<p>Leukemia is one of the cancers that affects the bone marrow and blood, characterized by the immature and abnormal proliferation of white blood cells in the bone marrow, which then flow into the bloodstream. It disrupts the normal blood function and the immune system. Many people, including children, are affected by leukemia. So, early detection and accurate diagnosis are important for timely treatment and improved patient outcomes. Peripheral blood smear and bone marrow image analysis play a crucial role in the initial diagnosis and prognostic decisions. In recent years, machine learning and deep learning techniques have been studied for the automated leukemia detection and classification using microscopic images. These approaches are developed to reduce human error and facilitate early detection. This survey analyzes a systematic review of the literature on the detection and classification of acute leukemia, along with a comprehensive overview of computational learning processes, including preprocessing, augmentation, segmentation, and feature extraction. The review also covers performance metrics and validation methods. It also addresses the challenges specific to leukemia detection and classification models, including dataset availability, model generalizability, interpretability for clinical deployment, and future research considerations.</p>