Generation of RBC, WBC Subtype, and Platelet Count Report Using YOLO
Vaishnavi Abhyankar, Rashmi Kene · International Journal for Research in Applied Science and Engineering Technology · 2023
Abstract: Artificial intelligence introduced a way to combine machines’ computing ability with human intelligence. Machine learning is a sub-branch of AI that consists of different algorithms to implement concepts of AI in practical terms. But when a machine has to process numerous amounts of data, deep learning algorithms come into the picture. It is observed that when a computing system has to deal with image data then neural network algorithms give efficient ways to process them and draw unique patterns from them. Object detection is the task of identifying required objects from an image. This type of technology plays a crucial role in medical image processing. Some algorithms can efficiently identify and classify objects from an image. It is observed that Fast R-CNN and Faster R-CNN, mask R-CNN, have given pretty good accuracy while performing such tasks. But when time is a concern for a system, such methods put an obstacle of training time and architectural complexity. The You Only Look Once (YOLO) object detection method has introduced a new way of processing images in a single pass. This algorithm is famous because of its speed and correctness. There are numerous models of YOLO developed now which include YOLOv1 to YOLOV8. This paper gives the performance of the latest version of YOLO on the blood cell dataset.