Classification of acute leukemia using medical-knowledge-based morphology and CD marker
Journal article
Authors/Editors
Strategic Research Themes
No matching items found.
Publication Details
Author list: Laosai J., Chamnongthai K.
Publisher: Elsevier Ltd
Publication year: 2018
Volume number: 44
Start page: 127
End page: 137
Number of pages: 11
ISSN: 1746-8094
eISSN: 1746-8094
Languages: English-Great Britain (EN-GB)
View in Web of Science | View on publisher site | View citing articles in Web of Science
Abstract
Classification of subtypes in acute lymphocytic leukemia (ALL) and acute myelogenous leukemia (AML) is a vital pre-process for selecting an appropriate treatment for acute-leukemia patients, and urgently requires an automatic expert system to assist medical experts. To develop the automatic system, a classification method of acute leukemia subtypes is required. Currently, the cluster of differentiation (CD) marker, proven by medical scientists as important genetic information, is clinically used to classify acute-leukemia subtypes by comparison with classification results using morphological features. In the medical field, blood cells are first classified into ALL, AML, and healthy groups by perceptron features such as number of nuclei, cytoplasmic ratio, and nucleus size, and the classified ALL and AML groups are then classified into subtypes such as L1, L2, M1, M2, and so on by using their nucleus features. We therefore propose a method of morphological cell-subtype classification based on the coarse-to-fine concept following current medical knowledge. This means ALL, AML, and healthy cell groups are first separated in the coarse step, and the cells already classified into ALL and AML groups are then categorized into their subtypes in the fine step. These subtypes, which represent morphological classification results, are finally used as candidates to confirm cell-subtypes with CD markers in the decision-making process. In performance evaluation of the proposed method, experiments with 200 and 300 acute-leukemia samples for training and testing respectively were performed, and the results indicate 99.67% accuracy, a 4.94% improvement compared with the conventional method. ฉ 2018 Elsevier Ltd
Keywords
Acute leukemia, Acute lymphocytic leukemia (ALL), Acute myelogenous leukemia (AML), Multiparameter flow cytometry
Contact Information