Generative adversarial network for automatic quantification of Coronavirus disease 2019 pneumonia on chest radiographs.

PURPOSE: To develop a generative adversarial network (GAN) to quantify COVID-19 pneumonia on chest radiographs automatically. MATERIALS AND METHODS: This retrospective study included 50,000 consecutive non-COVID-19 chest CT scans in 2015-2017 for training. Anteroposterior virtual chest, lung, and pneumonia radiographs were generated from whole, segmented lung, and pneumonia pixels from each CT scan. Two GANs were sequentially trained to generate lung images from radiographs and to generate pneumonia images from lung images. GAN-driven pneumonia extent (pneumonia area/lung area) was expressed from 0% to 100%. We examined the correlation of GAN-driven pneumonia extent with semi-quantitative Brixia X-ray severity score (one dataset, n = 4707) and quantitative CT-driven pneumonia extent (four datasets, n = 54-375), along with analyzing a measurement difference between the GAN and CT extents. Three datasets (n = 243-1481), where unfavorable outcomes (respiratory failure, intensive care unit admission, and death) occurred in 10%, 38%, and 78%, respectively, were used to examine the predictive power of GAN-driven pneumonia extent. RESULTS: GAN-driven radiographic pneumonia was correlated with the severity score (0.611) and CT-driven extent (0.640). 95% limits of agreements between GAN and CT-driven extents were -27.1% to 17.4%. GAN-driven pneumonia extent provided odds ratios of 1.05-1.18 per percent for unfavorable outcomes in the three datasets, with areas under the receiver operating characteristic curve (AUCs) of 0.614-0.842. When combined with demographic information only and with both demographic and laboratory information, the prediction models yielded AUCs of 0.643-0.841 and 0.688-0.877, respectively. CONCLUSION: The generative adversarial network automatically quantified COVID-19 pneumonia on chest radiographs and identified patients with unfavorable outcomes.

Comparison of CT findings of coronavirus disease 2019 (COVID-19) pneumonia caused by different major variants.

PURPOSE: To explore the CT findings and pneumonnia progression pattern of the Alpha and Delta variants of SARS-CoV-2 by comparing them with the pre-existing wild type. METHOD: In this retrospective comparative study, a total of 392 patients with COVID-19 were included: 118 patients with wild type (70 men, 56.8 ± 20.7 years), 137 with Alpha variant (93 men, 49.4 ± 17.0 years), and 137 with Delta variant (94 men, 45.4 ± 12.4). Chest CT evaluation included opacities and repairing changes as well as lesion distribution and laterality. Chest CT severity score was also calculated. These parameters were statistically compared across the variants. RESULTS: Ground glass opacity (GGO) with consolidation and repairing changes were more frequent in the order of Delta variant, Alpha variant, and wild type throughout the disease course. Delta variant showed GGO with consolidation more conspicuously than did the other two on days 1-4 (vs. wild type, Bonferroni corrected p = 0.01; vs. Alpha variant, Bonferroni corrected p = 0.003) and days 5-8 (vs. wild type, Bonferroni corrected p < 0.001; vs. Alpha variant, Bonferroni corrected-p = 0.003). Total lung CT severity scores of Delta variant were higher than those of wild type on days 1-4 and 5-8 (Bonferroni corrected p = 0.01 and Bonferroni corrected p = 0.005, respectively) and that of Alpha variant on days 1-4 (Bonferroni corrected p = 0.002). There was no difference in the CT findings between wild type and Alpha variant. CONCLUSIONS: Pneumonia progression of Delta variant may be more rapid and severe in the early stage than in the other two.

Deep Learning-Based Automatic CT Quantification of Coronavirus Disease 2019 Pneumonia: An International Collaborative Study.

OBJECTIVE: We aimed to develop and validate the automatic quantification of coronavirus disease 2019 (COVID-19) pneumonia on computed tomography (CT) images. METHODS: This retrospective study included 176 chest CT scans of 131 COVID-19 patients from 14 Korean and Chinese institutions from January 23 to March 15, 2020. Two experienced radiologists semiautomatically drew pneumonia masks on CT images to develop the 2D U-Net for segmenting pneumonia. External validation was performed using Japanese (n = 101), Italian (n = 99), Radiopaedia (n = 9), and Chinese data sets (n = 10). The primary measures for the system's performance were correlation coefficients for extent (%) and weight (g) of pneumonia in comparison with visual CT scores or human-derived segmentation. Multivariable logistic regression analyses were performed to evaluate the association of the extent and weight with symptoms in the Japanese data set and composite outcome (respiratory failure and death) in the Spanish data set (n = 115). RESULTS: In the internal test data set, the intraclass correlation coefficients between U-Net outputs and references for the extent and weight were 0.990 and 0.993. In the Japanese data set, the Pearson correlation coefficients between U-Net outputs and visual CT scores were 0.908 and 0.899. In the other external data sets, intraclass correlation coefficients were between 0.949-0.965 (extent) and between 0.978-0.993 (weight). Extent and weight in the top quartile were independently associated with symptoms (odds ratio, 5.523 and 10.561; P = 0.041 and 0.016) and the composite outcome (odds ratio, 9.365 and 7.085; P = 0.021 and P = 0.035). CONCLUSIONS: Automatically quantified CT extent and weight of COVID-19 pneumonia were well correlated with human-derived references and independently associated with symptoms and prognosis in multinational external data sets.

