Chronic lymphocytic leukemia cell segmentation from microscopic blood images using watershed algorithm and optimal thresholding

Abstract

Chronic lymphocytic leukemia (CLL) is the most common type of blood cancer in Canadian adults. CLL cells are abnormal lymphocytes, which tend to be slightly larger than normal resting lymphocytes and have a condensed appearance to their chromatin. There is a low number of related works on this disease. This paper presents a method to segment normal and CLL lymphocytes into two parts: nucleus, and cytoplasm using a watershed algorithm and optimal thresholding. The goal of this work is reducing the over and under segmentation error of the watershed algorithm by suppressing 1% of the local minima. We tested 140 microscopic lymphocyte images (normal and CLL), and the algorithm obtained 99.92% maximum accuracy for nucleus segmentation, and 99.85% maximum accuracy for cell segmentation. The cytoplasm can be extracted with a 99.63% maximum accuracy with simple mask subtraction. The code for the presented algorithm is shared on the MATLAB <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> file exchange website.