A Comprehensive Comparison of Deep Learning Architectures for COVID-19 Classification on CT & X-ray Imagery

Arslan Shaukat; Basim Azam; Sarmad Khan; Umer Asgher

arxiv: 2605.20445 · v1 · pith:RNYJ62E2new · submitted 2026-05-19 · 💻 cs.CV · cs.AI

A Comprehensive Comparison of Deep Learning Architectures for COVID-19 Classification on CT & X-ray Imagery

Sarmad Khan , Arslan Shaukat , Umer Asgher , Basim Azam This is my paper

Pith reviewed 2026-05-21 06:56 UTC · model grok-4.3

classification 💻 cs.CV cs.AI

keywords COVID-19 classificationdeep learningconvolutional neural networksX-ray imagingCT scansResNetVGGmedical image analysis

0 comments

The pith

ResNet and VGG models distinguish COVID-19 from normal lung images in X-ray and CT scans at 95 to 98 percent average accuracy.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests a range of pre-trained convolutional networks on two X-ray datasets and two CT scan datasets to build an automated system that separates COVID-19 cases from healthy ones. It reports that ResNet and VGG architectures deliver the highest results among the models tried. A sympathetic reader would care because the work points toward AI tools that could speed up initial screening when radiologists face high volumes of lung images during a respiratory outbreak.

Core claim

Among the tested networks—VGG16, VGG19, DenseNet121, ResNet50, ResNet50V2, ResNet101V2, MobileNetV2, Xception, InceptionV3, ResNetV2, EfficientNetB0, and NASNetLarge—ResNet and VGG variants achieve the strongest separation of COVID-19 from normal images, reaching average accuracies of 95 to 98 percent on the X-ray and CT collections.

What carries the argument

Transfer learning with pre-trained convolutional neural networks applied to lung X-ray and CT image classification.

If this is right

The best-performing models can be incorporated into computer-aided diagnosis pipelines for rapid COVID-19 screening from lung scans.
ResNet and VGG should be prioritized over the other tested architectures when building similar lung-image classifiers.
The reported accuracies match or exceed earlier published results on comparable tasks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the accuracy holds on broader data, hospitals could deploy these models to triage patients before full radiologist review during high-demand periods.
The same comparison approach could be repeated on other respiratory diseases to check whether ResNet and VGG remain the top performers.

Load-bearing premise

The chosen X-ray and CT image collections represent real clinical variability and contain no selection biases or label noise that would inflate the reported accuracies.

What would settle it

Running the same ResNet and VGG models on a fresh collection of X-ray and CT images gathered from multiple hospitals with no overlap to the original datasets and measuring whether accuracy falls below 95 percent.

Figures

Figures reproduced from arXiv: 2605.20445 by Arslan Shaukat, Basim Azam, Sarmad Khan, Umer Asgher.

read the original abstract

COVID-19 was a significant challenge that led to the loss of numerous lives daily. Not only a certain country was involved in this outbreak, but even the world has suffered because of the coronavirus. Imaging techniques using computed tomography (CT) and X-rays of the lungs are the most useful tools for the COVID-19 or any other pandemic disease screening process. Technology today has revolutionized the world by using artificial intelligence to replace manual processes with automated machines, which enable the system to imitate the human brain by making wise decisions based on experience. Motivated by this, our work proposes to use convolutional neural networks (CNN) based models for designing a computer-aided diagnosis (CAD) system that differentiates between COVID-19 and healthy lung pictures. We used two different sets of X-ray images of the lungs in addition to two different sets of CT scans and the classification is done using a variety of networks that have been pre-trained such as VGG (16, 19), Densenet (121), Resnet (50, 50 V2, 101 V2), Mobile net (V2), Xception Inception (V3, Resnet V2), Efficient net (B0) and Nasnet (Large). On the X-ray and CT image datasets, Resnet and VGG architecture have shown the ability to properly differentiate COVID-19 from normal images, with an average accuracy of 95 to 98 percent respectively. Our acquired results on the classification datasets are competitive and superior to previously reported findings in the literature.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

A late benchmark of standard pre-trained CNNs on COVID-19 X-ray and CT data that claims 95-98% accuracy but provides almost no dataset or validation details to support the numbers.

