Enhanced k-Means Clustering Algorithm for Malaria Image Segmentation

Abstract

Clustering is one of the most commonly used digital image segmentation technique in multifarious fields including medical image segmentation. In essence, this study proposes clustering algorithm to acquire good segmented images of Plasmodium Vivax malaria parasite species via unsupervised pixel segmentation. In this study, enhanced k-means (EKM) clustering algorithm which is an enhanced version of the conventional k-means (KM) clustering algorithm has been proposed for malaria slide image segmentation. In the proposed EKM clustering algorithm, the concept of variance and a new version of transferring process for clustered members are used to assist the assignation of data to the proper centre during the process of clustering, so that good segmented image can be generated. The satisfactory sensitivity together with the high specificity and accuracy values obtained from an average of 100 malaria images, indicates that the EKM algorithm has provided good segmentation performances as compared to k-means, fuzzy c-means and moving k-means clustering algorithms. Good segmented malaria parasite and clean segmented malaria image has been acquired using the proposed clustering algorithm. Hence, the proposed EKM clustering can be considered as an image segmentation tool for segmenting the malaria images.