White Blood Cells Segmentation and Classification Using Swarm Optimization Algorithms and Multilayer Perceptron

Abstract

This study proposes a segmentation and classification system for early detection of blood disease; the proposed system consists of three phases. The first phase is segmenting white blood cells using multi-level thresholding optimized by the butterfly optimization algorithm to select the optimal threshold value to increase the accuracy. The second phase is extracting geometric and shape features of the segmented cells. The third phase is using the gray wolf optimizer to adopt the weights and biases of the multilayer perceptron to enhance the accuracy of classification between normal and leukemia cells, classify the normal cells to their five categories, and classify the leukemia to their four categories. The proposed system applies to different data sets (ALL-IDB2, LISC, and ASH-Image bank) and overcomes the segmentation and classification problems of microscopic images and shows an outstanding segmentation result, 98.02%; and the average classification accuracy between normal and leukemia cells is 98.58%, between white blood cell categories is 98.9%, and between leukemia types is 98.93%.