An efficient CNN-based Automated Leukemia diagnosis Using microscopic blood smear images and Subtypes Classification

Abstract

Leukemia is a form of blood cancer that damages the cells in the blood and bone marrow of the human body. It produces cancerous blood cells that disturb the human's immune system and significantly affect bone marrow's production ability to effectively create different varieties of blood cells like red blood cells (RBCs) and white blood cells (WBC), and platelets. Different kinds of manual methods have been used, but all these techniques are slow, labour-intensive, inaccurate, and need a lot of human experience and dedication. To deal with such manual methods, different researchers used different machine learning algorithms to classify the cells into normal and blast cells. However, still, the problem is complex blood characteristics. In this paper, we have proposed a robust diagnosis system to classify leukemia and its subtypes. Acute lymphocytic leukemia (ALL) is classified into subtypes based on FAB classification, such as L1, L2 and L3 types with better performance. Our model outperformed as compared to other state-of-the-art approaches.