Comparative Study of Feature Selection for White Blood Cell Differential Counts in Low Resolution Images

Habibzadeh, Mehdi; Krzyżak, Adam; Fevens, Thomas

doi:10.1007/978-3-319-11656-3_20

Cited by 9 publications

(2 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is seen that the best models, ResNet-50, DenseNet-121, and DenseNet-169, incur the largest inference times. is is not surprising, given that they [46] 10-Fold CV Private 93.9 Neural network + PCA [47] 75%: 25% Kanbilim 95.0 W-net [48] 10-Fold CV Private 97.0 W-net [48] 10-Fold CV LISC + private 96.0 Linear SVM [49] 10-Fold CV CellaVision 85.0 ResNet-50 DenseNet-169 Model Figure 11: Time for the DNN models to perform inference on the validation data. e original data without data augmentation given in Table 2 is used for this experiment.…”

Section: Resultsmentioning

confidence: 99%

Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network

Almezhghwi

Serte

2020

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

White blood cells (leukocytes) are a very important component of the blood that forms the immune system, which is responsible for fighting foreign elements. The five types of white blood cells include neutrophils, eosinophils, lymphocytes, monocytes, and basophils, where each type constitutes a different proportion and performs specific functions. Being able to classify and, therefore, count these different constituents is critical for assessing the health of patients and infection risks. Generally, laboratory experiments are used for determining the type of a white blood cell. The staining process and manual evaluation of acquired images under the microscope are tedious and subject to human errors. Moreover, a major challenge is the unavailability of training data that cover the morphological variations of white blood cells so that trained classifiers can generalize well. As such, this paper investigates image transformation operations and generative adversarial networks (GAN) for data augmentation and state-of-the-art deep neural networks (i.e., VGG-16, ResNet, and DenseNet) for the classification of white blood cells into the five types. Furthermore, we explore initializing the DNNs’ weights randomly or using weights pretrained on the CIFAR-100 dataset. In contrast to other works that require advanced image preprocessing and manual feature extraction before classification, our method works directly with the acquired images. The results of extensive experiments show that the proposed method can successfully classify white blood cells. The best DNN model, DenseNet-169, yields a validation accuracy of 98.8%. Particularly, we find that the proposed approach outperforms other methods that rely on sophisticated image processing and manual feature engineering.

show abstract

Section: Resultsmentioning

confidence: 99%

Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network

Almezhghwi

Serte

2020

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…The textural information extraction from preprocessed images can be carried out by various methods [7], [8]. However, image description by means of pseudo-Zernike (PZ) moments [9] was chosen for the cell subtype identification because it was proven to be a reliable method for the recognition of shapes [10], characters [11], [12], faces [13], [14], [15], and viruses [16].…”

Section: Introductionmentioning

confidence: 99%

Leukocyte subtypes classification by means of image processing

Ryabchykov¹,

Ramoji²,

Bocklitz³

et al. 2016

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

The classification of leukocyte subtypes is a routine method to diagnose many diseases, infections, and inflammations. By applying an automated cell counting procedure, it is possible to decrease analysis time and increase the number of analyzed cells per patient, thereby making the analysis more robust. Here we propose a method, which automatically differentiate between two white blood cell subtypes, which are present in blood in the highest fractions. We apply generalized pseudo-Zernike moments to transfer morphological information of the cells to features and subsequently to a classification model. The first results indicate that information from the morphology can be used to obtain efficient automatic classification, which was demonstrated for the leukocyte subtype classification of neutrophils and lymphocytes. The approach can be extended to other imaging modalities, like different types of staining, spectroscopic techniques, dark field or phase contrast microscopy.

show abstract

Galaxies image classification using artificial bee colony based on orthogonal Gegenbauer moments

2018

View full text Add to dashboard Cite

Comparative Study of Feature Selection for White Blood Cell Differential Counts in Low Resolution Images

Cited by 9 publications

References 44 publications

Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network

Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network

Leukocyte subtypes classification by means of image processing

Galaxies image classification using artificial bee colony based on orthogonal Gegenbauer moments

Contact Info

Product

Resources

About