Enhanced Leukemic Cell Classification Using Attention-Based CNN and Data Augmentation

TL;DR

This paper introduces an attention-based CNN with data augmentation for accurate, efficient leukemic cell classification, demonstrating significant performance improvements and interpretability on clinical datasets.

Contribution

It presents a novel deep learning pipeline combining EfficientNetV2-B3 with attention mechanisms, optimized for robustness, interpretability, and computational efficiency in medical image analysis.

Findings

Achieved 97.89% F1-score and accuracy on the C-NMC 2019 dataset.

Outperformed existing methods by up to 4.67%.

Reduced model parameters by 89% compared to VGG16.

Abstract

We present a reproducible deep learning pipeline for leukemic cell classification, focusing on system architecture, experimental robustness, and software design choices for medical image analysis. Acute lymphoblastic leukemia (ALL) is the most common childhood cancer, requiring expert microscopic diagnosis that suffers from inter-observer variability and time constraints. The proposed system integrates an attention-based convolutional neural network combining EfficientNetV2-B3 with Squeeze-and-Excitation mechanisms for automated ALL cell classification. Our approach employs comprehensive data augmentation, focal loss for class imbalance, and patient-wise data splitting to ensure robust and reproducible evaluation. On the C-NMC 2019 dataset (12,528 original images from 62 patients), the system achieves a 97.89% F1-score and 97.89% accuracy on the test set, with statistical validation through 100-iteration Monte Carlo experiments confirming significant improvements (p < 0.001) over baseline methods. The proposed pipeline outperforms existing approaches by up to 4.67% while using 89% fewer parameters than VGG16 (15.2M vs. 138M). The attention mechanism provides interpretable visualizations of diagnostically relevant cellular features, demonstrating that modern attention-based architectures can improve leukemic cell classification while maintaining computational efficiency suitable for clinical deployment.