MRI Image Segmentation Using Active Contour and Fuzzy C-Means Algorithm

Abstract

Interpretation of MRI images is difficult due to inherent noise and inhomogeneity. Segmentation is considered as vitally important step in medical image analysis and classification. Several methods are employed for medical image segmentation such as clustering method, thresholding method, region growing etc. In this paper, attention has been focused on clustering method such as Fuzzy C-means clustering algorithm that has been widely used for medical image segmentation. This algorithm was combined then with Active Contour method. Active Contours have been widely used as attractive image segmentation methods because they produce sub regions with continuous boundaries. The algorithms have been implemented and tested on MRI images. The comparison is made with existing conventional Fuzzy C-means method. Experimental results show that the proposed hybrid method significantly improves the Peak Signal to Noise Ratio (PSNR) and Mean Squared Error (MSE) for medical image segmentation.