Feature Extraction using CNN for Peripheral Blood Cells Recognition

Ammar, Mohammed; Daho, Mostafa El Habib; Harrar, Khaled; Laidi, Amel

doi:10.4108/eai.20-10-2021.171548

Cited by 4 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, a classification layer computes weighted classification tasks, employing mutually exclusive tasks and cross-entropy loss for classification. The network is trained using stochastic gradient descent with momentum (sgdm) [42]. By incorporating a momentum term, sgdm overcomes the oscillation issues of standard stochastic gradient descent.…”

Section: Methods B-training and Testingmentioning

confidence: 99%

EMU-Net: Automatic Brain Tumor Segmentation and Classification Using Efficient Modified U-Net

Aly,

Alotaibi

2023

CMC

View full text Add to dashboard Cite

Tumor segmentation is a valuable tool for gaining insights into tumors and improving treatment outcomes. Manual segmentation is crucial but time-consuming. Deep learning methods have emerged as key players in automating brain tumor segmentation. In this paper, we propose an efficient modified U-Net architecture, called EMU-Net, which is applied to the BraTS 2020 dataset. Our approach is organized into two distinct phases: classification and segmentation. In this study, our proposed approach encompasses the utilization of the gray-level co-occurrence matrix (GLCM) as the feature extraction algorithm, convolutional neural networks (CNNs) as the classification algorithm, and the chi-square method for feature selection. Through simulation results, the chi-square method for feature selection successfully identifies and selects four GLCM features. By utilizing the modified U-Net architecture, we achieve precise segmentation of tumor images into three distinct regions: the whole tumor (WT), tumor core (TC), and enhanced tumor (ET). The proposed method consists of two important elements: an encoder component responsible for down-sampling and a decoder component responsible for up-sampling. These components are based on a modified U-Net architecture and are connected by a bridge section. Our proposed CNN architecture achieves superior classification accuracy compared to existing methods, reaching up to 99.65%. Additionally, our suggested technique yields impressive Dice scores of 0.8927, 0.9405, and 0.8487 for the tumor core, whole tumor, and enhanced tumor, respectively. Ultimately, the method presented demonstrates a higher level of trustworthiness and accuracy compared to existing methods. The promising accuracy of the EMU-Net study encourages further testing and evaluation in terms of extrapolation and generalization.

show abstract

Section: Methods B-training and Testingmentioning

confidence: 99%

EMU-Net: Automatic Brain Tumor Segmentation and Classification Using Efficient Modified U-Net

Aly,

Alotaibi

2023

CMC

View full text Add to dashboard Cite

show abstract

“…The proposed DLBCNet is compared to other state-of-the-art methods on the same public dataset, including CNN-AdaboostM1 [18] and the Xception-LSTM [20]. The comparison results of the proposed DLBCNet with other state-of-the-art methods are provided in Table 10.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 99%

“…The Con-SVM model could achieve 85.96% accuracy. Ammar et al [18] applied seven different combinations of CNN models with other traditional classifiers to classify blood cells, including KNN, SVM, and AdaboostM1. Finally, the CNN-AdaboostM1 yielded 88% accuracy.…”

Section: Introductionmentioning

confidence: 99%

DLBCNet: A Deep Learning Network for Classifying Blood Cells

Zhu

Ren

et al. 2023

BDCC

View full text Add to dashboard Cite

Background: Blood is responsible for delivering nutrients to various organs, which store important health information about the human body. Therefore, the diagnosis of blood can indirectly help doctors judge a person’s physical state. Recently, researchers have applied deep learning (DL) to the automatic analysis of blood cells. However, there are still some deficiencies in these models. Methods: To cope with these issues, we propose a novel network for the multi-classification of blood cells, which is called DLBCNet. A new specifical model for blood cells (BCGAN) is designed to generate synthetic images. The pre-trained ResNet50 is implemented as the backbone model, which serves as the feature extractor. The extracted features are fed to the proposed ETRN to improve the multi-classification performance of blood cells. Results: The average accuracy, average sensitivity, average precision, average specificity, and average f1-score of the proposed model are 95.05%, 93.25%, 97.75%, 93.72%, and 95.38%, accordingly. Conclusions: The performance of the proposed model surpasses other state-of-the-art methods in reported classification results.

show abstract

“…Acevedo et al recently recommended that two pre-trained models, VGG16 and Inception-V3, be used to categorize normal blood cell pictures, with model VGG16 claiming the highest accuracy of 96.20% (Acevedo et al, 2019). In another work (Ammar et al, 2021), a CNN-based model is designed with four convolutional layers and three fully connected layers. The Adaboost algorithm was used as a feature extractor, and classification was done using machine learning strategies such as k-nearest neighbours, support vector machine with linear kernel, and support vector machine with RBF kernel function, with the Adaboost algorithm achieving the highest accuracy of 88.8%.…”

mentioning

confidence: 99%

Microcell‐Net: A deep neural network for multi‐class classification of microscopic blood cell images

Dwivedi

Dutta

2023

Expert Systems

View full text Add to dashboard Cite

Blood‐related diseases are one of the major concerns in the biomedical domain and most of the disease symptoms are reflected through the analysis of blood cells. The diagnosis by experts in laboratories is very costly and time‐consuming, thus artificial intelligence‐based systems can help in the automatic diagnosis and monitoring of an individual's health. In this study, a CNN‐based architecture Microcell‐Net is proposed which is trained on a microscopic image dataset of peripheral blood cells in eight different classes. The images have several inter‐class and intra‐class diversity with different magnification levels and the noise present in the images makes the classification task significantly challenging. Experimental results indicated that the proposed model can efficiently classify various types of microscopic blood cells with good accuracy. The experimental findings accomplished 98.76% validation accuracy and 97.65% test accuracy in complex background conditions. The performance of the model is compared with other state‐of‐the‐art models and the proposed deep neural network performs significantly better than others. The proposed model can be utilized in a real‐time diagnosis system because it is fast, automatic and efficient, which can assist in taking clinical decisions and early diagnosis of haematological disorders.

show abstract

Feature Extraction using CNN for Peripheral Blood Cells Recognition

Cited by 4 publications

References 13 publications

EMU-Net: Automatic Brain Tumor Segmentation and Classification Using Efficient Modified U-Net

EMU-Net: Automatic Brain Tumor Segmentation and Classification Using Efficient Modified U-Net

DLBCNet: A Deep Learning Network for Classifying Blood Cells

Microcell‐Net: A deep neural network for multi‐class classification of microscopic blood cell images

Contact Info

Product

Resources

About