Leukemia, a severe form of blood cancer, disrupts white blood cells, weakening the immune system. Early detection and diagnosis are critical. This study utilizes artificial intelligence, specifically K-means and Gaussian Mixture Model (GMM) clustering algorithms, to classify and predict four types of leukemia—Chronic Myelogenous Leukemia (CML), Acute Lymphocytic Leukemia (ALL), Non-Hodgkin Lymphoma (NHL), and Hodgkin Lymphoma (HL) along with benign cases. Using blood sample reports, we compare the performance of these algorithms based on accuracy, precision, and recall. Accuracy measures the correctness of predictions, precision evaluates the avoidance of false positives, and recall assesses the identification of all leukemia instances. In the course of this study, we have developed a conceptual framework that outlines the end-to-end process, from data acquisition to final classification. This framework integrates clustering techniques to optimize leukemia diagnosis, providing a systematic approach for analysing blood sample data. Our findings highlight the strengths and weaknesses of K-means and GMM, guiding the selection of the most effective algorithm for reliable leukemia diagnosis through AI and machine learning techniques.