AUTOMATED INTEGRATED CLUSTERING ALGORITHM FOR MAMMOGRAPHIC MASS SEGMENTATION

Abstract

Segmentation plays an important role in mammographic image processing by facilitating the delineation of regions of
interest. An automated Histon based integrated clustering algorithm is presented in this paper for the detection masses in
mammographic images by integrating K-means clustering algorithm with Fuzzy C-means algorithm. Initially Histon of the input
image was calculated and given as initial centroid for K-means clustering algorithm and Fuzzy C-means algorithm was applied
to segment the mass. The performance of proposed algorithm was evaluated using area overlap measure. The morphological
features are extracted from the segmented mass and 85% classification accuracy was obtained using SVM classifier.