SEMINAR accuracy of the methods is not high

 

 

 

 

SEMINAR REPORT ON

 

 

 

Malaria Parasite Detection using different Machine Learning Classifier

 

 

 

 

SUBMITTED TO

Department of Computer Science & Engineering

 

 

by

 

 

Avinash Kashyap 170913007

 M.Tech -2nd Sem CSE

 

 

 

 

 

 

Under the Guidance of

 

Dr. Ashalatha Nayak

Head of Department

Computer Science and Engineering

Manipal Institute of Technology

2017-2018

 

 

 

 

 

 

 

 

 

 

 

 

 

CHAPTER 1: INTRODUCTION

 

 

Malaria is a deadly disease caused by mosquito bites is very common in the tropical and sub-tropical countries because hot, humid environments are most amenable to mosquito growth and survival. Malaria is caused by the bites of infected female Anopheles mosquitoes. Statistics shows that there were around 212 million malaria cases and roughly 429000 malaria deaths according to a report by WHO (World Health Organization). In India the stats are 0.7-1.6 million confirmed cases and 400-1000 deaths annually. Countries are facing a lot of financial loss because of the disease. The mosquito bite introduces the parasites from the mosquito’s saliva into a person’s blood. There are 5 different parasite species which are responsible for causing malaria in human body. Of the 5 parasites only 2 are of greatest threat to human body namely P. falciparum and P. vivax which are responsible for malaria related human deaths. Symptoms of malaria usually appear after 10-15 days of infective mosquito bite. Symptoms are fever, headache and chills may be mild and difficult to recognize as malaria. It is very important to Identify the presence of malaria in patient to cure them. The methods used for detecting malaria over the years is to collect blood samples of patient and study them under the microscope and various other kits have been developed for detection but they have some limitations. The accuracy of the methods is not high which may cause risk at the patient health.

 

In this work automated detection of malaria parasite is done using image processing techniques and Machine learning algorithms. Different Machine learning techniques for classification are used which shows the accuracy rate of 99.8% in the detection of malaria parasite in blood smear images.

 

 

 

 

 

 

 

 

 

 

 

CHAPTER 2: BACKGROUND

 

Malaria caused by the bites of female Anopheles mosquito and other causes can be an organ transplant, a transfusion or use of shared needles and syringes. The symptoms of malaria such as high fever, headache, chills are visible after 10 days of infective bite. To tackle this deadly disease WHO (World Health Organization) has recommended different techniques to control, prevent and eradicate the disease. The early detection of malaria is very difficult because of the symptoms showing up after 10 days, it is required to get a method which is fast and accurate in the early stages of malaria in past years different methods such as rapid diagnostic test (RDT) detection method and polymerase chain reaction method (PCR)

 

 

 

 

 

“Malaria in India,” Malaria Site, 09-Jan-2018. Online. Available: https://www.malariasite.com/malaria-india/. Accessed: 22-Jan-2018.

 

“Fact sheet about Malaria,” World Health Organization. Online. Available: http://www.who.int/mediacentre/factsheets/fs094/en/. Accessed: 22-Jan-2018.