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...

Full description

Saved in:

Bibliographic Details
Published in:	Pathogens (Basel) Vol. 12; no. 1; p. 17
Main Authors:	Al-Shourbaji, Ibrahim, Kachare, Pramod H, Abualigah, Laith, Abdelhag, Mohammed E, Elnaim, Bushra, Anter, Ahmed M, Gandomi, Amir H
Format:	Journal Article
Language:	English
Published:	Switzerland MDPI AG 22-12-2022 MDPI
Subjects:	Artificial neural networks batch normalized convolutional neural network (BNCNN) Chest chest X-ray Classification Comparative analysis Complexity Computer applications Coronaviruses COVID-19 Datasets deep learning Disease transmission Feature extraction Feature maps Image analysis Image processing Learning Machine learning Mathematical models Medical research Neural networks Parameters Pneumonia X-rays COVID-19 batch normalized convolutional neural network (BNCNN) deep learning chest X-ray classification
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	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.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	2076-0817 2076-0817
DOI:	10.3390/pathogens12010017