Paper Info
Title
A Multi-Objective Randomly Updated Beetle Swarm and Multi-Verse Optimization for Brain Tumor Segmentation and Classification
Abstract
This paper plans to develop the optimal brain tumor classification model with diverse intelligent methods. The main phases of the proposed model are '(a) image pre-processing, (b) skull stripping, (c) tumor segmentation, (d) feature extraction and (e) classification'. At first, pre-processing of the image is performed by converting the image from red green blue to gray followed by median filtering. Further, skull stripping is done for removing the extra-meningeal tissue from the head image, which is done by Otsu thresholding. As the main contribution, the tumor segmentation is done by the optimized threshold-based tumor segmentation using multi-objective randomly updated beetle swarm and multi-verse optimization (RBS-MVO). The objective constraints considered for the segmentation of the tumor is entropy and variance. Next, the feature extraction techniques like gray level co-occurrence matrix, local binary pattern and gray-level run length matrix is accomplished to extract the set of features. The classification side uses the combination of neural network (NN) and deep learning model called convolutional neural network (CNN) for tumor classification. The extracted features are subjected to NN, and the segmented image is taken as input to CNN. In addition, the weight function of NN and hidden neurons of CNN is optimized by the RBS-MVO.
Year
DOI
Venue
2022
10.1093/comjnl/bxab171
COMPUTER JOURNAL
Keywords
DocType
Volume
brain tumor segmentation, skill stripping, optimized threshold-based tumor segmentation, feature extraction, neural network, convolutional neural network, randomly updated Beetle Swarm and multi-verse optimization
Journal
65
Issue
ISSN
Citations 
4
0010-4620
0
PageRank 
References 
Authors
0.34
0
2
Name
Order
Citations
PageRank
Katukuri Arun Kumar100.34
Ravi Boda200.34