AUTOMATIC DETECTION AND CLASSIFICATION OF MALARIA (PLASMODIUM FALCIPARUM) PARASITE GROWTH STAGES USING SHAPE ATTRIBUTES

Abstract

Malaria is a leading cause of deaths globally. Rapid and accurate diagnosis of the disease is a key to its effective treatment and management. Identification of plasmodium parasites life stages and species forms part of the diagnosis. In this study, a technique for detecting and identifying the P. Falciparum parasite life stages of microscopic images of thin blood smears stained with Giemsa was developed. The technique entailed designing and training Artificial Neural Network (ANN) classifiers to perform the classification of infected erythrocytes into their respective stages. Six morphological features were selected for ANN as inputs. These are Surface roundness, perimeter, maximum diameter, minimum diameter, area and equivalent diameter. Hence, the neuron numbers of the input layer were six. The output neurons considered in this study were three that corresponded to the three predefined growth stages of p. falciparum in malarial RBC considered in this study. The network with 20 hidden layers was trained by 80% (27 images) of the total data set. Then, the performance of the trained network was validated by 10% (3 images) and tested by using 10% (3 images) of the total 33 data set. The result of ANN classifier confusion matrix on the morphological features showed that from the total test examples of 33 images, 32 images (97.0%) were correctly classified and 1(3.0%) was misclassified