key: cord-0742516-7c69lk2p
authors: Kumar, Aayush; Tripathi, Ayush R; Satapathy, Suresh Chandra; Zhang, Yu Dong
title: SARS-Net: COVID-19 Detection from Chest X-Rays by Combining Graph Convolutional Network and Convolutional Neural Network
date: 2021-08-25
journal: Pattern Recognit
DOI: 10.1016/j.patcog.2021.108255
sha: ce906992b69cc4417fd07261c4a6826d23fe4eca
doc_id: 742516
cord_uid: 7c69lk2p

COVID-19 has emerged as one of the deadliest pandemics that has ever crept on humanity. Screening tests are currently the most reliable and accurate steps in detecting severe acute respiratory syndrome coronavirus in a patient, and the most used is RT-PCR testing. Various researchers and early studies implied that visual indicators (abnormalities) in a patient's Chest X-Ray (CXR) or computed tomography (CT) imaging were a valuable characteristic of a COVID-19 patient that can be leveraged to find out virus in a vast population. Motivated by various contributions to open-source community to tackle COVID-19 pandemic, we introduce SARS-Net, a CADx system combining Graph Convolutional Networks and Convolutional Neural Networks for detecting abnormalities in a patient's CXR images for presence of COVID-19 infection in a patient. In this paper, we introduce and evaluate the performance of a custom-made deep learning architecture SARS-Net, to classify and detect the Chest X-ray images for COVID-19 diagnosis. Quantitative analysis shows that the proposed model achieves more accuracy than previously mentioned state-of-the-art methods. It was found that our proposed model achieved an accuracy of 97.60% and a sensitivity of 92.90% on the validation set.

COVID-19, also known as the coronavirus disease 2019, is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), known as 2019 novel coronavirus (2019-nCoV) previously. The cases started spreading worldwide, mainly from Wuhan, China, where first cases were reported in late December 2019 .

The severe acute respiratory syndrome coronavirus 2 has a comparatively more significant risk on humanity due to its high reproductive number than SARS coronavirus (mean R0 for SARS coronavirus 2 is 3.28) (Zhao S et al., 2020) . The mean serial interval was calculated as 3.96 days, considerably shorter than mean serial interval calculated for SARS (8.4 days) or MERS (14.6 days) (Du Z et al., 2020) . Various measures have been taken to suppress its exponential Reproductive Number (R0) and Series Interval. It is crucial to immediately isolate the infected person from uninfected population if there is an absence of specific therapeutic drugs or vaccines for COVID-19.

While screening tests have been very useful in finding infection in a patient, the patient gets quick treatment and is isolated to stop the virus spread. RT-PCR testing has been instrumental during initial stage of virus spread and is considered gold standard. (Bell et al.) However, it has some flaws that might emerge once the most affected countries enter community transmission phase.

As the virus enters community transmission phase, there'll be a need for finding an optimal tradeoff between testing accuracy (performance) and time taken for testing to be done. With many cases emerging every day, it will be harmful to the human body to wait for a few days before screening tests. Therefore, some feasible alternative method is very much needed. One such method is COVID-19 detection through radiography images. Chest radiography imaging includes chest X-ray (CXR) or Computed Tomography (CT) imaging, that is conducted by some radiologists who further look for visual indicators (abnormalities) associated with SARS COVID-19 infection. Chest CT and CXR imaging modalities have been used to diagnose Pneumonia for a long time. (Raoof et al., 2012) They are currently the best method available because of their less complexity and high availability, which aids in the faster diagnosis of a patient. They are the most commonly used imaging modality for diagnosing and screening various chest-related diseases and play a vital role in clinical and epidemiological studies. Recent studies have shed light on Chest CT, demonstrating typical radiographic features of most covid infected patients, including ground-glass opacity, multifocal patchy consolidation, and interstitial changes with a peripheral distribution. As per a study conducted by (Daniel J Bell et al., 2020) chest radiography is less sensitive than chest CT; it is typically the first-line imaging modality used for patients suspected of COVID-19. Of the Chest radiographs of patients with covid requiring hospitalization, 69% (Wong et al., 2019) of total patients had an abnormal chest radiograph when admitted. After admission, approximately 80% had abnormalities at any time during hospitalization. The above findings were most extensive about 10-12 days after symptoms.

