Leukaemia/ Blood Cancer Classification, Detection, And Evaluation

Abstract

Leukaemia is a form of blood cancer that begins in the bone marrow and causes the body to make abnormal blood cells. Acute leukaemia is a kind of blood cancer that is linked to bone marrow malfunction. Young people and children have a higher risk of getting it. Due to the fast cell growth, this kind of leukaemia produces, prompt treatment is essential. The four most frequent types of leukaemia are acute lymphoblastic leukaemia, acute myeloid leukaemia, chronic lymphocytic leukaemia, and chronic myeloid leukaemia (CML). Acute lymphoblastic leukaemia (ALL) and acute myeloid leukaemia (AML) are the two most common kinds of leukaemia, and they are also the leading causes of mortality from the disease (AML).The computerized approach may help doctors decide on the best course of action by classifying these two forms of leukaemia on pictures of blood slides. This research aims to demonstrate the design of a convolutional neural network (CNN) that can differentiate between blood slides such as those showing ALL, AML, and healthy blood (HBS). A total of 2,415 images from 16 datasets were used in the studies, which achieved an accuracy and precision of 97.18 and 97.23 percent, respectively. The suggested model's efficacy was measured against that of state-of-the-art methods like those based on CNNs. The project's other objective is the development of a system for identifying and categorizing leukaemia. Because of the vast structural differences between leukemic and normal cells, several features are retrieved from the segmented lymphocyte pictures for identification purposes.