Agaian, SosMadhukar, Monica2024-02-122024-02-1220129781267843159https://hdl.handle.net/20.500.12588/4331This item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.Acute Leukemia (AL) is a type of cancer that affects the blood and the bone marrow. It is characterized by proliferation of immature white blood cells in the bone marrow. Detection and classification of white blood cells is a challenge in image processing, as manual data analysis is time consuming and most often not accurate. Research in this area is essential as a fully automated classifier tool can prove to be an effective ancillary for the physicians. This thesis focuses on developing a robust classifier system that a) demonstrates automated classification of peripheral blood smear images of Acute Leukemia containing multiple nuclei and b) validates the segmented images using multiple cross-validation methods. We also present the results interactively using a Graphical User Interface (GUI). The presented method involves: (1) acquisition of spectral images; (2) preprocessing of the acquired images; (3) calculation of future vectors including Nucleus Energy and Hausdorff Dimension [detection of single and touching cells in the scene]; (4) segmentation of the cells using color-based clustering; and (5) post-processing of the segmented regions. The system is evaluated for two types of Acute Leukemia - Acute Lymphoblastic Leukemia (ALL) and Acute Myelogenous Leukemia (AML). The database comprises of a set of 98 images for ALL (with 39 abnormal samples and 59 normal samples) obtained from Charles Fabio Scotti, Department of Information Technology - Università degli Studi di Milano and 80 images of AML (with 40 abnormal samples and 40 normal samples) obtained from American Society of Hematology. The system performs segmentation using k-means clustering and classifies the images using SVM classifier. The results show that 98% of the cases were correctly classified by the system for both Acute Lymphoblastic Leukemia as well as Acute Myelogenous Leukemia. This corroborates that the system presented acts as an efficient tool in classifying the subtypes of Acute Leukemia.112 pagesapplication/pdfAcute LeukemiaClassificationImage ProcessingSegmentationElectrical engineeringBiomedical engineeringThesis