A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images
Pre-trained machine learning models have recently been widely used to detect COVID-19 automatically from X-ray images. Although these models can selectively retrain their layers for the desired task, the output remains biased due to the massive number of pre-trained weights and parameters. This pape...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Pathogens |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-0817/12/1/17 |
| _version_ | 1827622716567453696 |
|---|---|
| author | Ibrahim Al-Shourbaji Pramod H. Kachare Laith Abualigah Mohammed E. Abdelhag Bushra Elnaim Ahmed M. Anter Amir H. Gandomi |
| author_facet | Ibrahim Al-Shourbaji Pramod H. Kachare Laith Abualigah Mohammed E. Abdelhag Bushra Elnaim Ahmed M. Anter Amir H. Gandomi |
| author_sort | Ibrahim Al-Shourbaji |
| collection | DOAJ |
| description | Pre-trained machine learning models have recently been widely used to detect COVID-19 automatically from X-ray images. Although these models can selectively retrain their layers for the desired task, the output remains biased due to the massive number of pre-trained weights and parameters. This paper proposes a novel batch normalized convolutional neural network (BNCNN) model to identify COVID-19 cases from chest X-ray images in binary and multi-class frameworks with a dual aim to extract salient features that improve model performance over pre-trained image analysis networks while reducing computational complexity. The BNCNN model has three phases: Data pre-processing to normalize and resize X-ray images, Feature extraction to generate feature maps, and Classification to predict labels based on the feature maps. Feature extraction uses four repetitions of a block comprising a convolution layer to learn suitable kernel weights for the features map, a batch normalization layer to solve the internal covariance shift of feature maps, and a max-pooling layer to find the highest-level patterns by increasing the convolution span. The classifier section uses two repetitions of a block comprising a dense layer to learn complex feature maps, a batch normalization layer to standardize internal feature maps, and a dropout layer to avoid overfitting while aiding the model generalization. Comparative analysis shows that when applied to an open-access dataset, the proposed BNCNN model performs better than four other comparative pre-trained models for three-way and two-way class datasets. Moreover, the BNCNN requires fewer parameters than the pre-trained models, suggesting better deployment suitability on low-resource devices. |
| first_indexed | 2024-03-09T11:30:35Z |
| format | Article |
| id | doaj.art-77ccde86515240388b0c3eb195baf7cb |
| institution | Directory Open Access Journal |
| issn | 2076-0817 |
| language | English |
| last_indexed | 2024-03-09T11:30:35Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Pathogens |
| spelling | doaj.art-77ccde86515240388b0c3eb195baf7cb2023-11-30T23:52:36ZengMDPI AGPathogens2076-08172022-12-011211710.3390/pathogens12010017A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray ImagesIbrahim Al-Shourbaji0Pramod H. Kachare1Laith Abualigah2Mohammed E. Abdelhag3Bushra Elnaim4Ahmed M. Anter5Amir H. Gandomi6Department of Computer Science, University of Hertfordshire, Hatfield AL10 9AB, UKDepartment of Electronics & Telecommunication Engineering, Ramrao Adik Institute of Technology, Nerul, Navi Mumbai 400706, Maharashtra, IndiaComputer Science Department, Prince Hussein Bin Abdullah Faculty for Information Technology, Al al-Bayt University, Mafraq 25113, JordanDepartment of Information Technology and Security, Jazan University, Jazan 45142, Saudi ArabiaDepartment of Computer Science, College of Science and Humanities in Al-Sulail, Prince Sattam bin Abdulaziz University, Riyadh 11671, Saudi ArabiaEgypt-Japan University of Science and Technology (E-JUST), Alexandria 21934, EgyptFaculty of Engineering and IT, University of Technology Sydney, Ultimo, NSW 2007, AustraliaPre-trained machine learning models have recently been widely used to detect COVID-19 automatically from X-ray images. Although these models can selectively retrain their layers for the desired task, the output remains biased due to the massive number of pre-trained weights and parameters. This paper proposes a novel batch normalized convolutional neural network (BNCNN) model to identify COVID-19 cases from chest X-ray images in binary and multi-class frameworks with a dual aim to extract salient features that improve model performance over pre-trained image analysis networks while reducing computational complexity. The BNCNN model has three phases: Data pre-processing to normalize and resize X-ray images, Feature extraction to generate feature maps, and Classification to predict labels based on the feature maps. Feature extraction uses four repetitions of a block comprising a convolution layer to learn suitable kernel weights for the features map, a batch normalization layer to solve the internal covariance shift of feature maps, and a max-pooling layer to find the highest-level patterns by increasing the convolution span. The classifier section uses two repetitions of a block comprising a dense layer to learn complex feature maps, a batch normalization layer to standardize internal feature maps, and a dropout layer to avoid overfitting while aiding the model generalization. Comparative analysis shows that when applied to an open-access dataset, the proposed BNCNN model performs better than four other comparative pre-trained models for three-way and two-way class datasets. Moreover, the BNCNN requires fewer parameters than the pre-trained models, suggesting better deployment suitability on low-resource devices.https://www.mdpi.com/2076-0817/12/1/17chest X-rayCOVID-19deep learningbatch normalized convolutional neural network (BNCNN)classification |
| spellingShingle | Ibrahim Al-Shourbaji Pramod H. Kachare Laith Abualigah Mohammed E. Abdelhag Bushra Elnaim Ahmed M. Anter Amir H. Gandomi A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images Pathogens chest X-ray COVID-19 deep learning batch normalized convolutional neural network (BNCNN) classification |
| title | A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images |
| title_full | A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images |
| title_fullStr | A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images |
| title_full_unstemmed | A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images |
| title_short | A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images |
| title_sort | deep batch normalized convolution approach for improving covid 19 detection from chest x ray images |
| topic | chest X-ray COVID-19 deep learning batch normalized convolutional neural network (BNCNN) classification |
| url | https://www.mdpi.com/2076-0817/12/1/17 |
| work_keys_str_mv | AT ibrahimalshourbaji adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT pramodhkachare adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT laithabualigah adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT mohammedeabdelhag adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT bushraelnaim adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT ahmedmanter adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT amirhgandomi adeepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT ibrahimalshourbaji deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT pramodhkachare deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT laithabualigah deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT mohammedeabdelhag deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT bushraelnaim deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT ahmedmanter deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages AT amirhgandomi deepbatchnormalizedconvolutionapproachforimprovingcovid19detectionfromchestxrayimages |