Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{181049,
        author = {Vinod A M and Ananya Kumar and Lekhana V and Kruthi H A and Malegere Ramesh Vinay},
        title = {UNet++ : A Deep Learning Approach to Leukemia Cell Segmentation},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {3252-3259},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=181049},
        abstract = {The recent verticals in computer-aided leukemia detection have been leveling the automation of white blood cell (WBC) segmentation and classification from peripheral blood smear images for early and accurate diagnosis. Deep learning approaches mainly consisting of U-Net and its variants-U-Net++, U-Net-VGG16/VGG19, U-Net-ResNet, WBC-Net have been reported to have very high segmentation accuracy, with Dice metric going up to 90% and IoU up to 83%. Some of these works are also considered hybrid methods using feature extraction (like GLCM, morphological, and geometric descriptors) combined with classical classifiers such as Random Forest, SVM, Naive Bayes, and K-NN for classifying subtypes of chronic and acute lymphoblastic leukemias (ALL). Even pre-processing operations such as color space transformation, watershed segmentation, and marker-based algorithms help greatly to improve image quality and segmentation results. A better insight into transfer learning and the design of appropriate loss functions put the advantage on deep learning models over traditional ones, while results produced in the deployment evaluation indicate that simpler variations of U-Net may deliver more consistent results in practice. All of these techniques combined could be the bundle of automation-leukemia diagnosis services.},
        keywords = {White Blood Cell (WBC) Segmentation, Deep Learning, U-Net and its Variants (U-Net++, U-Net-VGG16/VGG19, U-Net-ResNet, WBC-Net), Leukemia Classification, Feature Extraction (GLCM, Morphological and Geometric Descriptors), Transfer Learning.},
        month = {June},
        }