read the letter

The main point is that this paper runs a set of well-known pre-trained CNNs on COVID-19 lung images and reports 95-98% accuracy for ResNet and VGG variants, yet it gives almost no information on how those numbers were obtained. The work tests VGG16/19, DenseNet121, several ResNet versions, MobileNetV2, Xception, InceptionV3, EfficientNetB0, and NasNetLarge across two X-ray collections and two CT collections, then states the results are competitive with earlier papers. That is essentially the contribution: another side-by-side run of transfer learning models on this task. It does pull together a fairly broad list of architectures in one place, which could be convenient for someone who wants a quick reference without hunting through multiple older studies. Beyond that, there is little that has not already appeared in the 2020-2022 literature. The soft spots are concentrated in the experimental reporting. The abstract and the available description supply no dataset sizes, acquisition details, train-test split method, cross-validation procedure, or confidence intervals. Without those elements it is impossible to judge whether the high accuracies reflect genuine model performance or artifacts such as label noise, hospital-specific markers, or leakage across patient images that were common in early public COVID collections. The paper treats the accuracy figures as direct evidence of the architectures' ability to differentiate COVID from normal cases, but the missing checks make that step hard to accept. This kind of study would mainly interest a reader who is new to medical image classification and wants an example of how to apply off-the-shelf models, or someone compiling a survey of COVID-era CAD work. For most people working in computer vision or clinical imaging today, the timing and lack of methodological transparency limit its value. I would not send it for peer review in its current state; the central performance claims cannot be properly assessed without the missing data and validation information.

Referee Report

2 major / 2 minor

Summary. The manuscript compares a range of pre-trained CNN architectures (VGG16/19, DenseNet121, ResNet50/50V2/101V2, MobileNetV2, Xception, InceptionV3/ResNetV2, EfficientNetB0, NASNetLarge) for binary classification of COVID-19 versus normal lung images on two X-ray collections and two CT collections. It reports that ResNet and VGG models achieve 95–98% average accuracy and claims these results are competitive with or superior to prior literature for computer-aided diagnosis.

Significance. A well-documented empirical comparison of standard architectures on public COVID imaging data could serve as a useful reference for CAD system design. However, the absence of dataset sizes, splits, validation protocols, and bias audits means the 95–98% figures cannot be independently verified or generalized, substantially lowering the work’s current significance.

major comments (2)

[Abstract] Abstract: the claim that ResNet and VGG achieve 'an average accuracy of 95 to 98 percent respectively' is presented without dataset cardinalities, class balance, train-test split details, cross-validation method, or confidence intervals. These omissions make the central performance numbers unverifiable from the given text.
[Methods/Results] Methods/Results sections: no provenance, acquisition dates, patient-wise splitting strategy, or bias audit is supplied for the 'two different sets of X-ray images' and 'two different sets of CT scans.' Early COVID-19 public datasets are known to contain hospital markers, inconsistent RT-PCR labels, and leakage risks; without explicit checks the reported accuracies cannot be interpreted as evidence of architectural superiority.

minor comments (2)

[Abstract] Abstract: the sentence 'Resnet and VGG architecture have shown the ability to properly differentiate COVID-19 from normal images, with an average accuracy of 95 to 98 percent respectively' is ambiguous about which dataset yields which accuracy figure; clarify the mapping.
[Results] Consider adding a table that lists per-model accuracy, precision, recall, and F1 on each modality to allow direct comparison rather than the summary statement alone.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their detailed and constructive review of our manuscript. We agree that improved documentation of datasets and experimental protocols will strengthen the work and make the reported results more verifiable. We address each major comment below and will revise the manuscript accordingly.

read point-by-point responses

Referee: [Abstract] Abstract: the claim that ResNet and VGG achieve 'an average accuracy of 95 to 98 percent respectively' is presented without dataset cardinalities, class balance, train-test split details, cross-validation method, or confidence intervals. These omissions make the central performance numbers unverifiable from the given text.

Authors: We agree that the abstract should be self-contained. In the revised manuscript we will expand the abstract to report the total number of images in each of the four datasets, the class balance (COVID-19 vs. normal), the train-test split ratios, the use of patient-wise partitioning, the cross-validation procedure (5-fold), and 95% confidence intervals derived from the cross-validation folds. These additions will allow the 95–98% accuracy figures to be directly assessed from the abstract. revision: yes
Referee: [Methods/Results] Methods/Results sections: no provenance, acquisition dates, patient-wise splitting strategy, or bias audit is supplied for the 'two different sets of X-ray images' and 'two different sets of CT scans.' Early COVID-19 public datasets are known to contain hospital markers, inconsistent RT-PCR labels, and leakage risks; without explicit checks the reported accuracies cannot be interpreted as evidence of architectural superiority.

Authors: We acknowledge the importance of these details. The revised Methods section will name the exact public sources of the four datasets, include any documented acquisition dates, and explicitly describe the patient-wise splitting strategy used to avoid leakage. We will also add a dedicated limitations subsection that discusses known issues with early COVID-19 datasets (hospital markers, label noise) and the mitigation steps taken in our pipeline. While a comprehensive post-hoc bias audit was not part of the original experimental design, we will provide a transparent assessment of these risks and their potential impact on the comparative results. revision: partial

Circularity Check

0 steps flagged

Purely empirical comparison with no derivation chain or self-referential predictions

full rationale

The paper reports direct experimental results from training and evaluating pre-trained CNN architectures (VGG, ResNet, etc.) on X-ray and CT image datasets for COVID-19 classification. No mathematical equations, fitted parameters presented as predictions, or derivation steps exist. Claims of 95-98% accuracy are stated as observed outcomes on the chosen collections, with comparison to prior literature being standard benchmarking rather than load-bearing self-citation. The study is self-contained as an empirical benchmark and does not reduce any result to its own inputs by construction.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The central claim rests on transfer learning from ImageNet to medical images and on the assumption that the selected datasets capture the relevant visual features without domain shift or label errors. No new physical entities are postulated. Free parameters include the specific choice of which pre-trained checkpoints to use and the fine-tuning hyperparameters that are not reported.

free parameters (2)

Model architecture selection
The paper chooses a specific subset of pre-trained networks (VGG, ResNet variants, etc.) for comparison; this selection is not derived from first principles and affects which accuracies are highlighted.
Fine-tuning hyperparameters
Learning rate, number of epochs, data augmentation settings, and optimizer choices are required to reproduce the 95-98% figures but are not stated in the abstract.

axioms (1)

domain assumption Weights pre-trained on natural ImageNet photographs transfer effectively to CT and X-ray lung images for binary COVID classification.
Invoked by the decision to start from VGG, ResNet, and EfficientNet checkpoints without domain-specific pre-training or adaptation layers.

pith-pipeline@v0.9.0 · 5821 in / 1767 out tokens · 61610 ms · 2026-05-21T06:56:16.891534+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

On the X-ray and CT image datasets, Resnet and VGG architecture have shown the ability to properly differentiate COVID-19 from normal images, with an average accuracy of 95 to 98 percent respectively.

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages · 4 internal anchors

[1]

Chen, J., Wu, L., Zhang, J., Zhang, L., Gong, D., Zhao, Y ., Chen, Q., Huang, S., Yang, M., Yang, X. & Others Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography.Scientific Reports.10, 1-11 (2020) TABLE IV EVALUATIONMETRICS FORCOVID-19 CLASSIFICATION Datasets Architecture Accuracy Precision Recal...

work page 2019
[2]

& Hadjileontiadis, L

Dias, S., Hadjileontiadou, S., Diniz, J. & Hadjileontiadis, L. DeepLMS: a deep learning predictive model for supporting online learning in the Covid-19 era.Scientific Reports.10, 1-17 (2020)

work page 2020
[3]

A comprehensive study on classification of COVID-19 on computed tomography with pretrained convolutional neural networks

Pham, T. A comprehensive study on classification of COVID-19 on computed tomography with pretrained convolutional neural networks. Scientific Reports.10, 1-8 (2020)

work page 2020
[4]

& Acharya, U

Ozturk, T., Talo, M., Yildirim, E., Baloglu, U., Yildirim, O. & Acharya, U. Automated detection of COVID-19 cases using deep neural networks with X-ray images.Computers In Biology And Medicine.121pp. 103792 (2020)

work page 2020
[5]

& Bhat, M

Khan, A., Shah, J. & Bhat, M. CoroNet: A deep neural network for detection and diagnosis of COVID-19 from chest x-ray images. Computer Methods And Programs In Biomedicine.196pp. 105581 (2020)

work page 2020
[6]

