A Review of Various Malaria Detection Techniques using Blood Smears
The traditional method of malaria detection requires manual counting of parasitized and uninfected cells. These methods depends heavily on human expertise, requires intensive labor, time consuming, less sensitive and more prone to errors. Such a technique could lead to poor results when applied to a large scale analysis due to human errors. In 2017, an estimated 219 million cases of malaria were detected which caused approximately 435000 deaths, mostly in developing countries. Thus the importance to develop new tools that facilitate easy diagnosis of malaria for areas with limited access to healthcare services cannot be overstated. State-of-the-art image-analysis based computer aided diagnosis (CADx) methods use machine learning (ML) techniques on microscopic images of the smears using hand-engineered features to analyze morphological, textural, and positional variations of the region of interest (ROI). In contrast, Convolutional Neural Networks (CNN), a class of deep learning (DL) models promise highly scalable and superior results with end-to-end feature extraction and classification. Automated malaria screening using DL techniques could, therefore, serve as an effective diagnostic aid. In this paper, we compare the state of the art algorithms for malaria detection and their performance toward classifying parasitized and uninfected cells to aid in improved disease screening.
Keywords - Malaria, Computer Aided Diagnosis (CADX), Convolution Neural Network (CONVNET/CNN), Machine Learning(ML), Deep Learning (DL)