Generation of Synthetic Chest X-ray Images and Detection of COVID-19: A Deep Learning Based Approach

Abstract

COVID-19 is a disease caused by the SARS-CoV-2 virus. The COVID-19 virus spreads when a person comes into contact with an affected individual. This is mainly through drops of saliva or nasal discharge. Most of the affected people have mild symptoms while some people develop acute respiratory distress syndrome (ARDS), which damages organs like the lungs and heart. Chest X-rays (CXRs) have been widely used to identify abnormalities that help in detecting the COVID-19 virus. They have also been used as an initial screening procedure for individuals highly suspected of being infected. However, the availability of radiographic CXRs is still scarce. This can limit the performance of deep learning (DL) based approaches for COVID-19 detection. To overcome these limitations, in this work, we developed an Auxiliary Classifier Generative Adversarial Network (ACGAN), to generate CXRs. Each generated X-ray belongs to one of the two classes COVID-19 positive or normal. To ensure the goodness of the synthetic images, we performed some experimentation on the obtained images using the latest Convolutional Neural Networks (CNNs) to detect COVID-19 in the CXRs. We fine-tuned the models and achieved more than 98% accuracy. After that, we also performed feature selection using the Harmony Search (HS) algorithm, which reduces the number of features while retaining classification accuracy. We further release a GAN-generated dataset consisting of 500 COVID-19 radiographic images.

Keywords: COVID-19 detection; chest X-ray; deep learning; feature selection; generative adversarial network; harmony search; synthetic data generation.

Figure 1

Flowchart representing the overall process.

Figure 2

Schematic diagram of ACGAN.

Figure 3

A schematic diagram of the skip connections in the ResNet architecture. (He et al. [22]).

Figure 4

A schematic diagram representing an inception block. (Chollet [23]).

Figure 5

A schematic diagram representing a simplified inception block. Compared to Figure 4, it only contains a single size of convolutions ($3\times 3$) and does not contain pooling layers. (Chollet [23]).

Figure 6

A strictly equivalent reformulation of the simplified inception block of Figure 5. Chollet [23].

Figure 7

A flowchart representing the HS algorithm.

Figure 8

The loss curve for a single CNN model (VGG16) for 20 epochs.

Figure 9

Some synthetic images that were generated using the present approach. (a) Synthetic images belonging to the COVID-19 infected CXR class generated by the GAN; (b) synthetic images belonging to the normal CXR class generated by the GAN.