W-Net: A CNN-based Architecture for White Blood Cells Image Classification

TL;DR

This paper introduces W-Net, a CNN-based architecture that accurately classifies white blood cells from microscopic images, achieving 97% accuracy on a large real-world dataset, aiding immune system diagnostics.

Contribution

The paper presents W-Net, a novel CNN model specifically designed for white blood cell classification, demonstrating high accuracy on a large-scale dataset.

Findings

W-Net achieves 97% average accuracy.

The method effectively classifies five WBC types.

It outperforms previous approaches in accuracy.

Abstract

Computer-aided methods for analyzing white blood cells (WBC) have become widely popular due to the complexity of the manual process. Recent works have shown highly accurate segmentation and detection of white blood cells from microscopic blood images. However, the classification of the observed cells is still a challenge and highly demanded as the distribution of the five types reflects on the condition of the immune system. This work proposes W-Net, a CNN-based method for WBC classification. We evaluate W-Net on a real-world large-scale dataset, obtained from The Catholic University of Korea, that includes 6,562 real images of the five WBC types. W-Net achieves an average accuracy of 97%.