Detection of Acute Lymphocytic Leukemia using Blast Cell Morphological Features

Precise detection of leukemia/ blood cancer is one of the major issues in the medical field so that the victims can be treated effectively. ‘Leukemia’ affects the lymphatic system of our body. Potent infection fighter cells that are white blood cells are involved in causing leukemia. If leukemia causes abnormalities in lymphocytes, it is known as lymphoblastic/ lymphocytic leukemia. If leukemia causes abnormalities in myelocytes, it is known as myeloblastic/ myelocytic leukemia. Leukemia can also be further classified as ‘Acute’ and ‘Chronic’. Acute leukemia is more decisive than chronic leukemia since it causes immediate death. In this paper, the detection and classification of Acute Lymphocytic Leukemia (ALL) and their subtypes are focused. Supervised classifiers such as SVM and KNN classifiers are evaluated for the detection and classification of cells. From the results obtained, it is determined that the KNN classifier is more suitable for classification when noise-free data is used.