Image Processing for Automated Leukemia Detection and Stage Prediction

Leukemia, which is a common disease nowadays, develops when the bone marrow produces an abundance of dysfunctional white blood cells. The hematologists utilize microscopic examination of human blood necessitates with the help of different processes like segmentation, classification, and grouping. The Haematologists visually examine the microscopic photos, which is a laborious and time-consuming operation. Computerized image processing technology is an essential tool in this application and can get beyond related visual examination limitations. In this communication, we proposed a microscopic imaging-based Leukemia detection algorithm. Early and prompt detection of Leukemia is extremely helpful in determining the best course of treatment. White blood cells are initially segmented using statistical characteristics like mean and standard deviation, which separates them from other blood components like erythrocytes and platelets. Geometrical characteristics like the area and periphery of the white circulation cell nucleus are examined for the diagnosis and prognosis process. A significant number of photographs have been successfully processed using the suggested method, with results that are encouraging for images of various qualities. With the help of the widely used simulation software MATLAB, we have introduced a novel method that results in the prediction of leukemia considering other factors like the percentage of myeloid cells, homogeneity, energy, correlation, contrast, etc.