Radiography examinations are usually done for any respiratory-related ailments and therefore it is readily available. Any symptomatic/asymptomatic COVID-19 patient will show respiratory ailments as a symptom. Therefore, when radiography examinations are performed over a patient, the radiologist can look for abnormalities in CXR images and detect the presence of the SARS-CoV-2 virus. As the COVID-19 pandemic progresses, there will be more reliance on radiography techniques than RT-PCR testing because of its advantages. The biggest challenge in detecting COVID-19 through radiography is the unavailability of experienced radiologists who can readily spot any abnormality in a patient's CXR image. As Pneumonia and COVID-19 have similar symptoms, a radiologist should correctly interpret the radiographic images as the visual indicators can be very minute. This challenge can be tackled using Computer-aided Diagnostic (CADx) systems, that can help a radiologist spot the visual indicators more efficiently and accurately. The pressure on an individual radiologist or a doctor also increases resulting in human error, which is not the case in machine-aided detection tailored to detect COVID-19 accurately.

CADx tools can lower the clinical burden by facilitating Classification and Interpretation using Machine Learning and AI. Deep Learning (DL) plays a vital role in medical image analysis due to its excellent feature extraction ability. DL models complete tasks by automatically analyzing multimodal medical images. Some examples of the application of DL include the diagnosis of diabetic retinopathy (P. Chudzik et al., 2018) , cancer detection and classification (B. Gecer et al., 2018) , polyp detection during colonoscopy (R. Zhang et al., 2018) , and multi-classification of multi-modality skin lesions (L. Bi et al., 2020) , Many articles provide systematic reviews of machine learning techniques in detecting COVID-19 and the potential of Neural Networks and DL for the task. (Albahri et al., 2020 , De Felice et al., 2020 . (Li, X et al., 2020) developed an on-device COVID-19 screening system that uses CXR images to identify COVID-19 infections. Afshar et al. developed a capsule-based network framework to identify COVID-19 infection in CXR images. (Khobahi et al., 2020) have worked on developing a semisupervised technique for the identification of COVID-19 infection.

Computer Vision refers to using a specific set of algorithms for processing visual data and how a computer might gain information and understanding from this process. Convolutional Neural Network (CNN), a class of DL algorithm, has been proven very useful in image recognition & classification, video analysis, medical image analysis & other vision-related tasks. Generally, due to the limited availability of data (especially in the case of medical data), pre-trained CNN models trained on large datasets like ImageNet are used to capitalize on the knowledge of generic features from the images for a target application. Advanced computer-aided diagnosis schemes are primarily based on state-of-theart methods. (Li, Y et al., 2020) employed learning of relation-aware representation (RAR) along with image-level representation. When evaluating each data point individually, RAR is a method that uses the relationships between data points collected as a whole to improve decisions unbiasedly. Taking inspiration from their work, we used RAR over an entire cohort of X-ray images to assess the associations between them. We determined chest-related diseases in each image unbiasedly. The above association can be effectively modeled if we treat each image as a 'node.' (Fukunaga et al., 2020) proposed a new AI framework, a graph convolutional network (GCN), that can learn RARs of nodes by studying graph structure and node features. This work aims to learn image-level features using a conventional CNN, and relation-aware representation (RAR) features using a Graph Convolutional Network. Through extensive experiments, we find the combination could outperform any network operating alone. This research aims to present a CADx system that uses a combination of CNN and GCN to make more accurate diagnoses for detecting COVID-19 from CXR images.

The rest of paper's organization is described as follows: Section 2 describes the Literature survey required for the study. The proposed Methodology is depicted and described in detail in Section 3.

The experiments, quantitative comparison, and analysis of results are discussed in Section 4. Finally, concluding remarks have been made in Section 5.

