Fine-Tuned Siamese Network with Modified Enhanced Super-Resolution GAN Plus Based on Low-Quality Chest X-ray Images for COVID-19 Identification

Abstract

Coronavirus disease has rapidly spread globally since early January of 2020. With millions of deaths, it is essential for an automated system to be utilized to aid in the clinical diagnosis and reduce time consumption for image analysis. This article presents a generative adversarial network (GAN)-based deep learning application for precisely regaining high-resolution (HR) CXR images from low-resolution (LR) CXR correspondents for COVID-19 identification. Respectively, using the building blocks of GAN, we introduce a modified enhanced super-resolution generative adversarial network plus (MESRGAN+) to implement a connected nonlinear mapping collected from noise-contaminated low-resolution input images to produce deblurred and denoised HR images. As opposed to the latest trends of network complexity and computational costs, we incorporate an enhanced VGG19 fine-tuned twin network with the wavelet pooling strategy in order to extract distinct features for COVID-19 identification. We demonstrate our proposed model on a publicly available dataset of 11,920 samples of chest X-ray images, with 2980 cases of COVID-19 CXR, healthy, viral and bacterial cases. Our proposed model performs efficiently both on the binary and four-class classification. The proposed method achieves accuracy of 98.8%, precision of 98.6%, sensitivity of 97.5%, specificity of 98.9%, an F1 score of 97.8% and ROC AUC of 98.8% for the multi-class task, while, for the binary class, the model achieves accuracy of 99.7%, precision of 98.9%, sensitivity of 98.7%, specificity of 99.3%, an F1 score of 98.2% and ROC AUC of 99.7%. Our method obtains state-of-the-art (SOTA) performance, according to the experimental results, which is helpful for COVID-19 screening. This new conceptual framework is proposed to play an influential role in addressing the issues facing COVID-19 examination and other diseases.

Keywords: COVID-19; Siamese network; adversarial learning; chest X-ray images; contrastive loss; deep learning; super-resolution.

Figure 1

Scaling of images at different resolutions to a fixed resolution using image scaling adaptive module.

Figure 2

Our proposed modified ESRGAN+ and Siamese convolutional neural network.

Figure 3

We adopted the fundamental structural configuration of ESRGAN+, where feature extraction and most computation is performed on the LR image feature. We redesigned the structure for better optimization and performance by making a few modifications to the generator structure. The transition from SRGAN to MESRGAN+ is equally showcased.

Figure 4

Comparison results of our proposed MESRGAN+ and other selected SOTA models with the same dataset. The PI value is reported on the left and the PSNR is reported on the right.

Figure 5

Contrastive loss function report for binary class and multi-class.

Figure 6

Cross-entropy loss function report for binary class and multi-class.

Figure 7

Accuracy report for our proposed model and selected pre-trained models for binary class and multi-class.

Figure 8

AUC report for our proposed model and selected pre-trained models for binary class and multi-class.

Figure 9

Sensitivity report for our proposed model and selected pre-trained models for binary class and multi-class.

Figure 10

Our proposed MESRGAN+ and Siamese Capsule Network (Siamese-CapsNet).

Figure 11

Performance accuracy in comparison with our proposed model and other pre-trained models and selected state-of-the-art COVID-19 models for binary class.

Figure 12

Performance accuracy in comparison with our proposed model and other pre-trained models and selected state-of-the-art COVID-19 models for multi-class.

Figure 13

Performance ROC in comparison with our proposed model and other pre-trained models and selected state-of-the-art COVID-19 models for binary class.

Figure 14

Performance ROC in comparison with our proposed model and other pre-trained models and selected state-of-the-art COVID-19 models for multi-class.