Open AccessAmalgamation of Clustering and Meta-heuristic Optimization Techniques for Automated MR Brain Analysis
Author(s) -
Senthilkumar Natarajan,
Vishnuvarthanan Govindaraj,
Kannapiran Balasubramanian,
Pallikonda Rajasekaran Murugan,
Arunprasath Thiyagarajan,
Anitha Narayanan,
J. Deny
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.b1159.1292s219
Subject(s) - computer science , cluster analysis , peak signal to noise ratio , artificial intelligence , heuristic , pattern recognition (psychology) , computation , noise (video) , image processing , image (mathematics) , data mining , algorithm
Interest in computer-assisted image analysis in increasing among the radiologist as it provides them the additional information to take decision and also for better disease diagnosis. Traditionally, MR image is manually examined by medical practitioner through naked eye for the detection and diagnosis of tumor location, size, and intensity; these are difficult and not sufficient for accurate analysis and treatment. For this purpose, there is need for additional automated analysis system for accurate detection of normal and abnormal tumor region. This paper introduces the new semi-automated image processing method to identify the brain tumor region in Magnetic Resonance Image (MRI) using c means clustering technique along with meta-heuristic optimization, based on Jaya optimization algorithm. The resultant performance of the proposed algorithm (FCM +JA) is examined with the help of key analyzing parameters, MSE-Mean Square Error, PSNR-Peak Signal to Noise Ratio, DOI-Dice Overlap Index and CPU memory utilization. The experimental results of this method show better and enhanced tumor region display in reduced computation time.