Diagnosis of Malignant Haematopoietic Diseases based on the Automation of Blood Microscopic Image Analysis

Abstract

Leukemia is one of the leading causes of death among human and the prognosis highly depends on the early detection and diagnosis of the disease. In clinical practice, the primary suspicion of the leukemia is determined by the manual microscopic evaluation of peripheral blood smear image. Since this diagnostic method is time-consuming, lacks standardized accuracy and prone to errors due to various human factors, there is a high demand for the automated system, which would minimize the human intervention. The aim of this thesis is to propose a computer-aided leukemia diagnostic system based on image processing and machine learning techniques. To detect and identify the leukemic cells in the blood smear, methodologies such as image preprocessing, image segmentation, features extraction and classification are implemented. The overall automated system achieves recognition rates of 96.53 % by using artificial neural network and 97.92 % by implementing the polynomial SVM classification. The proposed system is successfully implemented in software development environment LabVIEW.