Since COVID-19 has become widespread, many literature works have described the application of computer vision and DL in the diagnosis of the disease based on medical images and have achieved promising results. Chest CT and X-ray images have proven successful in detecting acute Pneumonia through various DL methods described in the literature (Nair et al., 2020 , Adam et al., 2020 . (Xu et al., 2020 ) developed a DL-based CNN prediction model to classify COVID-19 Pneumonia and influenza with an accuracy of 86.7%. CT images were used as an imaging modality by (Wang S et al., 2020) , who utilized a modified Inception TL model and obtained an accuracy of 89.5%. (Ni QQ et al., 2020 ) used a DL algorithm to characterize COVID-19 Pneumonia in chest CT images. A methodology based on DeTrac CNNs was presented for the classification of images into COVID-19 and non-COVID-19. The results were promising, with an accuracy of above 90%. (Abbas et al., 2020 ) (Luz et al., 2020) employed TL technique to use a class of CNNs known as Efficient-Nets to detect the pattern of COVID-19 in CXR images, that are famous for their high accuracy and low computation. They also used a hierarchical classifier for the task, making it suitable to embed medical equipment or mobile phones. (Farooq et al., 2020) use the TL strategy to fine-tune a ResNet-50 model in the classification of CXR images for detecting COVID-19 along with bacterial and viral Pneumonia. (Minaee et al., 2020) also used TL to develop Deep-Covid to predict COVID-19 from CXR images. They curated a dataset of 5000 CXR images by board-certified radiologists and used a subset of those images to train four CNNs, namely ResNet-18, ResNet-50, Squeeze-Net, and DenseNet-121. (Ucar et al., 2020) used Squeeze-Net with a Bayesian optimization additive for COVID-19 diagnosis. They report better performance and obtain higher COVID-19 diagnosis accuracy on their proposed network due to fine-tuned hyperparameters and augmented datasets. (Novitasari et al., 2020 ) used a combination of SVM and CNNs to detect COVID-19 from CXR images. (Rajaraman et al., 2020) presented a weakly labeled data augmentation method on COVID-19 CXR images. (Tartaglione et al., 2020) give insights into an extensive set of statistical results obtained from the currently available public datasets, besides providing results on a medium-size COVID CXR dataset. (Khan et al., 2020) propose Coro-Net, a CNN-based model to automate the detection of COVID-19 infection from CXR images. They use TL to train Xception CNN architecture, pretrained on the ImageNet dataset, and employ a 4-fold cross-validation method for evaluation. (Periera et al., 2020) used Inception-V3 to extract features from CXR images for COVID-19 and classifiers such as KNNs, Decision Trees, Random Forests, SVMs for their classification. They reported an F-1 score of 0.89 for the COVID-19 class, using hierarchical analysis for the task of COVID-19 detection. The authors of (Hussain et al., 2020 ) analyzed X-ray images and combined the texture and morphological features to classify them into COVID-19, bacterial Pneumonia, and non COVID-19 viral Pneumonia.

The contributions of this research entail the following points: (1) First, we designed a base network SARS-Net CNN (Net-1), a CNN consisting of convolution layers and Inception blocks.

(2) Next, we added various improvement techniques in the base network, Net-1, namely Coordinate-Convolutions (CC), anti-aliased (AA) CNNs, and rank-based stochastic pooling (RSP) to obtain Net-2, Net-3, and Net-4. The names of Net-2, Net-3, and Net-4 are SARS-Net CC, SARS-Net AA, and SARS-Net RSP, respectively. (3) Last, we developed SARS-Net (Net-5), where we combined Net-1 with our proposed 2-layer graph convolutional network (GCN). We prefer to use the Net-i naming convention for ease of understanding where i = 1 to 5. Further experiments showed that Net-5 gives the best results amongst all proposed five networks. In addition, Net-5 was superior to state-of-the-art approaches. 

The dataset used for training and validation is COVIDx, introduced in the COVID-NET paper . It consists of 13,975 CXR images spreading across 13,870 patient cases. So far, the COVIDx dataset is the most prominent openly available CXR dataset in terms of the number of COVID-19 positive patient cases.

Net-1 Base Network SARS-Net CNN CNN consisting of conv. layers and Inception blocks Net-2 ←Net-1 + CC SARS-Net CC Add coordinate-convolutions to Net-1

Net-3 ←Net-1 + AA SARS-Net AA Add anti-aliased CNNs to Net-1

Net-4 ←Net-1 + RSP SARS-Net RSP Use RSP to replace Max-Pooling in Net-1

Net-5 ←Net-1 + GCN SARS-Net Add 2-layer GCN to Net-1

The distribution of data samples from the COVIDx dataset used for training and testing is depicted in Table 2 . We use 90% of the data samples for training and validation, and the rest 10% is used for testing. To generate the COVIDx dataset, the CXR images were collected from various publicly available dataset repositories, namely:

a. COVID-19 Image Data Collection [Cohen et al., 2020] b. COVID-19 Chest X-ray Dataset Initiative c. ActualMed COVID-19 Chest X-ray Dataset Initiative d. For generating the final dataset for training and evaluation of SARS-Net, the following type of patient cases were used from each of the repositories: 

(a), (b), (c), (d) (a) (e)

All the variants of SARS-Net architecture were trained on the COVIDx dataset. First, as a preprocessing step, the CXR images were Centre cropped before training to remove the commonly-found embedded textual patient information. Then, various DA methods were tried out to decide the optimal methods which enhance the performance of SARS-Net and prevent overfitting. We tried several combinations of DA strategies and proceeded with the ones that gave the optimal model performance during training and evaluation. Initially, we relied on the augmentation techniques that are effective for CXR imaging modality and subsequently added other DA methods in the pipeline. DA helps us increase the number of samples and introduce variability in the dataset without collecting new samples. Furthermore, to train the architectures, DA was leveraged with the following augmentation types: Elastic, Translation, CutBlur, Rotation, Random Horizontal flip, Zoom, MotionBlur, Intensity shift, and CutNoise. Finally, the input images in each batch were normalized.

Most of the methods used in the literature so far for COVID-19 detection have used transfer learning techniques (Luz et al., Farooq et al., Minaee et al., Periera et al., Khan et al., 2020) . DL models previously trained on ImageNet dataset are initialized and fine-tuned according to requirements. The underlying principle in Transfer Learning (TL) lies in the fact that DL models transfer the weights learned while capturing generic features from ImageNet datasets for aforementioned tasks. The proposed SARS-Net CNN network architecture is shown in Fig. 3 , that employs parallel concatenation in a single block known as Inception Block, which was introduced by (Szegedy et al.,

As shown in Fig. 2 , each Inception block comprises four parallel paths leading to concatenation.

Different spatial sizes are extracted by first three paths, which use filter sizes of 1x1, 3x3, and 5x5.

We use a convolution of 1x1 in two middle paths to reduce the number of input channels. It also helps in reducing the model complexity. The fourth and last path uses a max-pooling of 3x3, followed by a 1x1 convolution, to change the channels. All paths use padding to maintain same height and width of the block's input and output. The outputs along each path are concatenated along the channel to constitute the Inception block output.

Mathematically speaking, for a given image I and filter K, the convolution operation can be expressed as:

Where ⌊ ⌋ is the floor function of There are also some special types of Convolution:

 Valid convolution: p=0  Same convolution: output size = input size  convolution: , it might be helpful to in some cases to shrink the number of channels without changing the other dimensions ( ). In the example below, we filled the filter with numbers for the sake of illustration. In a CNN, the Filter parameters are learned through backpropagation.

Our proposed SARS-Net CNN consists of a stack of four Inception blocks and a global average pooling to predict the final output. The dimensionality is reduced by using max-pooling layers in between the Inception blocks. The input is passed through convolutional blocks having a kernel size of 7x7, 3x3, and 1x1 with max-pooling between them. ReLU is used as an activation function for the convolutional blocks.

The Pooling operation can be expressed as: We can assert that:

In this section, we discuss various improvement techniques that we adopted over the base network, SARS-Net CNN, to formulate and design the other three nets, namely SARS-Net CC (Net-2), SARS-Net AA (Net-3), and SARS-Net RSP (Net-4).

For designing SARS-Net CC (Net-2), we add an extra module of CoordConv (Liu et al., 2018 ) that allows SARS-Net to learn either varying degrees of translational dependence or complete translational invariance without compromising computational and parametric efficiency of ordinary convolution.

The proposed CoordConv layer is a simple addition to the standard convolutional layer. It works by giving convolution access to its input coordinates using extra coordinate channels. Fig. 4 depicts the operation where two coordinates, i and j, are added. The i-coordinate channel is an h×w rank-1 matrix with its first row filled with 0's, its second row with 1's, its third with 2's, and j coordinate channel is similar but with columns filled in with constant values instead of rows. ( )

Blurred downsampling with a box filter is the same as average pooling. Replacing it with a more robust filter provides better shift-equivariance.

For SARS-Net RSP (Net-4), we replace the traditional module of max pooling with rank-based stochastic pooling (RSP) (Shi et al., 2016 ) that allows the SARS-Net to improve its performance without compromising its computational and parametric efficiency. Rank-based pooling is calculated based on the matrix ranks rather than their absolute values, as in the case of max and average pooling. 

As an improvement over our base model, we also use Graph Convolutional Network (GCN) to learn the Relation Aware Representation (RAR) from the CXR images to improve the performance of SARS-Net. The CNN prioritizes the local characteristics of the image and objects, such as edges, corners and interest points when the kernel performs the convolution over the visual feature maps;

thus, a lot of global information is lost. GCN helps in generalizing the regular convolution operation to graph convolution and works well with non-Euclidean data (Seti et al., 2020) . When we represent data patterns in a non-Euclidean way, we are giving it an inductive bias. We can prioritize and reward the model into learning certain patterns in data by changing its structure, given data of an arbitrary type, format and size. Generally, the inductive bias that is used is relational, in majority of current research pursuits and literature. We define an Adjacency Matrix (ADM) A R N×N for studying the relationship of the nodes {v i } of a graph G = (V, E), where there are N nodes v i V, i = 1,⋯, N and

Graph Convolutional Network is used to represent a graph G with the help of a Neural network (X, A) in which X R N×D , where D represents the feature dimension of every node, N represents the number of nodes, and AX represents the sum of features of all neighboring nodes. Thus, GCN can successfully learn the Relation Aware Representation (RAR) feature (Derr et al., 2020) .

A multi-layer GCN uses the layer-wise rule to update all the nodes' feature representation:

Where ̂ R N×N denotes the normalized form of the adjacency matrix A, and denotes the ReLU activation function. H (l) R N×dl represents the feature representation of the l-th layer. For finding the normalization matrix of A, i.e., A ↦ ̂, following steps are followed:

Firstly, the degree matrix d m R N×N that is a diagonal matrix is computed as:

Afterward, ̂ is deduced via degree matrix d m and ADM A (Kipf et al., 2017) .

Each image can be represented by its image-level features and its neighbor features by combining CNN and the two layer-GCN (Shi et al., 2019) . We use a two-layer Graph Convolutional Network: Fig. 3 is utilized as the individual image-level representation I R D where D = 100 in this work.

Once we obtain the individual image-level representations, we make them pass through k-means clustering (KMC) to obtain N cluster centroids (CCs) X R N×D . The Cluster Centroid correlation is used to display the relationships of the images. Then the adjacency matrix A R N×N is deduced as

KNN represents the cosine similarity (CS)-based kNN, its neighbor number is set to k KNN . 1, 2, 3, 4) . So, we have X a KNN(X b ), X b KNN(X a ).

Using the 'or' operation, we can conclude A ij = 1.

