IRO Journals
Journal of Innovative Image Processing

A Hybrid Vision Transformer Model for Leukemia Image Classification

Open Access
Published: 30 April, 2026
Pages: 575-595
  • Revathi B. , Revathi B.
    Artificial Intelligence and Data Science, Ramco Institute of Technology, Virudhunagar, India
    Artificial Intelligence and Data Science, Ramco Institute of Technology, Virudhunagar, India
  • Kaliappan M. , Kaliappan M.
    Artificial Intelligence and Data Science, Ramco Institute of Technology, Virudhunagar, India
    Artificial Intelligence and Data Science, Ramco Institute of Technology, Virudhunagar, India
  • Anandhi S.V. , Anandhi S.V.
    Artificial Intelligence and Data Science, Ramco Institute of Technology, Virudhunagar, India
    Artificial Intelligence and Data Science, Ramco Institute of Technology, Virudhunagar, India
  • Kezial Elizabeth S.K. Kezial Elizabeth S.K.
    Computer Science and Engineering, Mangayarkarasi College of Engineering, Madurai, India
    Computer Science and Engineering, Mangayarkarasi College of Engineering, Madurai, India
view PDF
view PDF

How to Cite

B., Revathi, Kaliappan M., Anandhi S.V., and Kezial Elizabeth S.K. 2026. “A Hybrid Vision Transformer Model for Leukemia Image Classification”. Journal of Innovative Image Processing 8 (2): 575-95. https://doi.org/10.36548/jiip.2026.2.008.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Classification
Vision Transformer
Kolmogorov Arnold Network
Multilayer Perceptron
XGBoost
Microscopic Images

Abstract

This work presents an automated method for distinguishing leukemic blast cells from normal bone marrow cells in microscopic blood smear images. The approach uses a Vision Transformer (ViT) as the main feature extractor, combined with a Kolmogorov–Arnold Network (KAN) representation and an XGBoost classifier for final classification. Typically, a multilayer perceptron (MLP) head is used to classify images in most Vision Transformer architectures. In this study, we explore the potential improvement of transformer feature discernibility for leukemia detection by replacing the conventional MLP representation with a KAN-based transformation. The pre-trained Vision Transformer processes CLS-token embeddings extracted from the images. Subsequently, these embeddings are classified using an XGBoost model and evaluated through the KAN representation. The proposed ViT–KAN–XGBoost model achieved an accuracy of 85.11% on the C-NMC 2019 dataset, which is higher than the 83.22% obtained with the ViT–MLP–XGBoost model. This suggests that adding KAN-based representations to transformer features can improve classification performance and may be useful for leukemia screening tasks.

