Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/269051
Title: Automated Malaria Detection Using Machine Learning Techniques
Researcher: CHAYA D. JAGTAP
Guide(s): USHA RANI NELAKUDITI
Keywords: 
University: Vignans Foundation for Science Technology and Research
Completed Date: 2019
Abstract: In order to diagnose malaria, a test that has traditionally been conducted which mainly entails preparation of a blood smear on a glass slide, staining the blood and examining the blood through the use of a microscope to observe parasite genus plasmodium. Although there are several other kinds of diagnostic test solutions which can be adopted, there are numerous shortcomings which are always observed when the microscopic analysis is carried out. Presently, the treatments are conducted based on symptoms, and upon the occurrence of false negatives, it might be fatal and may result in the creation of different kinds of implications. There have been a number of deaths which have been associated with malaria, and as a result, there is a dire need to ensure that there is early detection of malarial infection among the people. The present study mainly provides a review of the current contributions regarding computer-aided strategies, as well as microscopic image processing strategies for the detection of malaria. newlineThe present research work contributions of this thesis endeavored for proposing a new selection of feature and classification strategies to optimize the computer-aided malaria detection process. The hierarchy of the research contributions discussed in this doctoral thesis is listed below. newlineThe initial contribution carried a quantitative and qualitative analysis of the contemporary literature related to computer-aided malaria detection strategies, which is built on the concepts of machine learning and data mining. The review explored the limits and constraints of the existing methods, and scope to escalate the research further to overcome these limits. newlineFurther, contribution endeavored to denote a novel feature selection strategy that guarantees the variety of optimum features to identify the difference between infected with malaria and normal erythrocytes. Subsequently, the doctoral thesis explores a classifier that built on cuckoo search concept, which is further used to devise a scale to distinguish the infected and normal erythrocytes. newlineThe further contribution of the doctoral thesis is an ensemble classifier, which is an extension to the earlier contribution. The objective of this ensemble classifier is to overcome the constraint of the dimensionality of the values projected for the features, which is the high newline newlinemisclassification rate due to diversity in values projected for optimal features. newline newline
Pagination: 171
URI: http://hdl.handle.net/10603/269051
Appears in Departments:Department of Electronics and Communication Engineering

Files in This Item:
File Description SizeFormat 
10_chapter-7.pdfAttached File486.39 kBAdobe PDFView/Open
11_references.pdf662.06 kBAdobe PDFView/Open
12_biblography.pdf1.01 MBAdobe PDFView/Open
1_title.pdf495.02 kBAdobe PDFView/Open
2_certificate.pdf441.04 kBAdobe PDFView/Open
3_prelimenary pages.pdf842.06 kBAdobe PDFView/Open
4_chapter-1.pdf1.41 MBAdobe PDFView/Open
5_chapter-2.pdf1.06 MBAdobe PDFView/Open
6_chapter-3.pdf2.26 MBAdobe PDFView/Open
7_chapter-4.pdf1.65 MBAdobe PDFView/Open
8_chapter-5.pdf2.13 MBAdobe PDFView/Open
9_chapter-6.pdf1.18 MBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetric Badge: