Detection, Classification and Counting RBCs and WBCs Using Deep Learning

Blood is a highly complex and essential biological fluid in the human body that performs various vital functions. It comprises different components, including plasma, platelets, and various types of blood cells, each possessing unique properties and functions. Among these blood cells, red blood cells (RBCs) and white blood cells (WBCs) hold significant importance. In the medical world, detection, Identification, and counting of these RBCs and WBCs is important to evaluate a person's comprehensive health condition. Counting blood cells is one of the most common tests conducted at medical treatment facilities. Manually detecting and counting blood cell counts using a hemocytometer is time-consuming and tedious. The challenges here to be worked on and the goals to be achieved are related to Accurately detecting and counting two types of blood cells within a limited timeframe, this paper proposes a deep learning approach to this issue. The algorithm that detects and classifies objects “You Only Look Once” (YOLO) is being used in our approach. The output achieved from the work on medical images of blood cells show the detected cell's bounding boxes, classification results, and confidence levels, along with the count of each type of blood cell.