The node features X, and the adjacency matrix A is then passed to a two-layered-GCN, and we obtained H (2) R N×D if we set d 2 = D = 100. The H (2) is then combined with I via dot product

Through a linear projection with trainable weights W (2) R N×NC , in which N C denotes the number of classes, we can define

Where z R Nc , and W b represent the bias. Here, N C = 3 due to this tertiary classification problem, i.e.,

Normal, Non-COVID-19 or COVID-19. Hence, we only need to train (W (0) , W (1) , W (2) ) and their related biases for the two-layer Graph Convolutional Network. 

The proposed SARS-Net is trained and validated on the COVIDx dataset. The division of the COVIDx dataset for training and testing is depicted in Table 2 , where we split 90% of the data for training and validation (train split), and the rest 10% is the test split. Out of the 90% train split, we further divide 10% of the data for validation and use the rest for training. A scheduling LR policy is used to update the learning rate where the learning rate is decreased by a small amount when learning remains the same for an extended period. Adam optimizer is used for optimizing the SARS-Net.

The hyperparameters used for training were:

Various data augmentation techniques were used to prevent the overfitting of SARS-Net model on the training dataset, namely rotation, zoom, horizontal flip, and translation.

The SARS-Net model was built and evaluated using PyTorch, a DL library. The convolution operation used in the architecture can be summarized as a series of convolutional products using various kernels of filters applied to an input image. Finally, the output is passed through an activation function. In the end, we have fully connected layers for the classification of the convoluted feature outputs. 

The learned parameters at the layer are:

The operation of the Fully connected layer:

In general, considering the node as layer, we have the following equations:

The input the output of a convolution or pooling layer with the dimensions ( ).

We reshape the tensor to a 1D vector to plug it into the fully connected layer, having the dimension:

( ), thus:

( )

The learned parameters at the layer is:

 Weights with parameters  Bias with parameters

The algorithm for training the SARS-Net:

 Initialization of the model parameters and weights.

 For i =1,2,…N, (N=Total epochs):

1. Forward propagation:

i.

Compute the predicted value of through the DL model: ̂ .

ii. Evaluate the function: ( ) ∑ ( ̂ ) where m is total number of input images, stands for the model parameters, and is the cost function.

i. Apply a gradient descent method to update the parameters: ( ) Figure-10 : The training and testing workflow of the SARS-Net CADx system is depicted in the following flow chart.

We use Cross-Entropy as the loss function/criterion for training our SARS-Net because it minimizes the distance between two probability distributions, i.e., the actual and the predicted. This section shows how class imbalance can cause problems while training DL models by creating a bias against the majority class. We prove the above statements by using the formula for cross-entropy loss, and finally, we use class-specific weight factors to handle the data imbalance.

The cross-entropy loss contribution from the i th training data case is:

Where x i and y i are the input features and labels, and f(x i ) is the SARS-Net output, the probability that it is positive. Suppose we use a normal cross-entropy loss function with a highly unbalanced dataset.

In that case, the algorithm will be trained to prioritize the majority class (i.e., Non-COVID cases)

since it contributes more to the loss. For any case, i.e., (Y i = 0) or (1-Y i = 0), so only one of these terms contributes to the loss (the other term is multiplied by zero and becomes zero).

We can rewrite the overall average cross-entropy loss over the entire training set D of size as follows:

Using this formulation, we can see that if there is a large imbalance with very few positive training cases, the negative class dominates the loss. Summing the contribution over all the training cases for each class (i.e., pathological condition), we see that the contribution of each class (i.e., positive or negative) is:

freq neg-=

As we see in our case, positive COVID-19 contributions are significantly lower than negative ones.

However, we want contributions to be equal. One way of doing this is by multiplying each example from each class by a class-specific weight factor W pos , and W neg, so that the overall contribution of each class is the same.

