COVID-19 on Chest Radiographs: A Multireader Evaluation of an                     Artificial Intelligence System

Murphy, Keelin; Smits, Henk; Knoops, Arnoud J.G.; Korst, Michael B. J. M.; Samson, Tijs; Scholten, Ernst Th.; Schalekamp, Steven; Schaefer-Prokop, Cornelia; Philipsen, Rick H. H. M.; Meijers, Annet; Melendez, Jaime; Ginneken, Bram van; Rutten, M.J.C.M.

doi:10.1148/radiol.2020201874

Cited by 194 publications

(199 citation statements)

References 18 publications

Supporting

Mentioning

184

Contrasting

Unclassified

Order By: Relevance

“…This integrated approach may be a valid alternative when or where other more specific tests (in primis RT-PCR tests) are limited. To this end, an added value comes from the promising artificial intelligence techniques developed for improving the diagnostic accuracy of imaging and assisting radiologists and clinicians in the CXR evaluation as part of the COVID-19 triage process [24]. The results of our study should also warn that when dealing with the suspicion of a positive COVID-19 in a patient admitted to the emergency room a few days after the onset of symptoms and without severe alterations of CRP and LDH, physicians should not be surprised to be faced with a negative CXR.…”

Section: Discussionmentioning

confidence: 99%

Baseline chest X-ray in coronavirus disease 19 (COVID-19) patients: association with clinical and laboratory data

et al. 2020

View full text Add to dashboard Cite

Purpose To assess the reliability of CXR and to describe CXR findings and clinical and laboratory characteristics associated with positive and negative CXR. Methods Retrospective two-center study on consecutive patients admitted to the emergency department of two north-western Italian hospitals in March 2020 with clinical suspicion of COVID-19 confirmed by RT-PCR and who underwent CXR within 24 h of the swab execution. 260 patients (61% male, 62.8 ± 15.8 year) were enrolled. CXRs were rated as positive (CXR+) or negative (CXR−), and features reported included presence and distribution of airspace opacities, pleural effusion and reduction in lung volumes. Clinical and laboratory data were collected. Statistical analysis was performed with nonparametric tests, binary logistic regression (BLR) and ROC curve analysis. Results Sensitivity of CXR was 61.1% (95%CI 55–67%) with a typical presence of bilateral (62.3%) airspace opacification, more often with a lower zone (88.7%) and peripheral (43.4%) distribution. At univariate analysis, several factors were found to differ significantly between CXR+ and CXR−. The BLR confirmed as significant predictors only lactate dehydrogenase (LDH), C-reactive protein (CRP) and interval between the onset of symptoms and the execution of CXR. The ROC curve procedure determined that CRX+ was associated with LDH > 500 UI/L (AUC = 0.878), CRP > 30 mg/L (AUC = 0.830) and interval between the onset of symptoms and the execution of CXR > 4 days (AUC = 0.75). The presence of two out of three of the above-mentioned predictors resulted in CXR+ in 92.5% of cases, whereas their absence in 7.4%. Conclusion CXR has a low sensitivity. LDH, CRP and interval between the onset of symptoms and the execution of CXR are major predictors for a positive CXR.

show abstract

Section: Discussionmentioning

confidence: 99%

Baseline chest X-ray in coronavirus disease 19 (COVID-19) patients: association with clinical and laboratory data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…In resource-constrained settings with large throughputs to handle, such as with the COVID-19 pandemic, artificial intelligence (AI) may help expedite reading times, thus becoming an important asset in the clinical management of these patients. An AI system for the detection of COVID-19 was recently shown to be able to identify COVID-19 pneumonia on CXR with performance comparable to six independent radiologists [14]. However, the role of AI on CXR as a prognostic tool has not yet been evaluated in COVID-19 patients.…”

