Detection of Malaria by Using a CNN Model

Malaria is a dangerous and potentially fatal disease caused by Plasmodium parasites that are transferred to humans through the bites of infected Anopheles mosquitos. The most common and accepted technique of diagnosing malaria is for a competent technician to visually analyze a blood smear under a microscope for parasitized red blood cells. The individual conducting the inspection’s experience and knowledge is used to make a diagnosis. It’s also time-consuming and, in rare situations, prone to human mistakes. The research work has been processed into two stages: Stage one deals with dataset summarization, whereas Stage two deals with the proposed algorithm. We present a completely automated CNN-based model for identifying malaria in blood smear micrographs in this study. Our proposed algorithm has a ninety-six percent accuracy rate in detecting malarial parasites from microscopic images, which can be improved further when trained and tested on larger datasets.