Malaria Cell Detection Using Deep Learning Architectures

Malaria is a common and life-threatening disease in our country, with high-risk areas in several villages and hill tracts. Current detection methods are time-consuming and inaccessible. Our system analyzes digital images of blood cells to identify signs of malaria infection. By utilizing image processing techniques and the Deep Learning method, the system accurately detects malaria-infected cells. The project involves collecting a labeled dataset of infected and uninfected blood cell images, applying preprocessing techniques, extracting relevant features, training the machine learning model, and evaluating its performance. Early and automated malaria diagnosis can improve healthcare outcomes. In our machine learning model, we get the best accuracy from the EfficientNetV2L model, which is 97%.