Detection and Classification of Various Types of Leukemia Using Image Processing, Transfer Learning and Ensemble Averaging Techniques

Abstract

Leukemia, often known as blood cancer, is a frequent and deadly illness. The abrupt rise of White Blood Cells in the blood causes a dangerous illness. Leukemia starts in the bone marrow and spreads throughout the body. It is responsible for the development of a huge number of cells that are aberrant. Leukemia can be categorized into four types. They are Acute myeloid leukaemia, Acute lymphoblastic leukaemia, chronic lymphocytic leukaemia, and Chronic myeloid leukemia. The early detection of leukemia plays crucial role. The detecting process will be more accurate if the diagnosis is computer-assisted. This project aims at developing an approach for detecting leukemia using machine learning techniques such as image processing and transfer learning which provides ways to see if it's a case of AML, ALL, CML, or CLL. Leukemic and healthy lymphocytes have markedly different morphological components. The lymphocyte images have been used to extract a variety of characteristics. The leukemia is further classified into its types by using a hybrid model constructed with the ensemble averaging of Inception, Xception convolutional models.