Leukemia Classification using Transfer Learning Models
Srijit Kundu, Diptayan Jash, Rudrajit Dutta, Deeba Kannan, K. C. Prabu Shankar, Fitri Yakub · 2024
Leukemia, a complex cancer of the blood and bone marrow, involves the uncontrolled growth of abnormal, immature white blood cells. Accurate and timely classification of leukemia sub-types is crucial for effective treatment planning and patient management. Traditionally, this relies on microscopic blood cell analysis, a laborious process susceptible to human error. Automated classification systems using machine learning hold promise for improving efficiency and accuracy. This paper investigates the application of various transfer learning models for classifying leukemia sub types from blood images. Through rigorous experimentation and validation, we demonstrate the effectiveness and robustness of our approach in accurately classi-fying leukemia sub types. Among the models tested, InceptionV3performs the best, achieving an accuracy of 95.06%. Finally, we address the ethical considerations and future directions of this research area, highlighting its potential to improve diagnostic accuracy and expedite treatment decisions.