A Novel Approach to Detect Chronic Leukemia using Shape based Feature Extraction and Identification with Digital Image Processing

Abstract

In this paper, some shape based features like area, perimeter, roundness, standard deviation etc. are used to recognize different types of white blood cells like monocyte, lymphocytes, eosinophil, basophil, neutrophils etc. Using image processing techniques, result can be obtained within 3-4 minute. To perform shape base features operation, contrast of RGB image has to be increased for better detection of white cells. After recognition of each and every cell, classification is performed to detect either it is CML (Chronic Myelogenous Leukemia) or CLL (chronic Lymphocytic leukemia). This algorithm is performed on 30 images. Out of 30, it is successful on 28 images. So it gives accuracy of 93.33%.

Detection of Acute Lymphocytic Leukemia using Statistical Features

“…The following nine shape descriptors were extracted from each blast cell and its nucleus [12]: Area: It is represented by the number of pixels in the ROI. This feature was used to quantify the size of the ROI.…”

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

“…Vaghela et al (Vaghela et al, 2015) compared between histogram equalization, K means, Watershedtransformandshapebasedfeaturesindetectingandcountingleukemiaaffectedcells. Theexperimentshowedthattheshapebasedfeaturesmethodgavethebestaccuracyof97.8%,while histogramequalization,KmeansandWatershedgaveaccuracyof73.7%,72%and72.2%respectively.…”