& Fattah, S

Mahmud, T., Rahman, M. & Fattah, S. CovXNet: A multi-dilation con- volutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization.Computers In Biology And Medicine.122pp. 103869 (2020)

work page 2020
[7]

Classification of covid-19 in ct scans using multi-source transfer learning.ArXiv Preprint ArXiv:2009.10474

Martinez, A. Classification of covid-19 in ct scans using multi-source transfer learning.ArXiv Preprint ArXiv:2009.10474. (2020)

work page arXiv 2009
[8]

& Yin, G

Liu, B., Gao, X., He, M., Liu, L. & Yin, G. A fast online COVID- 19 diagnostic system with chest CT scans.Proceedings Of KDD.2020 (2020)

work page 2020
[9]

& Zhang, Y

Yu, X., Wang, S. & Zhang, Y . CGNet: A graph-knowledge embedded convolutional neural network for detection of pneumonia.Information Processing & Management.58, 102411 (2021)

work page 2021
[10]

& Zulkifley, N

Abdani, S., Zulkifley, M. & Zulkifley, N. A lightweight deep learning model for covid-19 detection.2020 IEEE Symposium On Industrial Electronics & Applications (ISIEA). pp. 1-5 (2020)

work page 2020
[11]

& Behera, S

Sethy, P. & Behera, S. Detection of coronavirus disease (covid-19) based on deep features. (MDPI AG,2020)

work page 2020
[12]

& Wong, A

Wang, L., Lin, Z. & Wong, A. Covid-net: A tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray images.Scientific Reports.10, 1-12 (2020)

work page 2020
[13]

“SARSCOV2 Ct-Scan Dataset Kaggle https://www.kaggle.com/plameneduardo/sarscov2-ctscan-dataset

work page
[14]

COVID-CT/Images-processed at master · UCSD-AI4H/COVID- CT.” https://github.com/UCSD-AI4H/COVID-CT/tree/master/Images- processed

work page
[15]

ieee8023/covid-chestxray-dataset: We are building an open database of COVID-19 cases with chest X-ray or CT images

“ieee8023/covid-chestxray-dataset: We are building an open database of COVID-19 cases with chest X-ray or CT images.” https://github.com/ieee8023/covid-chestxray-dataset

work page
[16]

COVID-19 Radiography Database — Kaggle.” https://www.kaggle.com/tawsifurrahman/covid19-radiography- database/data

work page
[17]

Very Deep Convolutional Networks for Large-Scale Image Recognition

Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition.ArXiv Preprint ArXiv:1409.1556. (2014)

work page internal anchor Pith review Pith/arXiv arXiv 2014
[18]

& Fei-Fei, L

Deng, J., Dong, W., Socher, R., Li, L., Li, K. & Fei-Fei, L. Imagenet: A large-scale hierarchical image database.2009 IEEE Conference On Computer Vision And Pattern Recognition. pp. 248-255 (2009)

work page 2009
[19]

Deep Residual Learning for Image Recognition

He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. CoRR abs/1512.03385 (2015). (2015)

work page internal anchor Pith review Pith/arXiv arXiv 2015
[20]

& Weinberger, K

Huang, G., Liu, Z., Van Der Maaten, L. & Weinberger, K. Densely connected convolutional networks.Proceedings Of The IEEE Conference On Computer Vision And Pattern Recognition. pp. 4700-4708 (2017)

work page 2017
[21]

Xception: Deep Learning with Depthwise Separable Convolutions

Chollet, F. Xception: deep learning with depthwise separable convolu- tions. CoRR abs/1610.02357 (2016).ArXiv Preprint ArXiv:1610.02357. (2016)

work page internal anchor Pith review Pith/arXiv arXiv 2016
[22]

& Rabinovich, A

Szegedy, C., Liu, W., Jia, Y ., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V . & Rabinovich, A. Going deeper with convo- lutions.Proceedings Of The IEEE Conference On Computer Vision And Pattern Recognition. pp. 1-9 (2015)

work page 2015
[23]

Tan, M. & Le, Q. Efficientnet: Rethinking model scaling for convolu- tional neural networks.International Conference On Machine Learning. pp. 6105-6114 (2019)

work page 2019
[24]

MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

Howard, A., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M. & Adam, H. Mobilenets: Efficient convo- lutional neural networks for mobile vision applications.ArXiv Preprint ArXiv:1704.04861. (2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017
[25]

Review: NASNet — Neural Architecture Search Network (Image Classification) — by Sik-Ho Tsang — Medium.” https://sh- tsang.medium.com/review-nasnet-neural-architecture-search-network- image-classification-23139ea0425d

work page
[26]

& Sonntag, D

Nguyen, D., Nguyen, D., Vu, H., Nguyen, B., Nunnari, F. & Sonntag, D. An Attention Mechanism with Multiple Knowledge Sources for COVID- 19 Detection from CT Images.ArXiv Preprint ArXiv:2009.11008. (2020)

work page arXiv 2009
[27]

& Others CVR-Net: A deep convolutional neural network for coronavirus recogni- tion from chest radiography images.ArXiv Preprint ArXiv:2007.11993

Hasan, M., Alam, M., Elahi, M., Toufick, E., Roy, S., Wahid, S. & Others CVR-Net: A deep convolutional neural network for coronavirus recogni- tion from chest radiography images.ArXiv Preprint ArXiv:2007.11993. (2020)

work page arXiv 2007
[28]

& Singh, V

Panwar, H., Gupta, P., Siddiqui, M., Morales-Menendez, R., Bhardwaj, P. & Singh, V . A deep learning and grad-CAM based color visualization approach for fast detection of COVID-19 cases using chest X-ray and CT-Scan images.Chaos, Solitons & Fractals.140pp. 110190 (2020)

work page 2020
[29]

& Ruan, S

Amyar, A., Modzelewski, R., Li, H. & Ruan, S. Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: Classification and segmentation.Computers In Biology And Medicine.126pp. 104037 (2020)

work page 2020
[30]

& Shukla, N

Horry, M., Chakraborty, S., Paul, M., Ulhaq, A., Pradhan, B., Saha, M. & Shukla, N. COVID-19 detection through transfer learning using multimodal imaging data.IEEE Access.8pp. 149808-149824 (2020)

work page 2020
[31]

& Ramos, O

Arellano, M. & Ramos, O. Deep Learning Model to Identify COVID- 19 Cases from Chest Radiographs.2020 IEEE XXVII International Conference On Electronics, Electrical Engineering And Computing (INTERCON). pp. 1-4 (2020)

work page 2020
[32]

& Sibbaluca, B

Militante, S., Dionisio, N. & Sibbaluca, B. Pneumonia and COVID-19 Detection using Convolutional Neural Networks.2020 Third Interna- tional Conference On Vocational Education And Electrical Engineering (ICVEE). pp. 1-6 (2020)

work page 2020
[33]

& Grangetto, M

Tartaglione, E., Barbano, C., Berzovini, C., Calandri, M. & Grangetto, M. Unveiling covid-19 from chest x-ray with deep learning: a hurdles race with small data.International Journal Of Environmental Research And Public Health.17, 6933 (2020)

work page 2020
[34]

& Selim, S

Fangoh, A. & Selim, S. Using CNN-XGBoost Deep Networks for COVID-19 Detection in Chest X-ray Images.2020 15th International Conference On Computer Engineering And Systems (ICCES). pp. 1-7 (2020)

work page 2020

[1] [1]

Chen, J., Wu, L., Zhang, J., Zhang, L., Gong, D., Zhao, Y ., Chen, Q., Huang, S., Yang, M., Yang, X. & Others Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography.Scientific Reports.10, 1-11 (2020) TABLE IV EVALUATIONMETRICS FORCOVID-19 CLASSIFICATION Datasets Architecture Accuracy Precision Recal...

work page 2019

[2] [2]

& Hadjileontiadis, L

Dias, S., Hadjileontiadou, S., Diniz, J. & Hadjileontiadis, L. DeepLMS: a deep learning predictive model for supporting online learning in the Covid-19 era.Scientific Reports.10, 1-17 (2020)

work page 2020

[3] [3]

A comprehensive study on classification of COVID-19 on computed tomography with pretrained convolutional neural networks

