The Study Of Brain Mri Segmentation Using Fuzzy Clustering Technology

Objective Improve and optimize the application of fuzzy c-means clustering algorithm on brain magnetic resonance image (MRI) segmentation by analyzing the parameters in the algorithm.Methods 50 brain magnetic resonance images were processed using the fuzzy c-means (FCM) image processing procedures. Four parameters were used for comparison:number of fuzzy clustering c, weighted index m, thresholdε, and number of recursions N. Only one parameter varied within a certain range each time. The range of fuzzy clustering number c was (3,4,5,6,7,8); the range of weighted index m was (1.1,1.2,1.3,2.5,2.6); the range of the threshold s was (0.1,0.01,0.001,0.0001,0.00001,0.000001); the range of number of recursions N was (5,10,20,30,40,50,60,70,75,80,90,100,200, 300,400,500,600,700,800,900,1000).Results and Conclusion The optimal brain MRI segmentation was achieved when the fuzzy clustering number c was 6, the weighted index m was 2.1, the threshold valueεwas 0.00001 and the number of recursions was 75.