Leukemia’s are cancers of the hematopoietic stem cells that cause the bone marrow. Acute and chronic are the two broad categories of leukemia. Acute leukemia is classified into two subtypes namely, Acute Lymphoblastic Leukemia (ALL) and Acute Myeloid Leukemia (AML). ALL is a childhood cancer that primarily affects children from 5 to 14 years old. If leukemia is predicted and identified in its early stage, appropriate treatment can be given to the patient. In this way, the disease can be cured, and patient’s recovery can be expedited and the lives of many leukemia patients can be saved. Leukemia cells are complex. Diagnosing leukemia disease through medical images is a difficult task. The primary goal of this research work is to extract the significant features from leukemia microscopic ALL images. The Fuzzy based Mathematical Clustering (FMC) algorithm is proposed in this work for feature extraction. From the experimental analysis, it is found that a proposed FMC has properly extracted GLCM, shape, and color features from leukemia images with a higher performance ratio than existing Gray Level Difference Method (GLDM), Gray Level Run Length Matrix (GLRLM), and Mean-shift feature extraction algorithms.

Medical Images, Leukemia, Acute Lymphoblastic Leukemia (ALL), Feature Extraction and GLCM.