Standardized reporting systems of chest computed tomography in a population with low coronavirus disease 2019 prevalence: A retrospective comparative study.

PURPOSE: To compare the diagnostic performance and interobserver agreement of three reporting systems for computed tomography findings in coronavirus disease 2019 (COVID-19), namely the COVID-19 Reporting and Data System (CO-RADS), COVID-19 Imaging Reporting and Data System (COVID-RADS), and Radiological Society of North America (RSNA) expert consensus statement, in a low COVID-19 prevalence area. METHOD: This institutional review board approval single-institutional retrospective study included 154 hospitalized patients between April 1 and May 21, 2020; 26 (16.9 %; 63.2 ± 14.1 years, 21 men) and 128 (65.7 ± 16.4 years, 87 men) patients were diagnosed with and without COVID-19 according to reverse transcription-polymerase chain reaction results, respectively. Written informed consent was waived due to the retrospective nature of the study. Six radiologists independently classified chest computed tomography images according to each reporting system. The area under receiver operating characteristic curves, sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and interobserver agreements were calculated and compared across the systems using paired t-test and kappa analysis. RESULTS: Mean area under receiver operating characteristic curves were as follows: CO-RADS, 0.89 (95 % confidence interval [CI], 0.87-0.90); COVID-RADS, 0.78 (0.75-0.80); and RSNA expert consensus statement, 0.88 (0.86-0.90). Average kappa values across observers were 0.52 (95 % CI: 0.45-0.60), 0.51 (0.41-0.61), and 0.57 (0.49-0.64) for CO-RADS, COVID-RADS, and RSNA expert consensus statement, respectively. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were the highest at 0.71, 0.53, 0.72, 0.96, and 0.56 in the CO-RADS; 0.56, 0.31, 0.54, 0.95, and 0.35 in the COVID-RADS; 0.83, 0.49, 0.61, 0.96, and 0.55 in the RSNA expert consensus statement, respectively. CONCLUSIONS: The CO-RADS exhibited the highest specificity, positive predictive value, which are especially important in a low-prevalence population, while maintaining high accuracy and negative predictive value, demonstrating the best performance in a low-prevalence population.

The role of chest imaging in the diagnosis, management, and monitoring of coronavirus disease 2019 (COVID-19).

Coronavirus disease 2019 (COVID-19) pandemic has posed a major public health crisis all over the world. The role of chest imaging, especially computed tomography (CT), has evolved during the pandemic paralleling the accumulation of scientific evidence. In the early stage of the pandemic, the performance of chest imaging for COVID-19 has widely been debated especially in the context of comparison to real-time reverse transcription polymerase chain reaction. Current evidence is against the use of chest imaging for routine screening of COVID-19 contrary to the initial expectations. It still has an integral role to play, however, in its work up and staging, especially when assessing complications or disease progression. Chest CT is gold standard imaging modality for COVID-19 pneumonia; in some situations, chest X-ray or ultrasound may be an effective alternative. The most important role of radiologists in this context is to be able to identify those patients at greatest risk of imminent clinical decompensation by learning to stratify cases of COVID-19 on the basis of radiologic imaging in the most efficient and timely fashion possible. The present availability of multiple and more refined CT grading systems and classification is now making this task easier and thereby contributing to the recent improvements achieved in COVID-19 treatment and outcomes. In this article, evidence of chest imaging regarding diagnosis, management and monitoring of COVID-19 will be chronologically reviewed.

Diagnostic prediction of COVID-19 based on clinical and radiological findings in a relatively low COVID-19 prevalence area.

