COVID-19 classification of X-ray images using deep neural networks

Abstract

Objectives In the midst of the coronavirus disease 2019 (COVID-19) outbreak, chest X-ray (CXR) imaging is playing an important role in diagnosis and monitoring of patients with COVID-19. We propose a deep learning model for detection of COVID-19 from CXRs, as well as a tool for retrieving similar patients according to the model’s results on their CXRs. For training and evaluating our model, we collected CXRs from inpatients hospitalized in four different hospitals. Methods In this retrospective study, 1384 frontal CXRs, of COVID-19 confirmed patients imaged between March and August 2020, and 1024 matching CXRs of non-COVID patients imaged before the pandemic, were collected and used to build a deep learning classifier for detecting patients positive for COVID-19. The classifier consists of an ensemble of pre-trained deep neural networks (DNNS), specifically, ReNet34, ReNet50¸ ReNet152, and vgg16, and is enhanced by data augmentation and lung segmentation. We further implemented a nearest-neighbors algorithm that uses DNN-based image embeddings to retrieve the images most similar to a given image. Results Our model achieved accuracy of 90.3%, (95% CI: 86.3–93.7%) specificity of 90% (95% CI: 84.3–94%), and sensitivity of 90.5% (95% CI: 85–94%) on a test dataset comprising 15% (350/2326) of the original images. The AUC of the ROC curve is 0.96 (95% CI: 0.93–0.97). Conclusion We provide deep learning models, trained and evaluated on CXRs that can assist medical efforts and reduce medical staff workload in handling COVID-19. Key Points • A machine learning model was able to detect chest X-ray (CXR) images of patients tested positive for COVID-19 with accuracy and detection rate above 90%. • A tool was created for finding existing CXR images with imaging characteristics most similar to a given CXR, according to the model’s image embeddings. Supplementary Information The online version contains supplementary material available at 10.1007/s00330-021-08050-1.

Cov-Net: A computer-aided diagnosis method for recognizing COVID-19 from chest X-ray images via machine vision

“…Currently in practice, reverse transcription-polymerase chain reaction (RT-PCR) is the most popular screening method of COVID-19 detection since the beginning of the epidemic, which is regarded as a reliable gold-standard. While in addition, imaging can serve as a complementary method to assist radiologist in diagnosing COVID-19 so as to achieve greater diagnostic certainty ( Ai et al, 2020 , Ismael and Şengür, 2021 , Kassani et al, 2021 ), where the most remarkable merits include convenient acquisition and time-saving detection ( Keidar et al, 2021 ). Consequently, the application of computer aided diagnosis (CAD) framework incorporating chest X-ray (CXR) or computed tomography (CT) images can circumvent issues associated with existing diagnostic procedures and enable enhanced recognition of COVID-19.…”

A Promising and Challenging Approach: Radiologists’ Perspective on Deep Learning and Artificial Intelligence for Fighting COVID-19

“…Wang et al [17] developed a deep learning system trained on a heterogeneous multicenter dataset containing 145,202 images to distinguish viral pneumonia from other types of pneumonia and absence of pneumonia, obtaining sensitivity, specificity, and an AUC of 87.04%, 92.94%, and 0.968, respectively. Wehbe et al and Keidar et al implemented comparable machine learning approaches for detection, achieving an AUC of 0.90 and 0.96, respectively [18,19].…”

Untitled

“…The authors in [22] proposed a deep learning classification model for detection of coronavirus using X-ray images based on deep features and SVM classifier. The research paper [23] built a deep learning classifier based on an ensemble of pre-trained deep neural networks (DNNS), specifically, ReNet34, ReNet50¸ ReNet152, and vgg16 for detecting patients' positive for COVID-19. In [24], the authors addressed a method for visual diagnosis of cases of COVID-19 on CXR images.…”