Automatic Detection of Malaria Using Convolutional Neural Network

Abstract

It's an infectious parasitic disease spread by mosquitoes that's caused by a parasite called (Plasmodium). This parasite infects and destroys red blood cells in the human body. Fever, anemia, spleen enlargement, headache, chills, and exhaustion are among the most common symptoms. Blood smears from individuals are still planned under the microscope by competent professionals and technicians for red blood cells contaminated with parasites, which is the classic approach for diagnosing malaria. The wait for a diagnosis is long. Deep learning algorithms have been used to infected human blood signals as a result, however the practical performance has so far been insufficient. The architecture of the convolutional neural network (CNN) was used in this paper to diagnose malaria illness using deep learning. , which includes 43,828 pictures of infected and uninfected blood cells were used to train on various weighted layers to assess the presence or absence of malaria in red blood cells from microscopic images collected of blood smears of infected human persons. A total accuracy of 96.87 was attained when the CNN model was tested. The proposed deep learning approach for detecting malaria was found to be effective based on the experimental results on the clinical data set.

Keywords: malaria ,deep learning , machine learning ,CNN