Automatic Detection of White Blood Cells from Microscopic Images for Malignancy Classification of Acute Lymphoblastic Leukemia

Abstract

Leukemia is a group of cancers that typically begins in the bone marrow that usually affects both the children and the adults and remains a leading cause of fatality to death around the whole world. Acute Lymphoblastic Leukemia (ALL) is the most common type of leukemia. ALL diagnosis is performed by expert hematologists from blood microscopic images manually. However, human inspection is lengthy, tedious, and repetitive and fully depends on the operator's skill; thus its statistical reliability becomes limited. A computer based system can be used as a supportive tool for the hematologists in order to enhance and accelerate the morphological diagnosis process. The aim of this paper is to design an automated system for the detection and classification of acute lymphoblastic leukemia by analyzing morphological, textural and color features from the blood microscopic images using image processing. The proposed system is designed and now ready for the practical usage in medical having a white blood cell detection accuracy of 93% and a malignancy classification accuracy of 93.6% and the system is tested with 108 blood microscopic images of ALL-IDB. The proposed system is successfully implemented in MATLAB R2016a and it is named as CAD-SFA (Computer Aided Diagnosis-System for ALL).