Malaria Detection System Using Convolutional Neural Network Algorithm

Malaria Detection System Using Convolutional Neural Network Algorithm

Kanika Gautam, Sunil Kumar Jangir, Manish Kumar, Jay Sharma
Copyright: © 2020 |Pages: 12
DOI: 10.4018/978-1-7998-3095-5.ch010
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Malaria is a disease caused when a female Anopheles mosquito bites. There are over 200 million cases recorded per year with more than 400,000 deaths. Current methods of diagnosis are effective; however, they work on technologies that do not produce higher accuracy results. Henceforth, to improve the prediction rate of the disease, modern technologies need to be performed for obtain accurate results. Deep learning algorithms are developed to detect, learn, and determine the containing parasites from the red blood smears. This chapter shows the implementation of a deep learning algorithm to identify the malaria parasites with higher accuracy.
Chapter Preview
Top

Literature Review

The work done in the past by many authors in the field of detecting malaria parasites in the red blood smears by using modern machine learning and deep learning methodologies have given better results as compared to the traditional methods being used. In 2015, Hanung Adi Nugroho et.al, carried out work in the domain of feature extraction and classification for malaria detection which gave a result of 87.8% accuracy. Similarly, Yuhang Dong et.al, in 2017 analyzed an automated system for malaria detection which results in an accuracy of 95% (Shen et al. 2017) (Dong, Jiang, Shen, David Pan, et al. 2017). The Table 1 below shows a comparative study of the work done by different authors in different domains to obtain better results.

Table 1.
Comparative study of different methodologies used by different authors
S.No.Author and YearApproach usedResults and Discussions
1.(Poostchi et al. 2018)Deep learning, Image Acquisition, Feature ExtractionEnables larger test suits on the patient level.
Automated microscopy turns out to be cheap, simple and reliable.
2.(Dong, Jiang, Shen, and Pan 2017)Deep learning, CNN, SVM, LeNet, AlexNet and GoogLeNet.Results came out with 95% accuracy. Features of the datasets were learned automatically
3.(Rahman et al. 2019)Data Splitting, Stain Normalisation, Min-Max Normalisation, Data AugmentationThe results were less accurate.
The layout of the original image changed.
4.(Rahman et al. 2019)K means clustering algorithm, classification, KNN classifier.Separation of infected cells.
Malaria parasites detected and classified.
5.(Nugroho, Akbar, and Murhandarwati 2016)Histogram, Feature Extraction, Classification, BackpropagationThe proposed method results in the accuracy of 87.8%, sensitivity of 81.7%, and specificity of 90.8% for detecting malaria parasites in red blood smears.
6.(Khan et al. 2017)Color translation, K-Means clustering.The feature set obtained b*-color channel from the CIE L*a*b* color space provides better clustering results.
7.(Pattanaik, Swarnkar, and Sheet 2017)Object Detection, Kalman Filter; Kernal Based Detection, Computer vision.The newly acquired representation enhances the detection performance of infected malaria parasites n thin blood smear images.

Complete Chapter List

Search this Book:
Reset