BACKGROUND: Distinguishing coronavirus disease 2019 (COVID-19) pneumonia from other lung diseases is often difficult, especially in a highly comorbid patient population in a low prevalence region. We aimed to distinguish clinical data and computed tomography (CT) images between COVID-19 and other lung diseases in an advanced care hospital. METHODS: We assessed clinical characteristics, laboratory data, and chest CT images of patients with COVID-19 and non-COVID-19 patients who were suspected of having COVID-19 between February 20 and May 21, 2020, at the University of Tokyo Hospital. RESULTS: Typical appearance for COVID-19 on CT images were found in 24 of 29 COVID-19 cases and 21 of 168 non-COVID-19 cases, according to the Radiological Society of North America Expert Consensus Statement (for predicting COVID-19, sensitivity 0.828, specificity 0.875, positive predictive value 0.533, negative predictive value 0.967). When we focused on cases with typical CT images, loss of taste or smell, and close contact with COVID-19 patients were exclusive characteristics for the COVID-19 cases. Among laboratory data, high fibrinogen (P < 0.01) and low white blood cell count (P < 0.01) were good predictors for COVID-19 with typical CT images in multivariate analysis. CONCLUSIONS: In a relatively low prevalence region, CT screening has high sensitivity to COVID-19 in patients with suspected symptoms. When chest CT findings are typical for COVID-19, close contact, loss of taste or smell, lower white blood cell count, and higher fibrinogen are good predictors for COVID-19.

Erratum: Chest CT Findings in Cases from the Cruise Ship "Diamond Princess" with Coronavirus Disease 2019 (COVID-19).

[This corrects the article DOI: 10.1148/ryct.2020200110.].

Comparison of Chest CT Grading Systems in Coronavirus Disease 2019 (COVID-19) Pneumonia.

PURPOSE: To compare the performance and interobserver agreement of the COVID-19 Reporting and Data System (CO-RADS), the COVID-19 imaging reporting and data system (COVID-RADS), the RSNA expert consensus statement, and the British Society of Thoracic Imaging (BSTI) guidance statement. MATERIALS AND METHODS: In this case-control study, total of 100 symptomatic patients suspected of having COVID-19 were included: 50 patients with COVID-19 (59±17 years, 38 men) and 50 patients without COVID-19 (65±24 years, 30 men). Eight radiologists independently scored chest CT images of the cohort according to each reporting system. The area under the receiver operating characteristic curves (AUC) and interobserver agreements were calculated and statistically compared across the systems. RESULTS: A total of 800 observations were made for each system. The level of suspicion of COVID-19 correlated with the RT-PCR positive rate except for the "negative for pneumonia" classifications in all the systems (Spearman's coefficient: ρ=1.0, P=<.001 for all the systems). Average AUCs were as follows: CO-RADS, 0.84 (95% confidence interval, 0.83-0.85): COVID-RADS, 0.80 (0.78-0.81): the RSNA statement, 0.81 (0.79-0.82): and the BSTI statement, 0.84 (0.812-0.86). Average Cohen's kappa across observers was 0.62 (95% confidence interval, 0.58-0.66), 0.63 (0.58-0.68), 0.63 (0.57-0.69), and 0.61 (0.58-0.64) for CO-RADS, COVID-RADS, the RSNA statement and the BSTI statement, respectively. CO-RADS and the BSTI statement outperformed COVID-RADS and the RSNA statement in diagnostic performance (P=.<.05 for all the comparison). CONCLUSIONS: CO-RADS, COVID-RADS, the RSNA statement and the BSTI statement provided reasonable performances and interobserver agreements in reporting CT findings of COVID-19.

Chest CT Findings in Cases from the Cruise Ship Diamond Princess with Coronavirus Disease (COVID-19).

PURPOSE: To evaluate the chest CT findings in an environmentally homogeneous cohort from the cruise ship Diamond Princess with coronavirus disease 2019 (COVID-19). MATERIALS AND METHODS: This retrospective study comprised 104 cases (mean age, 62 years ± 16 [standard deviation], range, 25-93 years) with COVID-19 confirmed with reverse-transcription polymerase change reaction findings. CT images were reviewed, and the CT severity score was calculated for each lobe and the entire lung. CT findings were compared between asymptomatic and symptomatic cases. RESULTS: Of 104 cases, 76 (73%) were asymptomatic, 41 (54%) of which had lung opacities on CT. Twenty-eight (27%) cases were symptomatic, 22 (79%) of which had abnormal CT findings. Symptomatic cases showed lung opacities and airway abnormalities on CT more frequently than asymptomatic cases [lung opacity; 22 (79%) vs 41 (54%), airway abnormalities; 14 (50%) vs 15 (20%)]. Asymptomatic cases showed more ground-glass opacity (GGO) over consolidation (83%), while symptomatic cases more frequently showed consolidation over GGO (41%). The CT severity score was higher in symptomatic cases than asymptomatic cases, particularly in the lower lobes [symptomatic vs asymptomatic cases; right lower lobe: 2 ± 1 (0-4) vs 1 ± 1 (0-4); left lower lobe: 2 ± 1 (0-4) vs 1 ± 1 (0-3); total score: 7 ± 5 (1-17) vs 4 ± 2 (1-11)]. CONCLUSION: This study documented a high incidence of subclinical CT changes in cases with COVID-19. Compared with symptomatic cases, asymptomatic cases showed more GGO over consolidation and milder extension of disease on CT.An earlier incorrect version appeared online. This article was corrected on April 8, 2020.© RSNA, 2020.