References

  1. Jaffe, R. "WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues." World health organization classification of tumours (2008): 358-360.
  2. Goda, Chinmayee, Rohan Kulkarni, Yaphet Bustos, Wenjun Li, Alexander Rudich, Ozlen Balcioglu, Sadie Chidester et al. "Cellular Taxonomy of the Preleukemic Bone Marrow Niche of Acute Myeloid Leukemia." Leukemia 39, no. 1 (2025): 51-63.
  3. Sasada, Keiko, Noriko Yamamoto, Hiroki Masuda, Yoko Tanaka, Ayako Ishihara, Yasushi Takamatsu, Yutaka Yatomi, Waichiro Katsuda, Issei Sato, and Hirotaka Matsui. "Inter-Observer Variance and the Need for Standardization in the Morphological Classification of Myelodysplastic Syndrome." Leukemia research 69 (2018): 54-59.
  4. Dosovitskiy, Alexey, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani et al. "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale." arXiv preprint arXiv:2010.11929 (2020).
  5. Kayalibay, Baris, Grady Jensen, and Patrick Van Der Smagt. "CNN-Based Segmentation of Medical Imaging Data." arXiv preprint arXiv:1701.03056 (2017).
  6. Liu, Ziming, Yixuan Wang, Sachin Vaidya, Fabian Ruehle, James Halverson, Marin Soljačić, Thomas Y. Hou, and Max Tegmark. "Kan: Kolmogorov-Arnold Networks." arXiv preprint arXiv:2404.19756 (2024).
  7. Chen, Tianqi, and Carlos Guestrin. "Xgboost: A Scalable Tree Boosting System." In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, 785-794.
  8. Halim, Nurul Hazwani Abd, Mohd Yusoff Mashor, and Rosline Hassan. "Classification of Acute Leukemia Based on Multilayer Perceptron." In Journal of Physics: Conference Series, vol. 1372, no. 1, IOP Publishing, 2019, 012044.
  9. Shah, Waqar Hussain, Abdullah Baloch, Rider Jaimes-Reátegui, Sohail Iqbal, Syeda Rafia Fatima, and Alexander N. Pisarchik. "Acute Lymphoblastic Leukemia Classification Using Persistent Homology." The European Physical Journal Special Topics 234, no. 15 (2025): 4583-4596.
  10. Hegde, Roopa B., Keerthana Prasad, Harishchandra Hebbar, and I. Sandhya. "Peripheral Blood Smear Analysis Using Image Processing Approach for Diagnostic Purposes: A Review." Biocybernetics and Biomedical Engineering 38, no. 3 (2018): 467-480.
  11. Cho, Priscilla, Sajal Dash, Aristeides Tsaris, and Hong-Jun Yoon. "Image Transformers for Classifying Acute Lymphoblastic Leukemia." In Medical Imaging 2022: Computer-Aided Diagnosis, vol. 12033, SPIE, 2022, 647-653.
  12. Rao, Pritam, Ashutosh Naik, Chirag Rana, and Sunil Ghane. "Segmentation of Nuclei using Transformer based Architecture." In 2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT), IEEE, 2022, 1-6.
  13. Zhang, Xian-Ya, Di Zhang, Wang Zhou, Zhi-Yuan Wang, Chao-Xue Zhang, Jin Li, Liang Wang, and Xin-Wu Cui. "Predicting Lymph Node Metastasis in Papillary Thyroid Carcinoma: Radiomics Using Two Types of Ultrasound Elastography." Cancer Imaging 25, no. 1 (2025): 13.
  14. Raju, Akella Subrahmanya Narasimha, K. Venkatesh, B. Padmaja, CH N. Santhosh Kumar, Pattabhi Rama Mohan Patnala, Ayodele Lasisi, Saiful Islam, Abdul Razak, and Wahaj Ahmad Khan. "Exploring Vision Transformers and XGBoost as Deep Learning Ensembles for Transforming Carcinoma Recognition." Scientific Reports 14, no. 1 (2024): 30052.
  15. Gupta, Ritu, Shiv Gehlot, and Anubha Gupta. "C-NMC: B-Lineage Acute Lymphoblastic Leukaemia: A Blood Cancer Dataset." Medical Engineering & Physics 103, no. 1 (2022): 103793.
  16. Prasad, Prakeerth. "Acute Lymphoblastic Leukemia Subtypes Detection Using Vision Transformer Model." In 2024 5th International Conference on Data Intelligence and Cognitive Informatics (ICDICI), IEEE, 2024, 1413-1418.
  17. Hornik, Kurt, Maxwell Stinchcombe, and Halbert White. "Multilayer Feedforward Networks are Universal Approximators." Neural networks 2, no. 5 (1989): 359-366.
  18. Agnihotri, Diya, Bhavpreet Kaur, and Law Kumar Singh. "Deep Learning Approaches for Blood Cancer Detection: A Comprehensive Review." In 2025 7th International Conference on Information Systems and Computer Networks (ISCON), IEEE, 2025, 1-6.
  19. Xie, Wentao, Xinye Jiang, Jingying Huang, Mingwei Qin, and Zhisheng Bi. "Research Advances in the Adjunctive Diagnosis of Acute Myeloid Leukemia." Frontiers in Oncology 15 (2025): 1634935.
  20. Maruf, Md, Md Mahbubul Haque, and Bishowjit Paul. "Deep Learning with Self-Attention and Enhanced Preprocessing for Precise Diagnosis of Acute Lymphoblastic Leukemia from Bone Marrow Smears in Hemato-Oncology." arXiv preprint arXiv:2508.17216 (2025).
  21. Kılıç, Şafak. "Attention-Based Dual-Path Deep Learning for Blood Cell Image Classification Using ConvNeXt and Swin Transformer." Journal of Imaging Informatics in Medicine (2025): 1-19.
  22. Revathi, B., M. Kaliappan, P. Chanthiya, And Sv Anandhi. "Superpixel Based Classification for Graph Neural Network (GNN) in Leukemia Images." Journal of Theoretical and Applied Information Technology 103, no. 23 (2025).
  23. Ramaneswaran, S., Kathiravan Srinivasan, PM Durai Raj Vincent, and Chuan-Yu Chang. "Hybrid Inception V3 XGBoost Model for Acute Lymphoblastic Leukemia Classification." Computational and Mathematical Methods in Medicine 2021, no. 1 (2021): 2577375.