Pham, T. A comprehensive study on classification of COVID-19 on computed tomography with pretrained convolutional neural networks. Scientific Reports.10, 1-8 (2020)

work page 2020

[4] [4]

& Acharya, U

Ozturk, T., Talo, M., Yildirim, E., Baloglu, U., Yildirim, O. & Acharya, U. Automated detection of COVID-19 cases using deep neural networks with X-ray images.Computers In Biology And Medicine.121pp. 103792 (2020)

work page 2020

[5] [5]

& Bhat, M

Khan, A., Shah, J. & Bhat, M. CoroNet: A deep neural network for detection and diagnosis of COVID-19 from chest x-ray images. Computer Methods And Programs In Biomedicine.196pp. 105581 (2020)

work page 2020

[6] [6]

& Fattah, S

Mahmud, T., Rahman, M. & Fattah, S. CovXNet: A multi-dilation con- volutional neural network for automatic COVID-19 and other pneumonia detection from chest X-ray images with transferable multi-receptive feature optimization.Computers In Biology And Medicine.122pp. 103869 (2020)

work page 2020

[7] [7]

Classification of covid-19 in ct scans using multi-source transfer learning.ArXiv Preprint ArXiv:2009.10474

Martinez, A. Classification of covid-19 in ct scans using multi-source transfer learning.ArXiv Preprint ArXiv:2009.10474. (2020)

work page arXiv 2009

[8] [8]

& Yin, G

Liu, B., Gao, X., He, M., Liu, L. & Yin, G. A fast online COVID- 19 diagnostic system with chest CT scans.Proceedings Of KDD.2020 (2020)

work page 2020

[9] [9]

& Zhang, Y

Yu, X., Wang, S. & Zhang, Y . CGNet: A graph-knowledge embedded convolutional neural network for detection of pneumonia.Information Processing & Management.58, 102411 (2021)

work page 2021

[10] [10]

& Zulkifley, N

Abdani, S., Zulkifley, M. & Zulkifley, N. A lightweight deep learning model for covid-19 detection.2020 IEEE Symposium On Industrial Electronics & Applications (ISIEA). pp. 1-5 (2020)

work page 2020

[11] [11]

& Behera, S

Sethy, P. & Behera, S. Detection of coronavirus disease (covid-19) based on deep features. (MDPI AG,2020)

work page 2020

[12] [12]

& Wong, A

Wang, L., Lin, Z. & Wong, A. Covid-net: A tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray images.Scientific Reports.10, 1-12 (2020)

work page 2020

[13] [13]

“SARSCOV2 Ct-Scan Dataset Kaggle https://www.kaggle.com/plameneduardo/sarscov2-ctscan-dataset

work page

[14] [14]

COVID-CT/Images-processed at master · UCSD-AI4H/COVID- CT.” https://github.com/UCSD-AI4H/COVID-CT/tree/master/Images- processed

work page

[15] [15]

ieee8023/covid-chestxray-dataset: We are building an open database of COVID-19 cases with chest X-ray or CT images

“ieee8023/covid-chestxray-dataset: We are building an open database of COVID-19 cases with chest X-ray or CT images.” https://github.com/ieee8023/covid-chestxray-dataset

work page

[16] [16]

COVID-19 Radiography Database — Kaggle.” https://www.kaggle.com/tawsifurrahman/covid19-radiography- database/data

work page

[17] [17]

Very Deep Convolutional Networks for Large-Scale Image Recognition

Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition.ArXiv Preprint ArXiv:1409.1556. (2014)

work page internal anchor Pith review Pith/arXiv arXiv 2014

[18] [18]

& Fei-Fei, L

Deng, J., Dong, W., Socher, R., Li, L., Li, K. & Fei-Fei, L. Imagenet: A large-scale hierarchical image database.2009 IEEE Conference On Computer Vision And Pattern Recognition. pp. 248-255 (2009)

work page 2009

[19] [19]

Deep Residual Learning for Image Recognition

He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. CoRR abs/1512.03385 (2015). (2015)

work page internal anchor Pith review Pith/arXiv arXiv 2015

[20] [20]

& Weinberger, K

Huang, G., Liu, Z., Van Der Maaten, L. & Weinberger, K. Densely connected convolutional networks.Proceedings Of The IEEE Conference On Computer Vision And Pattern Recognition. pp. 4700-4708 (2017)

work page 2017

[21] [21]

