Recent Advances in Segmentation and Classification of White Blood Cancer Cells in Bone Marrow Microscopic Images Using Deep Kronecker Neural Networks: A Systematic Review

Abstract

White blood cancer, particularly leukemia, originates in the bone marrow and leads to abnormal proliferation of white blood cells, significantly affecting the immune system. Accurate segmentation and classification of cancerous cells in bone marrow microscopic images are essential for early diagnosis and effective treatment planning. Traditional diagnostic approaches rely on manual examination by hematologists, which is time-consuming, subjective, and prone to variability.

Recent advancements in artificial intelligence, particularly deep learning, have enabled automated and highly accurate analysis of microscopic images. Convolutional Neural Networks (CNNs), U-Net variants, and hybrid models have demonstrated significant success in white blood cell segmentation and leukemia classification. More recently, Deep Kronecker Neural Networks have emerged as an efficient approach for handling high-dimensional data by reducing computational complexity while preserving structural information.

This systematic review analyzes recent studies from 2020 to 2023, focusing on segmentation and classification techniques for white blood cancer detection. Comparative analysis reveals that hybrid deep learning models and attention-based architectures achieve high accuracy, often exceeding 98%. However, challenges such as data scarcity, class imbalance, and lack of interpretability remain. Future research should focus on developing efficient, explainable, and clinically deployable AI systems for bone marrow image analysis.