Enhanced $k$-Means Clustering Algorithm for Detection of Human Intestinal Parasites

Helminthiases disease has been diagnosed as one of the most common health problems in children and adults that is caused by human intestinal parasites named helminth. An immediate diagnosis system is required for the identification and classification of helminth to avoid the development of parasites in human intestine. In response to the needs for a rapid and precise identification of helminth, this research proposes analyzing the segmentation performance analysis of unsupervised colour image segmentation of helminth parasites through enhanced
$k$
-means clustering (EKM) algorithm on the modified global contrast stretching (MGCS) and the modified linear contrast stretching (MLCS) enhancement techniques. By segmenting the helminth parasites using 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 obtained. A total of 100 helminth parasites images have been analyzed with the approach comprising of 50 images from each species of the Ascaris Lumbricoides Ova (ALO) and the Trichuris Trichiura Ova (TTO). Overall, the average results generated by the enhanced
$k$
-means clustering algorithm on MGCS is higher than MLCS. The proposed technique achieved 99.88% accuracy, 95.68% sensitivity alongside 99.95% specificity. Ultimately, the result demonstrates that the proposed technique is suitable for image segmentation of the helminth parasites images.