Deep learning pipelines for white blood cell classification: a comprehensive literature review

Abstract

This study aims to provide a comprehensive and structured overview of how DL frameworks have been utilized for the recognition and classification of white blood cell (WBC) images. The review is guided by three primary research questions: (1) What are the dominant trends in DL model adoption for WBC classification? (2) How do image preprocessing and segmentation techniques influence performance? (3) What are the current limitations and prospects in this domain? A systematic literature review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Studies were selected and analyzed based on relevance, methodological rigor, and contribution to the field. An analytical framework comprising 18 technical dimensions was applied to evaluate the DL pipelines used. Convolutional neural networks (CNN) were found to dominate the field, used in 81.6% of the reviewed studies. Among these, ResNet, Visual Geometry Group (VGG), and EfficientNet emerged as the most frequently applied architectures (32.9%, 18.4%, and 14.5%, respectively). Hybrid approaches (CNN-Support Vector Machine, CNN-Long Short-Term Memory, and more) accounted for 11.2% of the literature. Preprocessing techniques such as histogram equalization, normalization, and contrast enhancement were employed in 68% of the studies, while segmentation methods (U-Net, watershed, and more) were used in 54%. The highest reported classification accuracy (ACC) was 99.83% using EfficientNet on the BCCD dataset. This review introduces a structured analytical framework to benchmark existing studies and guide the development of more robust, explainable, and clinically deployable DL-based diagnostic tools for hematological imaging.