Automated white blood cells counting system for acute leukemia based on blood images

Leukemia is a blood cancer disease that contributes to the increment of mortality rate in Malaysia each year. Acute leukemia can be divided into two types which are acute myelogenous leukemia (AML) and acute lymphocytic leukemia (ALL). The production and development of acute leukemia cells happen rapidly and uncontrollable. Identification of technique for acute leukemia could be done fast and effective in the early stage, the proper treatment could be delivered. Hematologists or technologists will screen for leukemia using microscopic blood image. Screening and diagnosis with computer aided system becomes more popular choice amongst medical image processing researchers as the existing equipments which use other method than blood image analysis are too expensive to own. This paper presents automated counting procedures, including a series of image processing techniques such as segmentation, noise elimination and extraction to count the region of interest, in this case the white blood cells (WBC). The performance of the system is tested by comparing the results between manual counting and the total cells that the system detects. The proposed system is able to yield an average of 99.09% based on sensitivity