Title: An efficient CAD system for ALL cell identification from microscopic blood images

Abstract

Computer-aided diagnosis (CAD) becomes a common tool for identifying diseases, especially various cancers, from medical images. Thus, digital image processing plays a significant role in this research area. This paper concerns with developing an efficient automatic system for the identification of acute lymphoblastic leukemia (ALL) cells. The proposed approach involves two steps. The first step focuses on segmenting the white blood cells (WBCs). In the second step, significant features such as shape, geometrical, statistical, and discrete cosine transform (DCT) are extracted from the segmented cells. Various classification techniques are applied to the extracted features to classify the segmented cells into normal and abnormal cells. The performance of the proposed approach has been evaluated via extensive experiments conducted on the well-known ALL-IDB dataset of microscopic images of blood. The experimental results demonstrate that the proposed approach realizes an accuracy rate 97.45% and outperforms other existing approaches.