Abstract:

Abstract This research demonstrates a computer-assisted approach for identifying and segmenting brain tumors, which is based on methods of image registration and classification. The method may be found here. Image registration, transformation using Contourlet, extracting features, feature normalization, feature classification, and feature segmentation are all elements that make up this suggested research. For the purpose of tumor detection and segmentation using brain MRI, we first use a Genetic Algorithm (GA) to optimize the features that have been extracted, and then we use an Adaptive Neuro Fuzzy Inference System's (ANFIS) classification method to classify the features that have been extracted. The method that has been proposed for diagnosing brain cancers is given a quantitative examination in which the method's sensitivity, specificity, segmented precision, precision, accuracy, and Dice similarity coefficient are measured. In addition, the findings of this research provide a strategy for developing a framework for the diagnosis of brain tumors by combining a number of different classification approaches. The clarity of the pictures' low-resolution boundaries may be improved by integrating brain MRI scans taken from a data collection that is freely accessible to the public..

Keywords: