Akademik Veri Yönetim Sistemi
Journal of scientific reports-A (Online), sa.048, ss.42-54, 2022 (Hakemli Dergi)
Malaria is a disease that causes a parasite called plasmodium
to be transmitted to humans as a result of the bite of female anopheles’
mosquitoes. Malaria is detected by examining the blood sample taken from the
patient as a result of a microbiological examination under a microscope by
specialist physicians. Although microscopy is widely used, its efficiency is
low because it is time-consuming and depends on the interpretation of the
specialist physician. In recent years, deep learning methods used in the field
of computer vision increase the efficiency of specialist physicians by making a
significant contribution to the decision-making process in solving real-life
problems. In this study, ResNet architectures were preferred to quickly
classify the malaria parasite using deep learning methods. For the training and
testing of ResNet architectures, a dataset consisting of a total of 27558 red
blood cell images containing 13779 parasitized and 13779 uninfected were used.
Using this dataset, ResNet architectures were compared. As a result of the
comparison, the best success accuracy (94.09%) was obtained with the ResNet-50
v2 model.