AI-Driven Diagnosis of Leukemia Microscopic Blood Smear Images Using Vision Transformer Model

This paper provides insights from a systematic mapping study (SMS) and a systematic literature review (SLR) focused on deep learning techniques for leukemia diagnosis. By analyzing thirty studies, the review highlights a variety of methodologies that demonstrate promising levels of accuracy and sensitivity in detecting leukemia from blood smear images. The findings emphasize the importance of ongoing research to overcome current challenges and enhance the development of deep learning models that are accurate, reliable, and interpretable. Such advancements are essential for improving patient care and supporting informed decision-making in clinical settings. Integrating large language models (LLMs) into healthcare systems holds great potential to further transform diagnostics and treatment. LLMs’ ability to adapt to evolving data and contexts can drive advancements in disease prediction, personalized care, and operational efficiency.  However, addressing challenges like dataset limitations, ethical concerns, and practical implementation hurdles is critical to successfully adopting these technologies. With continued innovation and interdisciplinary collaboration, AI and machine learning can significantly reshape healthcare delivery, leading to improved outcomes and more equitable access to care. This paper highlights the potential of AI models in revolutionizing medical diagnosis and treatment. Significant progress has been made in improving healthcare facilities by utilizing the machine learning algorithms and more recent deep learning algorithms. It has opened the doors for precise and effective analysis of medical images helping in early identification of diseases like leukemia that we worked on in our project. Specially, ViT-B16, i.e., vision transformers, have a capability for medical image analysis with special benefits because of their ability to capture spatial relationships and global context in images. Moreover, our user-friendly graphical user interface has further expanded the possibilities for medical workers to identify diseases in seconds, making it a valuable tool for healthcare professionals. The promising results obtained from the vision transformer in leukemia detection suggest its potential for use in future diagnostic tasks in the field of medical and healthcare. For future work, our user interface can be further transformed into web application for professional use in healthcare. It can be progressed further by keeping main focus on getting relatively larger and diverse dataset that can enhance accuracy and performance of the model significantly for this purpose various data augmentation techniques can be utilized.

1.Robert W Miller, John L Young Jr, and BiljanaNovakovic, “Childhood cancer,” Cancer, vol. 75, no. S1, pp. 395–405, 1995. 2.Nizar Ahmed, AltugYigit, ZerrinIsik, and AdilAlpkocak, “Identification of leukemia subtypes from microscopic images using convolutional neural network,” Diagnostics, vol. 9, no. 3, pp. 104, 2019.  3.Rodrigo Veras, Andre Santana, Jo ´ ao Tavares, ˜ Justino Santos, and Vinicius Machado, “Convolution neural network models for acute leukemia diagnosis,” in 2020 International Conference on Systems, Signals and Image Processing (IWSSIP). IEEE, 2020. 4.Fatichah, C. Identification of acute lymphoblastic leukemia subtypes in touching cells based on enhanced edge detection. Int. J. Intell. Eng. Syst. 2020 5. De Sant, Y.F.D. Lightweight Classification of Normal Versus Leukemic Cells Using Feature Extraction. In Proceedings of the 2021 6.Suciati, N. Deep Learning for the Detection of Acute Lymphoblastic Leukemia Subtypes on Microscopic Images: A Systematic Literature Review. IEEE Access 2023 7. G Meena Devi and V Neelambary, “Computer-aided diagnosis of white blood cell leukemia using vgg16 convolution neural network,” in 2022 4th International Conference on Inventive Research in Computing Applications (ICIRCA). IEEE, 2022