Section: Introductionmentioning

confidence: 99%

Initial chest radiographs and artificial intelligence (AI) predict clinical outcomes in COVID-19 patients: analysis of 697 Italian patients

et al. 2020

View full text Add to dashboard Cite

Objective To evaluate whether the initial chest X-ray (CXR) severity assessed by an AI system may have prognostic utility in patients with COVID-19. Methods This retrospective single-center study included adult patients presenting to the emergency department (ED) between February 25 and April 9, 2020, with SARS-CoV-2 infection confirmed on real-time reverse transcriptase polymerase chain reaction (RT-PCR). Initial CXRs obtained on ED presentation were evaluated by a deep learning artificial intelligence (AI) system and compared with the Radiographic Assessment of Lung Edema (RALE) score, calculated by two experienced radiologists. Death and critical COVID-19 (admission to intensive care unit (ICU) or deaths occurring before ICU admission) were identified as clinical outcomes. Independent predictors of adverse outcomes were evaluated by multivariate analyses. Results Six hundred ninety-seven 697 patients were included in the study: 465 males (66.7%), median age of 62 years (IQR 52–75). Multivariate analyses adjusting for demographics and comorbidities showed that an AI system-based score ≥ 30 on the initial CXR was an independent predictor both for mortality (HR 2.60 (95% CI 1.69 − 3.99; p < 0.001)) and critical COVID-19 (HR 3.40 (95% CI 2.35–4.94; p < 0.001)). Other independent predictors were RALE score, older age, male sex, coronary artery disease, COPD, and neurodegenerative disease. Conclusion AI- and radiologist-assessed disease severity scores on CXRs obtained on ED presentation were independent and comparable predictors of adverse outcomes in patients with COVID-19. Trial registration ClinicalTrials.gov NCT04318366 (https://clinicaltrials.gov/ct2/show/NCT04318366). Key Points • AI system–based score ≥ 30 and a RALE score ≥ 12 at CXRs performed at ED presentation are independent and comparable predictors of death and/or ICU admission in COVID-19 patients. • Other independent predictors are older age, male sex, coronary artery disease, COPD, and neurodegenerative disease. • The comparable performance of the AI system in relation to a radiologist-assessed score in predicting adverse outcomes may represent a game-changer in resource-constrained settings.

show abstract

“…Other researchers have been able to apply pattern recognition to computed tomography (CT) lung scans and automatically diagnose the presence or absence of COVID-19 infection. The investigations verified the effectiveness of the software diagnosis by corroborating the presence or not of the virus with PCR tests [ 268 ]. InferVision, a Chinese startup, applies this technology in 45 hospitals from China, the USA and Europe and has performed more than 76,000 tests [ 269 ].…”

Section: Treatments For Covid-19mentioning

confidence: 85%

Biomedical Science to Tackle the COVID-19 Pandemic: Current Status and Future Perspectives

et al. 2020

View full text Add to dashboard Cite

The coronavirus infectious disease (COVID-19) pandemic emerged at the end of 2019, and was caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), which has resulted in an unprecedented health and economic crisis worldwide. One key aspect, compared to other recent pandemics, is the level of urgency, which has started a race for finding adequate answers. Solutions for efficient prevention approaches, rapid, reliable, and high throughput diagnostics, monitoring, and safe therapies are needed. Research across the world has been directed to fight against COVID-19. Biomedical science has been presented as a possible area for combating the SARS-CoV-2 virus due to the unique challenges raised by the pandemic, as reported by epidemiologists, immunologists, and medical doctors, including COVID-19’s survival, symptoms, protein surface composition, and infection mechanisms. While the current knowledge about the SARS-CoV-2 virus is still limited, various (old and new) biomedical approaches have been developed and tested. Here, we review the current status and future perspectives of biomedical science in the context of COVID-19, including nanotechnology, prevention through vaccine engineering, diagnostic, monitoring, and therapy. This review is aimed at discussing the current impact of biomedical science in healthcare for the management of COVID-19, as well as some challenges to be addressed.

show abstract

COVID-19 on Chest Radiographs: A Multireader Evaluation of an Artificial Intelligence System

Cited by 194 publications

References 18 publications

Baseline chest X-ray in coronavirus disease 19 (COVID-19) patients: association with clinical and laboratory data

Baseline chest X-ray in coronavirus disease 19 (COVID-19) patients: association with clinical and laboratory data

Initial chest radiographs and artificial intelligence (AI) predict clinical outcomes in COVID-19 patients: analysis of 697 Italian patients

Biomedical Science to Tackle the COVID-19 Pandemic: Current Status and Future Perspectives

Contact Info

Product

Resources

About