This study evaluates the performance of quality measures for the algorithms Modified Expectation Maximization (MEM) and Particle Swarm Optimization (PSO) in segmentation of medical images. Medical images of different modalities, such as computed tomography, magnetic resonance, X-ray and ultrasonic are considered for this study. The quality measures like peak signal to noise ratio (PSNR), average difference (AD), structural content (SC), image fidelity (IF), normalized correlation coefficient (NK), mean structural similarity index (MSSIM) and universal quality index (UQI) are calculated for medical images using MEM and PSO method. Experimental results sound profound for Modified Expectation Maximization (MEM) with average of 3dB increase in PSNR values than the PSO. Also, Figure of Merit (FOM) a performance measure for edge detection is considered for choosing the best technique of edge detection for medical images. Finally, Trend factor is set using aggregated quality values and FOM for the better segmentation as well as edge detection.
Harikumar, R. and Vinoth Kumar, B.
"Performance Analysis of Medical Image Segmentation and Edge Detection using MEM and PSO Algorithms,"
Applied Mathematics & Information Sciences: Vol. 09
, Article 51.
Available at: https://dc.naturalspublishing.com/amis/vol09/iss6/51