An efficient framework for Automatic segmentation and Analysis of complex and Overlapped white blood cells From microscopic blood images

Abstract

In the present scenario, an increasing number of advancements in the Computer-Aided Diagnosis (CAD) system are used to minimize the cost of treating diseases. Thus, the microscopic image processing occupies a major role in hematology, which analyzes White Blood Cells (WBCs or leukocytes) for the detection of blood diseases such as leukemia, anemia, cancer and other infectious diseases. Leukocytes morphology analysis is performed by visually examine in geometric structure of cells under microscope to predict the type of disease and the different stage of diseases. In market, the analysis of leukocytes in blood smear is performed with the help of manual and automated methods. In manual analysis, microscopic image screening is the common procedure of detecting certain blood diseases, where the pathologists examine the cell structures (morphological features) and the cellular distributions (placement of cells) under microscope to evaluate the state and stage of blood diseases. Therefore, the CAD system was developed with the use of image processing and machine learning algorithms. This system includes various steps such as color conversion, segmentation, feature extraction, counting, and classification. In this thesis, we have developed a framework to classify and count the leukocytes in microscopic images and developed framework supports the pathologist and boosts the CAD system accuracy. The process of analyzing leukocytes in microscopic blood images is the major concern of this research. In this thesis, first we work on WBCs segmentation from microscopic blood images. newline