White blood cells identification system based on convolutional deep neural learning networks. (January 2019)

• Record Type: Journal Article
• Authors: Shahin, A.I.; Guo, Yanhui; Amin, K.M.; Sharawi, Amr A.
• Is Part Of: Computer methods and programs in biomedicine. Volume 168(2019)
• Page Start: 69
• Page End: 80
• Publication Date: 2019-01
• Subjects: Blood smear image -- Deep learning -- Transfer deep learning -- WBCs identification -- Deep features visualization; ACM Automated cell morphology -- RBCs Red blood cells -- WBCs White blood cells -- FS Feature selection -- DL Deep learning -- GPU Graphic processor unit -- TLA Transfer learning approach -- DeCA Deep convolutional activation -- DeCAF Deep convolutional activation features -- ANN Artificial neural network -- CNN Convolutional neural network -- DCNNs Deep convolutional neural networks -- PS Pooling stride -- ReLU Rectified linear unit; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2017.11.015
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: Transfer learning approaches are employed to build a complete identification system for white blood cells (WBCs) in blood smear image. A complete end-to-end CNN architecture is proposed to identify the different WBCs types (basophil, eosinophil, lymphocyte, monocyte and neutrophil). The proposed CNN architecture (WBCsNet) is the first deep learning architecture for WBCs identification which achieved a higher classification accuracy more than the traditional WBCs identification systems or even the transfer learning approaches. The proposed WBCsNet is re-employed as a pre-trained network and achieves high accuracy scores for classifying limited balanced dataset. We present features visualization for the WBCsNet activation which reflect higher reasonable response than the pre-trained activated one. Abstract: Background and objectives: White blood cells (WBCs) differential counting yields valued information about human health and disease. The current developed automated cell morphology equipments perform differential count which is based on blood smear image analysis. Previous identification systems for WBCs consist of successive dependent stages; pre-processing, segmentation, feature extraction, feature selection, and classification. There is a real need to employ deep learning methodologies so that the performance of previous WBCs identification systems can be increased. Classifying small limited datasets through deep learning systems is a major challenge and should be investigated. Methods: In this paper, we propose a novel identification system for WBCs based on deep convolutional neural networks. Two methodologies based on transfer learning are followed: transfer learning based on deep activation features and fine-tuning of existed deep networks. Deep acrivation featues are extracted from several pre-trained networks and employed in a traditional identification system. Moreover, a novel end-to-end convolutional deep architecture called "WBCsNet" is proposed and built from scratch. Finally, a limited balanced WBCs dataset classification is performed through the WBCsNet as a pre-trained network. Results: During our experiments, three different public WBCs datasets (2551 images) have been used which contain 5 healthy WBCs types. The overall system accuracy achieved by the proposed WBCsNet is (96.1%) which is more than different transfer learning approaches or even the previous traditional identification system. We also present features visualization for the WBCsNet activation which reflects higher response than the pre-trained activated one. Conclusion: a novel WBCs identification system based on deep learning theory is proposed and a high performance WBCsNet can be employed as a pre-trained network. Graphical abstract: