Detection of leukemia and its types using image processing and machine learning

Leukemia (blood cancer) begins in the bone marrow and causes the formation of a large number of abnormal cells. The most common types of leukemia known are Acute lymphoblastic leukemia (ALL), Acute myeloid leukemia (AML), Chronic lymphocytic leukemia (CLL) and Chronic myeloid leukemia (CML). This thesis makes an effort to devise a methodology for the detection of Leukemia using image processing techniques, thus automating the detection process. The dataset used comprises of 220 blood smear images of leukemic and non leukemic patients. The Image segmentation algorithms that have been used are k means clustering algorithm, Marker controlled watershed algorithm and HSV color based segmentation algorithm. The morphological components of normal and Leukemic lymphocytes differ significantly; hence various features have been extracted from the segmented lymphocyte images. The leukemia is further classified into its types and sub types by making use of the SVM classifier, which is a Machine Learning classifier. This thesis aims at detecting leukemia and determine whether it is AML, CML, CLL or ALL; thus taking the classification process one step further in the field of research, because most of the previous methods have been limited to the detection of leukemia or classifying it into one or two subtypes.