COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images

Abstract

The Coronavirus Disease 2019 (COVID-19) pandemic continues to have a devastating effect on the health and well-being of the global population. A critical step in the fight against COVID-19 is effective screening of infected patients, with one of the key screening approaches being radiology examination using chest radiography. It was found in early studies that patients present abnormalities in chest radiography images that are characteristic of those infected with COVID-19. Motivated by this and inspired by the open source efforts of the research community, in this study we introduce COVID-Net, a deep convolutional neural network design tailored for the detection of COVID-19 cases from chest X-ray (CXR) images that is open source and available to the general public. To the best of the authors’ knowledge, COVID-Net is one of the first open source network designs for COVID-19 detection from CXR images at the time of initial release. We also introduce COVIDx, an open access benchmark dataset that we generated comprising of 13,975 CXR images across 13,870 patient patient cases, with the largest number of publicly available COVID-19 positive cases to the best of the authors’ knowledge. Furthermore, we investigate how COVID-Net makes predictions using an explainability method in an attempt to not only gain deeper insights into critical factors associated with COVID cases, which can aid clinicians in improved screening, but also audit COVID-Net in a responsible and transparent manner to validate that it is making decisions based on relevant information from the CXR images. By no means a production-ready solution, the hope is that the open access COVID-Net, along with the description on constructing the open source COVIDx dataset, will be leveraged and build upon by both researchers and citizen data scientists alike to accelerate the development of highly accurate yet practical deep learning solutions for detecting COVID-19 cases and accelerate treatment of those who need it the most.

COVIDNet-CT: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases From Chest CT Images

“…Inspired by Wang and Wong ( 19 ), a machine-driven design exploration strategy was leveraged to create the proposed COVIDNet-CT. More specifically, machine-driven design exploration involves the automatic exploration of possible network architecture designs and identifies the optimal microarchitecture and macroarchitecture patterns with which to build the deep neural network. As discussed in Wang and Wong ( 19 ), the use of machine-driven design exploration allows for greater flexibility and granularity in the design process as compared to manual design, and ensures that the resulting network satisfies the given operational requirements.…”

Web‐based efficient dual attention networks to detect COVID‐19 from X‐ray images

“…In Table 1, we compared the EDANet with three different deep models VGG‐19, ResNet‐50 and COVID‐Net in terms of test accuracy, model complexity (number of parameters) and computational complexity (number of MAC operations). It shows that EDANet outperforms the all tested models on the COVIDx [6 ] test data set. EDANet yields the accuracy of which is 13.0, and higher than VGG‐19, ResNet‐50 and COVID‐Net [6 ], respectively.…”

“…For the evaluation metrics, we calculated the test accuracy, sensitivity and positive predictive value (PPV) to compare the performance of the proposed model with the COVID‐net and other models in [6 ]. However, the test accuracy also compared with the architectural complexity (number of model parameters) and computational complexity (number of model multiply‐accumulate (MAC) operations) is shown in Table 1.…”

“…The data set of only 50 images (25 healthy and 25 COVID‐19 infected patients) are used in their experiment and obtained good performance of f1‐scores of 89 and 91% with the dense convolutional network for normal and COVID‐19, respectively. Recently, in COVID‐net [6 ], a new CNN architecture and an open‐access benchmark data set, COVIDx, is introduced for COVID‐19 detection from CXR images. The COVID‐net yields an accuracy of 90.3% and sensitivity of 91.0% for the class of the COVID‐19.…”

“…On the other hand, their implementation details are not publicly available. Instead, COVID‐net [6 ] make their source code and data set publicly available. Therefore, we claim to design a lightweight model with a low computational complexity that deploys with a web‐based publicly available application to help clinicians for identifying COVID‐19.…”