Automatic Malaria Parasite Detection and Classification using ANFIS

Abstract

Recent advancement in genomic technologies has opened a new realm for early detection of diseases that shows potential to overcome the drawbacks of manual detection technologies. Computer based malarial parasite analysis and classification has opened a new area for the early malaria detection that showed potential to overcome the drawbacks of manual strategies. This thesis presents a method for automatic classification of malarial infected cells. Blood cell segmentation and morphological analysis is a challenging due complexity of the blood cells. To improve the performance of malaria parasite segmentation and classification, we have used different set of features which are forward to the ANFIS classifier for malaria classification. the segmentation of clustered partially overlapping objects with a shape initially separated using marker controlled watershed segmentation accompanied with and overlapping cells concave point segmentation and contours are approximated using an ellipse. whereas ANFIS classifier for classification on different set of texture and shape features. This Study shows 96.33% and 96.31% recognition rates for both training and testing using ANFIS classifier.