Article: Segregation of MRI brain image using hybrid evolutionary clustering algorithm Journal: International Journal of Biomedical Engineering and Technology (IJBET) 2015 Vol.18 No.1 pp.30 - 51 Abstract: The present research is on computer-aided classification of the magnetic resonance brain images. The method proposed congregates on colour-converted hybrid clustering segmentation algorithm with hybrid feature selection approach based on Information Gain and Sequential Forward Floating Search (IGSFFS) and Multi-Class Support Vector Machine (MC-SVM) classifier technique to segregate the magnetic resonance brain images into three categories namely normal, benign and malignant. The present research acknowledges the colour-converted segmentation by new hybrid evolutionary clustering algorithm which is the mixture of weighted firefly and K-means algorithm to overcome local optima problems in firefly algorithm. Further random cluster initialisation is also modelled. The results of the simulation show that the performance of the proposed algorithm has better segmentation accuracy than the other algorithms such as colour-converted PSO-K-means and K-means clustering algorithm. The performance of the method is evaluated using classification accuracy, sensitivity, specificity and Receiver Operating Characteristic (ROC) curves. The results show that the highest classification accuracy of greater than 98% is obtained for the proposed diagnostic model, and this is very promising compared to the previously reported results. Inderscience Publishers - linking academia, business and industry through research

Title: Segregation of MRI brain image using hybrid evolutionary clustering algorithm

Authors: N. Rajalakshmi; V. Lakshmi Prabha

Addresses: Sanson Engineers Ltd., Coimbatore, Tamil Nadu, India ' Government College of Technology, Coimbatore 641 013, Tamil Nadu, India

Abstract: The present research is on computer-aided classification of the magnetic resonance brain images. The method proposed congregates on colour-converted hybrid clustering segmentation algorithm with hybrid feature selection approach based on Information Gain and Sequential Forward Floating Search (IGSFFS) and Multi-Class Support Vector Machine (MC-SVM) classifier technique to segregate the magnetic resonance brain images into three categories namely normal, benign and malignant. The present research acknowledges the colour-converted segmentation by new hybrid evolutionary clustering algorithm which is the mixture of weighted firefly and K-means algorithm to overcome local optima problems in firefly algorithm. Further random cluster initialisation is also modelled. The results of the simulation show that the performance of the proposed algorithm has better segmentation accuracy than the other algorithms such as colour-converted PSO-K-means and K-means clustering algorithm. The performance of the method is evaluated using classification accuracy, sensitivity, specificity and Receiver Operating Characteristic (ROC) curves. The results show that the highest classification accuracy of greater than 98% is obtained for the proposed diagnostic model, and this is very promising compared to the previously reported results.

Keywords: MRI scans; brain images; magnetic resonance imaging; biomedical engineering; colour-converted segmentation; K-means clustering; feature selection; firefly algorithm; multi-class SVM; MC-SVM; support vector machines; image segmentation; modelling; simulation.

DOI: 10.1504/IJBET.2015.069851

International Journal of Biomedical Engineering and Technology, 2015 Vol.18 No.1, pp.30 - 51

Received: 24 Sep 2014
Accepted: 21 Dec 2014
Published online: 14 Jun 2015 *

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

Title: Segregation of MRI brain image using hybrid evolutionary clustering algorithm

Keep up-to-date