation process and Segmented Image Display (BDSSD) taxonomy, which describes the major components that are required to implement Convolutional Neural Network for brain tumour diagnosis. This taxonomy helps to segment different MRI image data using pre-processing and feature extraction process. The proposed model has been evaluated on the basis of state-of-art models. Thirty state-of art solutions have been selected and the proposed BDSD taxonomy is validated, evaluated and verified based on system completeness, recognition and comparison. The BDSSD taxonomy has been presented so that all aspect is included and explained based on Convolutional Neural Network which helps in the accurate segmentation of brain tumour using different accuracy measures such as dice coefficient.