<![CDATA[Leukemia is a severe hematological malignancy that disrupts normal blood cell function, primarily affecting white blood cells (WBCs). Early and accurate Classification of white blood cells (WBCs) is essential for facilitating the accurate diagnosis of leukemia, thereby improving patient outcomes and reducing treatment costs. This paper provides a comprehensive review of recent deep learning and transfer learning approaches applied to WBC classification and leukemia diagnosis. Various models, including Convolutional Neural Networks (CNNs), Vision Transformers (ViT), and hybrid techniques combining handcrafted and learned features, are examined. Performance metrics such as accuracy, sensitivity, specificity, and F1-score are discussed across multiple datasets like BCCD, ALL-IDB, and Kaggle repositories. The study highlights the strengths of different models, addresses challenges such as class imbalance and data scarcity, and outlines future directions like the integration of multimodal data and real-time deployment. This review serves as a valuable resource for researchers and clinicians aiming to develop intelligent, automated systems for hematological disease diagnosis.]]>
Copyrights © 2026
