Diagnosis of brain tumor using image processing and determination of its type with RVM neural networks

Abstract

Typically, the diagnosis of a tumor is done through surgical sampling, which is more precise with existing methods. The difference is that this is an aggressive, time consuming and expensive way. In the statistical method, due to the complexity of the brain tissues and the similarity between the cancerous cells and the natural tissues, even a radiologist or an expert physician may also be in error in his diagnosis. Tumor diagnosis is done automatically and various results are achieved. The steps involved in these algorithms can be divided into two sections of the feature discovery and the classification of the samples. The methods generally are that, firstly, the properties of the image are extracted. These characteristics usually include static properties such as entropy, skewness, mean, energy, torque, correlation, etc., or the properties of other algorithms (instant conversion, histogram, etc.). The information obtained at this stage is applied to the sample classification process for decision making. This section is done with an advanced neural network such as RVM. Possible neural networks have the ability to classify more than one class and a kind of radar disease to extract features from MRI images using histogram or satellite conversion techniques, and then selecting appropriate features and ultimately using the system. Fuzzy Neural Network Diagnostics The decision making system of the fuzzy system is a conclusion that trains with these features and in the output, multiple images are given at different levels. In this research, using image and image processing, we try to find out exactly where the brain is placed. For this purpose, it is initially performed using preventive techniques such as enhancement of contrast, marginalization and morphological functions, and then using the neural network to perform a careful separation of the cancerous parts of the brain health sectors.