GWO-Based Fed-UNet-CNN Model for Leukocyte Classification Across Developmental Stages

Accurate leukocyte classification plays a crucial role in supporting hematological analysis and disease monitoring. In this study, we propose a deep learning-based computational framework for leukocyte classification across developmental stages using publicly available datasets. The workflow incorporates preprocessing techniques, including Contrast Limited Adaptive Histogram Equalization (CLAHE), followed by data augmentation using a Generative Adversarial Network (GAN) to improve dataset diversity and robustness. A U-Net architecture is employed for precise segmentation of leukocyte regions, and a Convolutional Neural Network (CNN) is used for feature extraction and classification. Additionally, a federated learning approach is integrated to enable collaborative model training across decentralized datasets while preserving data privacy. The proposed GWO-based Fed-UNet-CNN model demonstrates strong performance, achieving an overall accuracy of 99.29% on benchmark datasets. These results indicate the potential of the proposed approach as a computational decision-support tool for leukocyte classification. However, further validation using real-world clinical data is required before deployment in clinical settings.