Malaria remains as one of the major public health problems worldwide. About 228 million cases occurred in 2018 only, with Africa bearing about 93% of the cases. Asymptomatic population carrying the various forms of the parasite Plasmodium in endemic areas plays an important role in the spread of the disease. To tackle this battle, more sensitive and precise detection kits for malaria are crucial to better control the number of new malaria cases.
Computer-assisted algorithms have become a mainstay of biomedical applications to improve accuracy and reproducibility of repetitive tasks like manual segmentation and annotation. We propose a novel pipeline for red blood cell detection and counting in thin blood smear microscopy images, named RBCNet, using a dual deep learning architecture. RBCNet consists of a U-Net first stage for cell-cluster segmentation, followed by a second stage Faster R-CNN for detecting small cell objects within clusters, identified as connected components from the U-Net stage.