A Survey of Methods and Architectures for Segmentation and Classification of White Blood Cancer Cells in Bone Marrow Microscopic Images Using Deep Kronecker Neural Networks

Abstract

Artificial Intelligence (AI) has significantly advanced medical image analysis, particularly in diagnosing hematological malignancies such as leukemia, where early and accurate detection is vital for improving patient outcomes. Leukemia originates in the bone marrow and involves abnormal proliferation of white blood cells, making microscopic examination essential but often labor-intensive, subjective, and prone to variability among clinicians. This survey reviews recent developments from 2020 to 2023 in deep learning-based methods for segmentation and classification of cancerous white blood cells in bone marrow microscopic images. Advanced architectures, including Convolutional Neural Networks (CNNs), U-Net, Mask R-CNN, and transformer-based models, have demonstrated superior performance by automatically extracting complex features and improving classification accuracy, precision, and recall compared to traditional techniques. The survey also highlights emerging approaches such as Deep Kronecker Neural Networks (DKNN), which reduce computational complexity while preserving strong representational capabilities, making them suitable for high-dimensional medical data. Additionally, innovations like attention mechanisms, multimodal learning, and automated cytology systems have enhanced diagnostic efficiency and accuracy. Despite these advancements, challenges such as limited annotated datasets, class imbalance, domain variability, and lack of model interpretability persist. The study concludes by emphasizing future research directions, including explainable AI, integration of multimodal data, and development of real-time, clinically deployable systems to support reliable and efficient leukemia diagnosis.