A Novel Method for Automatic Detection of Malaria Parasite Stage in Microscopic Blood Image

Abstract

Malaria is caused due to the mosquito bite hence the parasite enter into blood through the saliva of the mosquito. The malaria parasite directly infects the red blood cells, therefore to design an automatic detection system, the red blood cells should be segmented from the artifacts and background in a microscopic image. Here in this paper watershed transform is used with the distance transform which separates even the overlapped red blood cells more efficiently, which are useful for the diagnosis of parasite and for the parasitemia too. The result shows improvement in diagnostic accuracy of detection of the parasite in Red Blood Cells and also describing the life cycle stage of the parasite. The accuracy, sensitivity and specificity achieved were as 97.7%, 97.4% and 97.7% respectively.

Improving Classification Accuracy Using Hybrid Machine Learning Algorithms on Malaria Dataset

“…Feature extraction is a method of identifying and classifying some of the specific points on an image [10].…”

Machine Learning and Its Applications for Protozoal Pathogens and Protozoal Infectious Diseases

“…An SVM method based on the watershed threshold algorithm for the detection of the lifecycle stage in microscopic blood images was proposed in Charpe et al. (2015) and shows 97.7% accuracy, 97.4% sensitivity, and 97.7% specificity.…”