The COVID-19 pandemic has wreaked havoc on the whole world, taking over half a million lives and capsizing the world economy in unprecedented magnitudes. With the world scampering for a possible vaccine, early detection and containment are the only redress. Existing diagnostic technologies with high accuracy like RT-PCRs are expensive and sophisticated, requiring skilled individuals for specimen collection and screening, resulting in lower outreach. So, methods excluding direct human intervention are much sought after, and artificial intelligence-driven automated diagnosis, especially with radiography images, captured the researchers’ interest. This survey marks a detailed inspection of the deep learning–based automated detection of COVID-19 works done to date, a comparison of the available datasets, methodical challenges like imbalanced datasets and others, along with probable solutions with different preprocessing methods, and scopes of future exploration in this arena. We also benchmarked the performance of 315 deep models in diagnosing COVID-19, normal, and pneumonia from X-ray images of a custom dataset created from four others. The dataset is publicly available at https://github.com/rgbnihal2/COVID-19-X-ray-Dataset . Our results show that DenseNet201 model with Quadratic SVM classifier performs the best (accuracy: 98.16%, sensitivity: 98.93%, specificity: 98.77%) and maintains high accuracies in other similar architectures as well. This proves that even though radiography images might not be conclusive for radiologists, but it is so for deep learning algorithms for detecting COVID-19. We hope this extensive review will provide a comprehensive guideline for researchers in this field.

In the updated version of the paper, more experiments and result are illustrated. We also introduce a new chapter (Image Preprocessing) in Section 7. We also added two new tables (Table 4 and Table 6), one figure (Figure 4) and renewed Table 1, 2, and 7 and Figure 5. We also edited many things throughout the paper: rewriting Abstract, Data Description, and Conclusion; Adding more datasets description in Section 5; updating method and analysis of result. The COVID-19 pandemic has wreaked havoc on the whole world, taking over half a million lives and capsizing the world economy in unprecedented magnitudes. With the world scampering for a possible vaccine, early detection and containment is the only redress. Existing diagnostic technologies with high accuracy like RT-PCRs are expensive and sophisticated, requiring skilled individuals for specimen collection and screening, resulting in lower outreach. So, methods excluding direct human intervention are much sought after, and artificial intelligence-driven automated diagnosis, especially with radiography images, captured the researchers' interest. This survey marks a detailed inspection of the deep-learning-based automated detection of COVID-19 works done to date, a comparison of the available datasets, methodical challenges like imbalanced datasets and others, along with probable solutions with different preprocessing methods, and scopes of future exploration in this arena. We also benchmarked the performance of 315 deep models in diagnosing COVID-19, Normal, and Pneumonia from x-ray images of a custom dataset created from four others. The dataset is publicly available at https://github.com/rgbnihal2/COVID-19-X-ray-Dataset. Our results show that DenseNet201 model with Quadratic SVM classifier performs the best (accuracy: 98.16%, sensitivity: 98.93%, specificity: 98.77%) and maintains high accuracies in other similar architectures as well. This proves that even though radiography images might not be conclusive for radiologists, but it is so for deep learning algorithms for detecting COVID-19. We hope this extensive review will provide a comprehensive guideline for researchers in this field.

Sudden prevalence of coronavirus disease-2019 (COVID-19) has badly impeded and collapsed the emerging global trend of economic development by its ongoing pandemic. The novel coronavirus also named severe acute respiratory syndrome virus – 2 (SARS-CoV-2) causes the disease COVID-19 that influences the health management of humans and the world commerce badly. It has affected the human social lives and education in underdeveloped countries and severely impeded industries, organizations, agriculture, etc. Three perceptible types of SARS-CoV-2 strains have been discovered. Each of them has specific receptors, and some of them are common in SARS and SARS-CoV-2. Among them, the ACE2 receptor is believed to be the central receptor of human infectious coronaviruses. It supports mainly to get access, enter into the cell, and causes the basic infection. Similarly, TMPRSS2 is also acting as a portal for a virus to get an approach to the cell and does not support metabolic processes like replication virus. ADAM17, which is a member of disintegrins and metalloproteases and is responsible for cell to cell and cell-array interconnections. These receptors can be important for prevention, vaccine development, and therapies. Several therapies in SARS-CoV-2 infected patients have been tried and suggested. Plasma and stem cell therapy reduce the severity of infection at certain levels in individual patients. In this review, we make an effort to cover all of the said aspects of COVID-19 in a very compressive and brief way. Finally, we shed light on vaccination and therapeutic approaches like plasma therapy and stem therapy and their future perspective with the whole discussion conclusion.

The ravage of COVID-19 is not merely limited to taking its toll with half a million fatalities. It has halted the world economy, disrupting normalcy of lives with supervening severity than any other global catastrophe of the last few decades. The majority of the vaccine discovery attempts are still on trial, making early detection and containment the only feasible redress. The existing diagnostic technique with high accuracy has the setbacks of being expensive and sophisticated, requiring skilled individuals for specimen collection and screening resulting in lower outreach. So, methods excluding direct human intervention are much sought after, and artificial intelligence-driven automated diagnosis, especially with radiography images, captures the researchers' interest. This survey marks a detailed inspection of the deep-learning-based automated detection of COVID-19 works done to date, methodical challenges along with probable solutions, and scopes of future exploration in this arena. We also provided a comparative quantitative analysis of the performance of 315 deep models in diagnosing COVID-19, Normal, and Pneumonia from x-ray images. Our results show that Densenet201 model with Quadratic SVM classifier performs the best (accuracy: 98.16\%, sensitivity: 98.93\%, specificity: 98.77\%) and maintains high accuracies in other similar architectures as well. This proves that even though radiography images might not be conclusive for radiologists, but it is so for deep learning algorithms for detecting COVID-19. We hope this extensive review will provide a comprehensive guideline for researchers in this field.

No abstract