Infection detection: Bridging the gap from pixels to malaria diagnosis

Abstract

Malaria, a serious disease that affects millions of people worldwide, continues to pose a significant global health challenge. remains a major global health challenge. In this context, we present a new method for detecting malaria in images of blood cells using long-term memory (LSTM) networks. Our primary goal is to develop an automated diagnostic system that can assist specialists in accurately identifying infected and uninfected blood cells. We have successfully captured complex sequence relationships within these images, which leads to high-resolution detection results. The achieved performance of our model reached an impressive accuracy of 98%, confirming its effectiveness in detecting malaria.