Clustering-Based Dual Deep Learning Architecture for Detecting Red Blood Cells in Malaria Diagnostic Smears

Using a dual deep learning architecture, we offer RBCNet, a new pipeline for red blood cell recognition and counting in thin blood smear microscopy images. RBCNet is made up of two stages: a U-Net first stage for cell-cluster or super-pixel segmentation, and a Faster R-CNN second stage for recognizing small cell objects within connected component clusters.