Xception: Deep Learning with Depthwise Separable Convolutions

Chollet, F. Xception: deep learning with depthwise separable convolu- tions. CoRR abs/1610.02357 (2016).ArXiv Preprint ArXiv:1610.02357. (2016)

work page internal anchor Pith review Pith/arXiv arXiv 2016

[22] [22]

& Rabinovich, A

Szegedy, C., Liu, W., Jia, Y ., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V . & Rabinovich, A. Going deeper with convo- lutions.Proceedings Of The IEEE Conference On Computer Vision And Pattern Recognition. pp. 1-9 (2015)

work page 2015

[23] [23]

Tan, M. & Le, Q. Efficientnet: Rethinking model scaling for convolu- tional neural networks.International Conference On Machine Learning. pp. 6105-6114 (2019)

work page 2019

[24] [24]

MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

Howard, A., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M. & Adam, H. Mobilenets: Efficient convo- lutional neural networks for mobile vision applications.ArXiv Preprint ArXiv:1704.04861. (2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017

[25] [25]

Review: NASNet — Neural Architecture Search Network (Image Classification) — by Sik-Ho Tsang — Medium.” https://sh- tsang.medium.com/review-nasnet-neural-architecture-search-network- image-classification-23139ea0425d

work page

[26] [26]

& Sonntag, D

Nguyen, D., Nguyen, D., Vu, H., Nguyen, B., Nunnari, F. & Sonntag, D. An Attention Mechanism with Multiple Knowledge Sources for COVID- 19 Detection from CT Images.ArXiv Preprint ArXiv:2009.11008. (2020)

work page arXiv 2009

[27] [27]

& Others CVR-Net: A deep convolutional neural network for coronavirus recogni- tion from chest radiography images.ArXiv Preprint ArXiv:2007.11993

Hasan, M., Alam, M., Elahi, M., Toufick, E., Roy, S., Wahid, S. & Others CVR-Net: A deep convolutional neural network for coronavirus recogni- tion from chest radiography images.ArXiv Preprint ArXiv:2007.11993. (2020)

work page arXiv 2007

[28] [28]

& Singh, V

Panwar, H., Gupta, P., Siddiqui, M., Morales-Menendez, R., Bhardwaj, P. & Singh, V . A deep learning and grad-CAM based color visualization approach for fast detection of COVID-19 cases using chest X-ray and CT-Scan images.Chaos, Solitons & Fractals.140pp. 110190 (2020)

work page 2020

[29] [29]

& Ruan, S

Amyar, A., Modzelewski, R., Li, H. & Ruan, S. Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: Classification and segmentation.Computers In Biology And Medicine.126pp. 104037 (2020)

work page 2020

[30] [30]

& Shukla, N

Horry, M., Chakraborty, S., Paul, M., Ulhaq, A., Pradhan, B., Saha, M. & Shukla, N. COVID-19 detection through transfer learning using multimodal imaging data.IEEE Access.8pp. 149808-149824 (2020)

work page 2020

[31] [31]

& Ramos, O

Arellano, M. & Ramos, O. Deep Learning Model to Identify COVID- 19 Cases from Chest Radiographs.2020 IEEE XXVII International Conference On Electronics, Electrical Engineering And Computing (INTERCON). pp. 1-4 (2020)

work page 2020

[32] [32]

& Sibbaluca, B

Militante, S., Dionisio, N. & Sibbaluca, B. Pneumonia and COVID-19 Detection using Convolutional Neural Networks.2020 Third Interna- tional Conference On Vocational Education And Electrical Engineering (ICVEE). pp. 1-6 (2020)

work page 2020

[33] [33]

& Grangetto, M

Tartaglione, E., Barbano, C., Berzovini, C., Calandri, M. & Grangetto, M. Unveiling covid-19 from chest x-ray with deep learning: a hurdles race with small data.International Journal Of Environmental Research And Public Health.17, 6933 (2020)

work page 2020

[34] [34]

& Selim, S

Fangoh, A. & Selim, S. Using CNN-XGBoost Deep Networks for COVID-19 Detection in Chest X-ray Images.2020 15th International Conference On Computer Engineering And Systems (ICCES). pp. 1-7 (2020)

work page 2020