Extraction of Nucleolus Candidate Zone in White Blood Cells of Peripheral Blood Smear Images Using Curvelet Transform. (15th May 2012)

• Record Type: Journal Article
• Authors: Soltanzadeh, Ramin; Rabbani, Hossein; Talebi, Ardeshir
• Other Names: Crum Bill Academic Editor.
• Is Part Of: Computational and mathematical methods in medicine. Volume 2012(2012)
• Publication Date: 2012-05-15
• Subjects: Medicine -- Computer simulation -- Periodicals; Medicine -- Mathematical models -- Periodicals; 610.11
• DOI: 10.1155/2012/574184
• Languages: English
• ISSNs: 1748-670X

Abstract

Abstract : The main part of each white blood cell (WBC) is its nucleus which contains chromosomes. Although white blood cells (WBCs) with giant nuclei are the main symptom of leukemia, they are not sufficient to prove this disease and other symptoms must be investigated. For example another important symptom of leukemia is the existence of nucleolus in nucleus. The nucleus contains chromatin and a structure called the nucleolus. Chromatin is DNA in its active form while nucleolus is composed of protein and RNA, which are usually inactive. In this paper, to diagnose this symptom and in order to discriminate between nucleoli and chromatins, we employ curvelet transform, which is a multiresolution transform for detecting 2D singularities in images. For this reason, at first nuclei are extracted by means of K -means method, then curvelet transform is applied on extracted nuclei and the coefficients are modified, and finally reconstructed image is used to extract the candidate locations of chromatins and nucleoli. This method is applied on 100 microscopic images and succeeds with specificity of 80.2% and sensitivity of 84.3% to detect the nucleolus candidate zone. After nucleolus candidate zone detection, new features that can be used to classify atypical and blast cells such as gradient of saturation channel are extracted.