Developing an efficient VGG19-based model and transfer learning for detecting acute lymphoblastic leukemia (ALL)

Abstract

Acute lymphoblastic leukemia (ALL) is a form of blood cancer that affects the lymphoid cells, leading to the excessive proliferation of immature lymphocytes. A pathologist typically examines the bone marrow to recognize the specific type of leukemia cells present. However, This time-honoured approach takes a lot of effort and time and may not always yield accurate results due to variations in specialist expertise. As a result, there is a need for automated methods that can increase efficiency and accuracy in identifying leukemia cells. Deep learning techniques have shown promise in this regard, as they can analyze images of leukemia cells and make predictions about their type. In our study, we utilized the VGG19 convolutional neural network (CNN) model to analyze images from the ALL-IDB-1 dataset of ALL. Our results demonstrate a remarkable accuracy rate of 99.49%, indicating that our proposed model outperformed other tested models in simplicity and performance. These findings suggest that machine learning and deep learning techniques may offer an effective way to streamline the identification of leukemia cells and improve patient outcome.