An Enhanced Fuzzy C-Means Clustering (ECFMC) Algorithm for Spot Segmentation

In this paper, we developed an efficient approach for the automated analysis of microarray images. The proposed approach consists of three steps: pre-processing, gridding and segmentation. Initially, the microarray images are preprocessed by hill climbing algorithm and gridding is applied to accurately identify the location of each spot while extracting spot intensities from the microarray images. In the segmentation stage, we developed an efficient Enhanced Fuzzy C-means (EFCMC) algorithm. The advantages of the proposed clustering algorithm compared with the conventional clustering techniques in segmentation of microarray image are given as: (1) it effectively detects the absent spot since the inclusion of neighborhood pixel information. (2) It is quite independent of the various noises (3) the detection of accurate spot can be done. By taking these advantages, the proposed approach effectively segmented the microarray images with preserving the image details. The proposed method is tested and evaluated using the Microarray image Database and the results confirmed the efficiency of the proposed approach.