COVID-19 prediction through X-ray images using transfer learning-based hybrid deep learning approach

Abstract

Over the past few months, the campaign against COVID-19 has developed into one of the world's most sought anti-toxin treatment scheme. It is fundamental to distinguish cases of COVID-19 precisely and quickly to help avoid this pandemic from taking a wrong turn with a proper medical reasoning and solution. While Reverse-Transcription Polymerase Chain Reaction (RT-PCR) has been useful in detection of corona virus, chest X-Ray techniques has proven to be more successful and beneficial at detection of the effects of virus. With the increase in COVID patients and the X-Rays done, it is currently possible to classify the X-Ray reports with transfer learning. This paper presents a novel approach, i.e., Hybrid Convolutional Neural Network (HDCNN), which integrates Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) architecture for the finding of COVID-19 using the chest X-Ray. The transfer learning approach, namely slope weighted activation class planning (Grad-CAMs), is used with HDCNN to display images responsible for taking decisions. In this study, HDCNN is compared with other CNNs such as Inception-v3, ShuffleNet, SqueezeNet, VGG-19 and DenseNet. As a result, HDCNN has achieved an accuracy of 98.20%, precision of 97.31%, recall of 97.1% and F1 score of 0.97. Compared to other current deep learning models, the HDCNN has achieved better results, and this can be used for diagnosis purpose after proper approvals.

Stacking Ensemble and ECA-EfficientNetV2 Convolutional Neural Networks on Classification of Multiple Chest Diseases Including COVID-19

“… CNN+ ELMs+ ChOA 2 99.11% 5 Thaseen et al (2022) ( 9 ) 13,808 X-ray images. Ensemble CNN model 3 99.00% 6 Kumar et al (2021) ( 10 ) 6000 X-ray images. HCNN 3 98.20% 7 Musallam et al (2022) ( 11 ) 7512 X-ray images.…”

“…The results showed that the three-category accuracy of COVID-19, general pneumonia and normal was 99.00%. A hybrid convolutional neural network (HCNN) combining CNN and RNN on classification of three-category accuracy of COVID-19, general pneumonia and normal was proposed with the classification accuracy of 98.20% ( 10 ).…”

A novel COVID diagnosis and feature extraction based on discrete wavelet model and classification using X-ray and CT images

“…Kumar et al [ 30 ] proposed the hybrid convolutional neural network (HDCNN), which fuses convolutional neural network (CNN) and recurrent neural network (RNN) structure for detecting COVID-19 cases by utilizing chest x-ray images. With the increase in COVID-19 cases and X-rays, it is necessary to classify the x-ray with the help of transfer learning.…”

“… For efficient net Accuracy- 93.48% Precision value for COVID and normal class – 93% and 94% Recall value for COVID and normal class- 93% and 93% F1-score value for COVID and normal class- 93% and 94% Chen [ 6 ] CNN + HOG classifier A combination of 1 × 1 convolution blocks makes the structural design of CNN lighter with a minimum number of parameters. The overfitting issues is not handled properly by the drop-out layer • Accuracy- 92.95% • Recall- 85% • Specificity- 82% • Precision- 91.5% Demir [ 9 ] DeepCoroNet Deals with big datasets For real-world applications, it requires high memory bandwidth – Kumar et al [ 30 ] Hybrid CNN By using HCNN, this system exhibits very promising outcome This model built over very minimum samples. • Accuracy- 98.20% • Precision- 97.31% • Recall- 97.1% • F1-score-0.97 Elaziz et al, [ 1 ] MobileNetV3 and Aquila optimizer algorithm Early diagnosis was performed efficiently It is unsuitable for agriculture, remote sensing, galaxy classification, and other image classification tasks.…”

“…It leads to cost complexity and needs lots of processing time to extract the features. In [ 30 ], an HDCNN is proposed that includes CNN and RNN for detecting COVID-19 cases. The classification stage classifies the input data into 700 steps and 64 for each step, which may cause learning complexity.…”