Abstract:<p>White blood cells (WBC) are important parts of our immune system and they protect our body against infections by eliminating viruses, bacteria, parasites and fungi. There are five types of WBC. These are called Lymphocytes, Monocytes, Eosinophils, Basophils and Neutrophils. The number of WBC types and the total number of WBCs provide important information about our health status. Diseases such as leukemia, AIDS, autoimmune diseases, immune deficiencies, blood diseases can be diagnosed based on the number of WBCs. In this study, a computer-aided automated system that can easily identify and locate WBC types in blood images has been proposed. Current blood test devices usually detect WBCs with traditional image processing methods such as preprocessing, segmentation, feature extraction, feature selection and classification. Deep learning methodology is superior to traditional image processing methods in literature. In addition, traditional methods require the appearance of the whole object to be able to recognize objects. Contrary to traditional methods, convolutional neural networks (CNN), a deep learning architecture, can extract features from a part of an object and perform object recognition. In this case, a CNN-based system shows a higher performance in recognizing partially visible cells for reasons such as overlap or only partial visibility of the image. Therefore, it has been the motivation of this study to increase the performance of existing blood test devices with deep learning method. Blood cells have been identified and classified by Regional Based Convolutional Neural Networks. Designed architectures have been trained and tested by combining BCCD data set and LISC data set. Regional Convolutional Neural Networks (R - CNN) has been used as a methodology. In this way, different cell types within the same image have been classified simultaneously with a detector. While training CNN which is the basis of R - CNN architecture; AlexNet, VGG16, GoogLeNet, ResNet50 architectures have been tested with full learning and transfer learning. At the end of the study, the system has showed 100% success in determining WBC cells. ResNet50, one of the CNN architectures, has showed the best performance with transfer learning. Cell types of Lymphocyte were determined with 99.52% accuracy rate, Monocyte with 98.40% accuracy rate, Basophil with 98.48% accuracy rate, Eosinophil with 96.16% accuracy rate and Neutrophil with 95.04% accuracy rate.</p>

Leukocyte Recognition with Convolutional Neural Network

LeukocyteMask: An automated localization and segmentation method for leukocyte in blood smear images using deep neural networks.

Efficient leukocytes detection and classification in microscopic blood images using convolutional neural network coupled with a dual attention network

A Peripheral Blood WBC Classification with Convolutional Neural Network

Improved U-net-based leukocyte segmentation method

Segmentation, feature extraction and classification of leukocytes leveraging neural networks, a comparative study

White blood cell classification and counting using convolutional neural network

A novel algorithm for segmentation of leukocytes in peripheral blood

Subclass Separation of White Blood Cell Images Using Convolutional Neural Network Models

Hybrid adversarial-discriminative network for leukocyte classification in leukemia

Deep learning-based image annotation for leukocyte segmentation and classification of blood cell morphology

WBC-AMNet: Automatic Classification of WBC Images Using Deep Feature Fusion Network Based on Focalized Attention Mechanism.

ResNeXt-CC: a novel network based on cross-layer deep-feature fusion for white blood cell classification

White blood cells detection and classification based on regional convolutional neural networks

Efficient Classification of White Blood Cell Leukemia with Improved Swarm Optimization of Deep Features

Segmentation and Classification of White Blood Cells Using the UNet

High-Efficiency Classification of White Blood Cells Based on Object Detection

White Blood Cell Image Segmentation Based on Color Component Combination and Contour Fitting

White blood cell classification via a discriminative region detection assisted feature aggregation network

A Novel White Blood Cell Segmentation Scheme Based on Feature Space Clustering

Improved White Blood Cells Classification Based on Pre-trained Deep Learning Models