Abstract—The biggest cancer disease invading children based on the health ministry 2015 is blood cancer or leukemia. One of the type leukemias is acute leukemia which consists of acute lymphoblastic leukemia (ALL) and acute myelogenous leukemia (AML). Acute leukemia can be diagnosed according to the calculation of a complete blood in the bone marrow, but the calculation process still has several problems, such as when leukemia blood cells are manually counted by microscope, it needs more power, takes too much time, and costs very expensive. This disease can be identified and classified by combining neural network and imaging processing techniques. Learning Vector Quantization (LVQ.1) is used as the neural network approach by extracting leukemia cells of ALL and AML. The image extraction used in this study is to use the color extraction of Hue saturation value color space and the texture feature of Gray level co-occurrence matrix. The experimental results show that the highest accuracy achieved by the proposed algorithm in identifying ALL is about 93.33% trained with 80% training data and tested with 20% testing data. On average, the proposed work yields about 70.31% accuracy to identify both blood cell types. In this sense, the proposed algorithm can classify ALL and AML well.
Index Terms—Acute leukemia classification, learning vector quantization, blood cell imaging extraction, digital image processing.
The authors are with the Department of Electrical Engineering Ming Chi University of Technology, Taiwan (e-mail: M07128021@o365.mcut.edu.tw, joannechen@mail.mcut.edu.tw, pooh61215@gmail.com, fadhilah.syafria@uin-suska.ac.id).
[PDF]
Cite: Faisal Asadi, Chiung-An Chen, Ting-Wei Liu, and Fadhilah Syafria, " Acute Leukemia (ALL and AML) Classification Using Learning Vector Quantization (LVQ.1) With Blood Cell Imagery Extraction," International Journal of Modeling and Optimization vol. 9, no. 3, pp. 171-176, 2019.