To implement this, we want: =

which can be implemented by:

This way, we balance the contribution of positive and negative labels. After computing the weights, our final weighted loss for each training case is:

To evaluate the efficiency and demonstrate the usefulness of the proposed SARS-Net, we carry out extensive quantitative analysis to better understand its detection performance. We calculate the test As mentioned in the Inception blocks, SARS-Net uses parallel concatenation to extract information in parallel, employing different kernel sizes ranging from 1 x 1 to 11 x 11. Besides, it has long-range connectivity and uses repeated blocks to learn complex features of the COVID-19 infected CXR images and produce good results compared to other CNN architectures. It is also advantageous to use various sized kernels in DL models to extract complex features.

As can be inferred from the above values, the addition of Graph Convolutional Network (GCN) enhances the performance of the proposed SARS-Net as GCNs can quickly learn the RARs among the test set. Therefore, classifiers with GCNs provide more precise results than those without GCNs. 

We compare the performance of SARS-Net with various other SOTA model literature that adopt a similar methodology and report the results in terms of Accuracy and Sensitivity for COVID-19

detection. Table 7 shows the results in Accuracy and Sensitivity for detecting COVID-19 of our

proposed SARS-Net model compared to other State of the art literature studies. As is shown, SARS-Net yielded the following performance in terms of Sensitivity = 92.90 and Accuracy = 97.60. 

To get a better insight into how DL models make their decision, we analyze the output of GRAD-CAM, which highlights the areas of the input image in the form of heat maps that impact SARS-Net decisions. To interpret SARS-Net predictions, we produce the heatmaps to highlight the image areas most indicative of the disease using the Class Activation maps (Zhou et al., 2016) , generated by a method known as GRAD-CAM. The left portion of Fig. 11 (A) shows an input image fed into the SARS-Net, and the right portion (B) showcases a heatmap pointing out the areas of the image that indicate COVID-19. Figure 11 : Output of GRADCAM for the COVID-19 Detection.

The fully trained convolutional network, SARS-Net, is fed with an image to generate the Class Activation Mappings, and the feature maps are extracted from the output of the last convolutional layer. Suppose W c.k is the last classification layer's weight for feature map k belonging to class c and be the feature map. To generate a mapping of the most significant salient features of an input image belonging to a class c, we calculate the weighted sum of the feature maps using the weights connected to them. Specifically, a mapping is calculated as:

In the last step to obtain the Class activation mapping to visualize the most critical features of an input image, we upscale the mapping according to the dimensions of the image, followed by overlaying the image.

In this paper, we introduce SARS-Net, a CADx system combining Graph Convolutional Network and Convolutional Neural Network for COVID-19 detection from CXR images. Extensive experiments showed that the proposed SARS-Net model attains the best results among all other proposed networks and attains superior performances to state-of-the-art methods. The SARS-Net is a combination of the SARS-Net CNN model and a 2L-GCN model. Here, SARS-Net CNN aids to extract image-level features, whereas GCN helps to extract relation-awareness features. The combination of both two networks helps increase the performance of our proposed SARS-Net model. Medical image processing has been gaining much attention recently due to the emergence of deeper and highaccuracy networks that can compete against humans and speed up medical research to a great extent.

COVID-19 detection has always been a crucial step for diagnosis in the present pandemic scenario, and recently many computer-aided analysis approaches are being used for quick and more reliable analysis. In this paper, the proposed SARS-Net model incorporates the latest profound DL 

We all authors declare that we have no conflict of Interest.

A Novel Coronavirus from Patients with Pneumonia in China

COVID-19 image data collection

Figure 1 COVID-19 chest x-ray data initiative

Actualmed COVID-19 chest x-ray data initiative

COVID-19 radiography database

RSNA pneumonia detection challenge

Preliminary estimation of the basic reproduction number of novel coronavirus (2019-nCoV) in China, from 2019 to 2020: a data-driven analysis in the early phase of the outbreak

Serial interval of COVID-19 among publicly reported confirmed cases

Detection of SARS-CoV-2 in different types of clinical specimens

Liming Xia. Initial CT findings and temporal changes in patients with the novel coronavirus pneumonia (2019-nCoV): a study of 63 patients in Wuhan

