Combined Spatial FCM Clustering and Swarm Intelligence for Medical Image Segmentation

Thiyam Ibungomacha Singh; Romesh Laishram  and Sudipta Roy

doi:10.17485/ijst/2016/v9i45/99962

Article

Combined Spatial FCM Clustering and Swarm Intelligence for Medical Image Segmentation

VIEWS 806
PDF 211

Abstract
Full-Text HTML
Full-Text PDF
How to Cite

Indian Journal of Science and Technology

DOI: 10.17485/ijst/2016/v9i45/99962

Year: 2016, Volume: 9, Issue: 45, Pages: 1-7

Original Article

Combined Spatial FCM Clustering and Swarm Intelligence for Medical Image Segmentation

Thiyam Ibungomacha Singh^{1 *}, Romesh Laishram² and Sudipta Roy³

¹Department of Computer Science and Engineering, Manipur Institute of Technology, Takyelpat - 795001, Manipur, India; [email protected] ²Department of Electronics and Communication Engineering, Manipur Institute of Technology, Takyelpat - 795001, Manipur, India; [email protected] ³Department of Computer Science and Engineering, Assam University, Silchar - 788011, Assam, India; [email protected]

*Author for correspondence
Thiyam Ibungomacha Singh Department of Computer Science and Engineering, Manipur Institute of Technology, Takyelpat - 795001, Manipur, India; [email protected]

This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

Objectives: The development of image processing tools for medical image processing has recently generated lot of interest. Medical image segmentation is one such area of focus for many researchers over the years. Methods/Statistical Analysis: In this work we have proposed an algorithm which is a combination of Fuzzy C-Means Clustering (FCM) with spatial constraints which is called spatial FCM (SFCM) and swarm intelligence optimization methods for medical image segmentation. The swarm intelligence algorithm that we have considered in this work is the Artificial Bee Colony (ABC) optimization. Findings: The algorithm is applied to brain MRI image segmentation and compared with other existing algorithm and the validation of the algorithm is evaluated by cluster a validity function which is an indication of how good a clustering result is. The results show that the combined algorithm i.e. ABCSFCM has better performance and improve the cluster validity functions as compared to SFCM. Applications/Improvements: The result is quite promising and although the proposed algorithm is tested on brain MRI image it can be extended to other problems of interest. The other variants of FCM and other natured inspired optimization are worth investigating for further improvements.

Keywords: ABC, ABCSFCM, MRI, SFCM, Segmentation