Attention-Aware Residual Network Based Manifold Learning for White Blood Cells Classification

• Published in: IEEE journal of biomedical and health informatics 2021-04, Vol.25 (4), p.1206-1214
• Main Authors: Huang, Pu, Wang, Jing, Zhang, Jian, Shen, Yajuan, Liu, Cong, Song, Weiqing, Wu, Shangshang, Zuo, Yuwei, Lu, Zhiming, Li, Dengwang
• Format: Article
• Language: English
• Subjects: Ablation; Attention-aware; Classification; Computer architecture; Convolution; Embedding; Feature extraction; Image classification; Learning; Leukemia; Leukocytes; Machine learning; Manifold learning; Manifolds; Manifolds (mathematics); Mathematical models; Parameters; Residual network; Riemann manifold; Task analysis; Training; White blood cells
• ISSN: 2168-2194; 2168-2208
• DOI: 10.1109/JBHI.2020.3012711

The classification of six types of white blood cells (WBCs) is considered essential for leukemia diagnosis, while the classification is labor-intensive and strict with the clinical experience. To relieve the complicated process with an efficient and automatic method, we propose the A ttention-aware R esidual Network based M anifold L earning model (ARML) to classify WBCs. The proposed ARML model leverages the adaptive attention-aware residual learning to exploit the category-relevant image-level features and strengthen the first-order feature representation ability. To learn more discriminatory information than the first-order ones, the second-order features are characterized. Afterwards, ARML encodes both the first- and second-order features with Gaussian embedding into the Riemannian manifold to learn the underlying non-linear structure of the features for classification. ARML can be trained in an end-to-end fashion, and the learnable parameters are iteratively optimized. 10800 WBCs images (1800 images for each type) is collected, 9000 images and five-fold cross-validation are used for training and validation of the model, while additional 1800 images for testing. The results show that ARML achieving average classification accuracy of 0.953 outperforms other state-of-the-art methods with fewer trainable parameters. In the ablation study, ARML achieves improved accuracy against its three variants: without manifold learning (AR), without attention-aware learning (RML), and AR without attention-aware learning. The t-SNE results illustrate that ARML has learned more distinguishable features than the comparison methods, which benefits the WBCs classification. ARML provides a clinically feasible WBCs classification solution for leukemia diagnose with an efficient manner.