An Efficient Malaria Parasite Detection Using Convolution Neural Networks

Abstract

Malaria disease is a life frightening blood disease which is instigated by female anopheles mosquito bite which is considered as an endemic in many parts of the world [12]. Manually microscopists usually examine the blood smears and the accuracy depends on the experience and the expertise in classifying the blood cells. This article focuses on detecting malaria from the microscopic images of the red blood cell by using deep learning models like Convolutional Neural Networks. The dataset has been taken from NIH Malaria Dataset which is available in National Library of Medicine [6]. It consists of both malaria parasite images and uninfected microscopic images. Authors implemented an application in which it uses Deep Convolutional Networks and performs feature extraction and classification from the red blood cells and classifies whether the blood cell is affected by malaria parasite or not.