ABSTRACT
A current COVID-19 detection tool is CXR imaging, which has been developing since 2019 to provide early diagnosis;it can be performed in any health unit and is more affordable than Real Time Polymerase Chain Reaction (RT-PCR) tests. However, diagnosis with Chest X Ray (CXR) images had not achieved the predictive capacity required to replace the RT-PCR test;previous studies with a limited number of images have evaluated their models. This research seeks to contribute to the detection of COVID-19 from CXR images, with the evaluation of a convolutional neural network model from CXR images, through the use of open source code on a free dataset of approximately 30 thousand images. The algorithm and mathematical model used was DenseNet-201. The results of the experiment show a precision and accuracy of more than 95% and specificity, sensitivity, predictive ability and F1 measurement of more than 90%.Copyright © 2022, Anka Publishers. All rights reserved.
ABSTRACT
The detection of COVID-19 by CXR imaging is a support tool for physicians and specialists since the pandemic and has been evolving rapidly because it provides early diagnosis, can be performed in any health center, and is more affordable than Real-Time Polymerase Chain Reaction (RT-PCR) tests. However, Chest X-Ray (CXR) imaging had not achieved the predictive capacity needed to replace the RT-PCR test;previous studies have evaluated their models with a limited amount of images. This study aims to contribute to the evaluation of a convolutional neural network (CNN) model to detect COVID-19 from CXR images, using open source and a free dataset containing approximately 30,000 images. The mathematical model or algorithm used was VGGNet-16. The results of the experiments show accuracy and precision of more than 95% and sensitivity, specificity, F1-measure,andthedictive ability of more than 90%.Copyright © 2022, Anka Publishers. All rights reserved.
ABSTRACT
BACKGROUND: The COVID-19 pandemic has raised numerous questions among patients with immune-mediated inflammatory diseases regarding potential reciprocal effects of COVID-19 and their underlying disease, and potential effects of immunomodulatory therapy on outcomes related to COVID-19. The seroprevalence of SARS-CoV-2 and factors associated with symptomatic COVID-19 in patients with immune-mediated inflammatory diseases are still unclear. The Euro-COVIMID study aimed to determine the serological and clinical prevalence of COVID-19 among patients with immune-mediated inflammatory diseases, as well as factors associated with COVID-19 occurrence and the impact of the pandemic in its management. METHODS: In this multicentre cross-sectional study, patients aged 18 years or older with a clinical diagnosis of rheumatoid arthritis, axial spondyloarthritis, systemic lupus erythematosus, Sjögren's syndrome, or giant cell arteritis were recruited from six tertiary referral centres in France, Germany, Italy, Portugal, Spain, and the UK. Demographics, comorbidities, treatments, and recent disease flares, as well as information on COVID-19 symptoms, were collected through a questionnaire completed by participants. SARS-CoV-2 serology was systematically tested. The main outcome was the serological and clinical prevalence of COVID-19. Factors associated with symptomatic COVID-19 were assessed by multivariable logistic regression, and incidence of recent disease flares, changes in treatments for underlying disease, and the reasons for treatment changes were also assessed. This study is registered with ClinicalTrials.gov, NCT04397237. FINDINGS: Between June 7 and Dec 8, 2020, 3136 patients with an immune-mediated inflammatory disease answered the questionnaire. 3028 patients (median age 58 years [IQR 46-67]; 2239 [73·9%] women and 789 [26·1%] men) with symptomatic COVID-19, serological data, or both were included in analyses. SARS-CoV-2 antibodies were detected in 166 (5·5% [95% CI 4·7-6·4]) of 3018 patients who had serology tests. Symptomatic COVID-19 occurred in 122 (4·0% [95% CI 3·4-4·8]) of 3028 patients, of whom 24 (19·7%) were admitted to hospital and four (3·3%) died. Factors associated with symptomatic COVID-19 were higher concentrations of C-reactive protein (odds ratio 1·18, 95% CI 1·05-1·33; p=0·0063), and higher numbers of recent disease flares (1·27, 1·02-1·58; p=0·030), whereas use of biological therapy was associated with reduced risk (0·51, 0·32-0·82; p=0·0057). At least one disease flare occurred in 654 (21·6%) of 3028 patients. Over the study period, 519 (20·6%) of 2514 patients had treatment changes, of which 125 (24·1%) were due to the pandemic. INTERPRETATION: This study provides key insights into the epidemiology and risk factors of COVID-19 among patients with immune-mediated inflammatory diseases. Overall, immunosuppressants do not seem to be deleterious in this scenario, and the control of inflammatory activity seems to be key when facing the pandemic. FUNDING: Pfizer, Sanofi, Amgen, Galapagos, and Lilly.