Automated Malaria Parasite Detection for Legal Blindness Accessibility using Hybrid Deep Learning Techniques
J. Senthil Kumar, S. Pandiarajan, J.Karthika Sorna Ilakkia · 2023
Mosquitoes carry malaria, a serious blood disease that can be fatal, contagious, and life-threatening. The most common and accepted method of diagnosing malaria involves looking at blood smears under a microscope to visually check for red blood cells that are infected with parasites. The results of this system depend on the examiner's knowledge and experience, are limited in scope, and take a lot of time to complete. In order to form an opinion, blood smears from malaria patients were subjected to automatic image recognition technologies based on image processing. However, the performance in practice has been underwhelming so far. This encourages us to develop the most straightforward, efficient, and rapid malaria diagnosis methods. Using image processing and prompt testing, our main goal is to develop a model that can categorize cells as infected or uninfected. This model will be able to distinguish between various cell types in thin blood smears on typical microscope slides. In order to find the parasite on the picture of the infected cell using deep learning techniques.