Title: Brain tumour detection using self-adaptive learning PSO-based feature selection algorithm in MRI images

Authors: A.R. Kavitha; C. Chellamuthu

Addresses: Department of IT, Jerusalem College of Engineering, Chennai, India ' Department of EEE, R.M.K. Engineering College, Chennai, India

Abstract: In this paper, we propose a brain tumour classification scheme to classify the breast tissues as normal or abnormal. At first, we segment the region of interest (ROI) from the medical image using modified region growing algorithm (MRGA). Feature matrix is generated using grey-level co-occurrence matrix (GLCM) to the entire detailed coefficient from 2D-DWT of the region of interest (ROI). To derive the relevant features from the feature matrix, we take the self-learning particle swarm optimisation (SLPSO) algorithm. In SLPSO, four upgrading strategies are utilised to adaptively redesign the velocity of every particle to guarantee its differences and robustness. The relevant features are used in a feed forward neural network (FFNN) classifier for classification. The method yield very encouraging result in terms of classification accuracy using a neural network. In experimental result most cases, the classification accuracy improved on previously reported results.

Keywords: region of interest; ROI; modified region growing; MRG; co-occurrence matrix; grey-level co-occurrence matrix; GLCM; 2D-DWT; self-learning particle swarm optimisation; SLPSO; features; feed forward neural network; FFNN; classification.

DOI: 10.1504/IJBIDM.2019.100469

International Journal of Business Intelligence and Data Mining, 2019 Vol.15 No.1, pp.71 - 97

Received: 13 Mar 2017
Accepted: 30 Apr 2017
Published online: 29 Jun 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article