Frequency and distribution of chest radiographic findings in patients positive for COVID-19

Interpretation of plain chest roentgenogram

Weakly Labeled Data Augmentation for Deep Learning: A Study on COVID-19 Detection in Chest X-Rays

A deep learning approach to characterize 2019 coronavirus disease (COVID-19) pneumonia in chest CT images

detection of covid-19 chest x-ray using support vector machine and convolutional neural network

Role of biological data mining and machine learning techniques in detecting and diagnosing the novel coronavirus (COVID-19): a systematic review

Coronavirus Disease (COVID-19): A Machine Learning Bibliometric Analysis

A British Society of Thoracic Imaging statement: considerations in designing local imaging diagnostic algorithms for the COVID-19 pandemic

Going deeper with convolutions

Detection and classification of cancer in whole slide breast histopathology images using deep convolutional networks

Classification of COVID-19 in chest X-ray images using DeTraC deep convolutional neural network

Exudate segmentation using fully convolutional neural networks and inception modules

Multi-label classification of multi-modality skin lesion via hyperconnected convolutional neural network

Polyp detection during colonoscopy using a regression-based convolutional neural network with a tracker

Automated detection of covid-19 cases using deep neural networks with x-ray images

Named-entity recognition in sports field based on a character-level graph convolutional network

Portable chest x-ray in coronavirus disease-19 (covid-19): A pictorial review

A modified deep convolutional neural network for detecting COVID-19 and pneumonia from chest X-ray images based on the concatenation of Xception and ResNet50V2

Automatically discriminating and localizing COVID-19 from communityacquired Pneumonia on chest X-rays

Covid-mobile expert: On-device covid-19 screening using snapshots of a chest xray

Covid-caps: A capsule network-based framework for identification of covid-19 cases from x-ray images

Coronet: A deep network architecture for semi-supervised task-based identification of covid-19 from chest x-ray images

Towards an effective and efficient deep learning model for covid-19 patterns detection in x-ray images

Deep-covid: Predicting covid-19 from chest x-ray images using deep transfer learning

COVIDiagnosis-net: Deep Bayes-squeeze net-based diagnosis of the coronavirus disease 2019 (COVID-19) from x-ray images

Unveiling covid-19 from chest x-ray with deep learning: a hurdles race with small data

Machine-learning classification of texture features of portable chest X-ray accurately classifies COVID-19 lung infection

Rank-based pooling for deep convolutional neural networks

Covid-19 identification in chest xray images on flat and hierarchical classification scenarios

Coronet: A deep neural network for detection and diagnosis of covid-19 from chest x-ray images

Covid-resnet: A deep learning framework for screening of covid19 from radiographs

Learning deep features for discriminative localization

Covid-net: A tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray images

Graph Convolutional Networks for Cervical Cell Classification

Semi-supervised classification with graph convolutional networks

A deep learning system to screen novel coronavirus disease 2019 pneumonia

A deep learning algorithm using CT images to screen for Corona virus disease (COVID-19)

Epidemic graph convolutional network

Graph-based relation-aware representation learning for clothing matching

Prediction of the Health Effects of Food Peptides and Elucidation of the Mode-of-action Using Multi-task Graph Convolutional Neural Network

Improved Breast Cancer Classification Through Combining Graph Convolutional Network and Convolutional Neural Network

An intriguing failing of convolutional neural networks and the coordconv solution

Authors Biography Aayush Kumar: He is in currently doing his under graduate study in Computer Science Engineering at KIIT Deemed to be University

He is in currently doing his under graduate study in Computer Science Engineering at KIIT Deemed to be University, Bhubaneswear, India. He has shown interest for AI in Healthcare

Machine Learning, Data Mining and Cognitive Sciences. He has developed two new optimization algorithms known as Social Group Optimization (SGO) published in Springer Journal and SELO

SM'15) received his BE in Information Sciences in 2004, and MPhil in Communication and Information Engineering in