Tran V.-N.Ismail W.Hassan R.Yoshitaka A.2024-05-282024-05-28201697818900000002154482410.1109/WAC.2016.75830232-s2.0-84993978020https://www.scopus.com/inward/record.uri?eid=2-s2.0-84993978020&doi=10.1109%2fWAC.2016.7583023&partnerID=40&md5=fb1d26b0b4f2bd47bf438164609ff141https://oarep.usim.edu.my/handle/123456789/9446Leukemia is a cancer of white blood cells that affect the blood forming cells in the body. Acute Myeloid Leukemia (AML) is a form of leukemia and are caused by replacement of normal bone marrows with leukemic cells, which cause a drop in red blood cells, platelets, and normal white blood cells. Early classification of the subtype of AML cells is necessary for proper treatment management. We classify the subtype based on the features of AML cells, which include the nuclei and cytoplasm. In this paper, we developed an automate method for the nuclei and cytoplasm detection from the blood cells images that are captured as microscope images. In contrast to other methods that focus on identifying the nuclei, we proposed a method based on the color conversion, intensity threshold and gradient magnitude. Our method detected both the nuclei and the cytoplasm at the same time. We test our method on 301 images, which contain 643 AML cells. The accuracy of both nuclei and cytoplasm detection is over 82.9% (increase 17% when was compared with the existent method). � 2016 TSI Enterprise Inc (TSI Press).en-USAcute Myeloid Leukemia (AML)cytoplasmgradient magnitudenucleiAutomationBloodCytologyDiseasesAcute myeloid leukemiaAutomated methodscytoplasmGradient magnitudeIntensity thresholdnucleiTreatment managementWhite blood cellsCellsConference Paper2016-October7583023