Published Paper Details:

Determining cancer's stage (extent) is necessary for most cancer types. The size of the tumor and the scope to which cancer has spread control the location. However, this can be useful for determining a patient's prognosis and treatment. In contrast, Acute Lymphocytic Leukemia (ALL) does not typically form tumors. Instead, it usually affects all of the bone marrow in the body, which can spread to other organs, such as the liver and spleen. So, unlike most other cancers, ALL is not staged.Image segmentation is the most crucial part in image processing techniques. Nu- merous segmentation techniques are used to segment digital images into smaller re- gions called segments, consisting of sets of pixels in order to analyze important infor- mation from the images. Segmentation simplifies the process of information retrieval from the region of interest .It helps in converting the digital image into more relevant information and easier to analyze. This presents a comparative analysis of existing segmentation techniques and its modification to form new segmentation techniques to overcome some of the drawbacks of the existing image segmentation approaches.Leukemia, a type of cancer affecting the blood and bone marrow, requires early detection and accurate classification for effective treatment. Traditional diagnostic methods often involve manual examination of blood smears, which is time-consuming and subjective. In this study, we present a MATLAB-based approach for automated leukemia detection using image processing and machine learning techniques. We preprocess microscopic blood smear images, segment leukemia cells using K-means clustering and Otsu's thresholding method, and extract features to characterize cell morphology and texture. A K-nearest neighbors (KNN) classifier is employed to clas- sify segmented cells into leukemia and non-leukemia categories. The performance of the proposed methodology is evaluated using accuracy, precision, recall, and F1-score metrics. Our results demonstrate the effectiveness of the computational approach in accurately identifying and classifying leukemia cells, with implications for improving diagnostic efficiency and patient outcomes in leukemia management.