Automated leukemia detection using hausdorff dimension in blood microscopic images

Abstract

A cute lymphocytic leukemia (ALL) is a malignant disease characterized by the accumulation of lymphoblast in the bone marrow. An improved scheme for ALL detection in blood microscopic images is presented here. In this study features i.e. hausdorff dimension and contour signature are employed to classify a lymphocytic cell in the blood image into normal lymphocyte or lymphoblast (blasts). In addition shape and texture features are also extracted for better classification. Initial segmentation is done using K-means clustering which segregates leukocytes or white blood cells (WBC) from other blood components i.e. erythrocytes and platelets. The results of K-means are used for evaluating individual cell shape, texture and other features for final detection of leukemia. Fractal features i.e. hausdorff dimension is implemented for measuring perimeter roughness and hence classifying a lymphocytic cell nucleus. A total of 108 blood smear images were considered for feature extraction and final performance evaluation is validated with the results of a hematologist.