Abstract
Bioengineering is the art of applying engineering principles, techniques, and technologies to biology and medicine for general healthcare applications. Analyzing human biological samples such as blood, has become essential for physicians to diagnose and follow diseases evolution. Traditional blood cells counting techniques used in laboratories are time consuming and laborious. They can lead to inaccurate results due to the human intervention in this complicated process. In this paper, we propose an automated blood cells counting framework using convolutional neural network (CNN), instance segmentation, transfer learning, and mask R-CNN techniques. Red and white blood cells are identified, classified, and counted from microscopic blood smear images. The obtained results reveal highly detection rate of different blood cells. In addition, unlike other state-of-the-art techniques, our proposed method has the ability to identify overlapped and faded cells.