Plasmodium falciparum Detection in Cell Images Using Convolutional Neural Network

Abstract

Malaria is one of the major burdens to global health, which results in countless deaths every year. It is caused by a group of Plasmodium parasites which spreads through the bite of the female anopheles mosquito. The infected mosquito first bites the host, and the parasite enters the bloodstream which proceeds to go to the liver. From the liver, the parasites grow and multiply in the red blood cells. The infected red blood cells eventually burst and release more parasites. The diagnosis of malaria is done by doing a blood test where the count of the parasite is found by examining thin blood smears under a microscope. This method is also used for testing drug resistance, measuring drug effectiveness, and classifying disease severity. Microscopic diagnostic methods are cumbersome and thus require a lot of skill and experience to execute. In this study, we propose the use of a deep convolutional neural network to detect the presence of red blood cell images. This would assist in automating the detection of malaria from red blood cell images and aid in early diagnosis.