A Deep Batch Normalized Convolution Approach for Improving COVID-19 Detection from Chest X-ray Images

Abstract

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.

Impact of the COVID-19 pandemic and COVID vaccination campaign on imaging case volumes and medicolegal aspects

“…The reduction of imaging case volume was justified by the absolute need to protect healthcare personnel and patients from infectious risk. For avoiding this issues technology and science plaied a significant part in particular innovative tools as teleradiology and artificial intelligence (AI) were implemented and for a better care for affected patients all over the world ( 20 , 21 ). AI is made to act and think like a human brain, automating many tasks by imitating its thought processes.…”

A Real Time Method for Distinguishing COVID-19 Utilizing 2D-CNN and Transfer Learning

“…Al-Waisy et al [43] proposed the COVID-CheXNet system, a hybrid deep-learning architecture that successfully diagnosed COVID-19 patients with an accuracy rate of 99.99%. As for binary and three-class classifications, Al-Shourbaji et al [44] proposed a batch-normalized convolutional neural network (BNCNN) model and used other pre-trained models such as VGG-16, VGG-19, Inception-V3, and ResNet-50 to detect COVID-19 and other lung diseases from the CXR images. Their results showed that the BNCNN model outperformed the pretrained models, with accuracies of 99.27% for the binary class and 96.84% for three-class classifications, respectively.…”

Impact of the COVID-19 pandemic on imaging case volumes in King Abdullah University Hospitals (KAUH)

“…Moreover, elective imaging services were stopped to decrease the virus’s transmission ( 8 ). This decision had a significant impact on hospitals’ level of financial revenue.…”