Large-scale screening of COVID-19 from community acquired pneumonia using infection size-aware classification.

The worldwide spread of coronavirus disease (COVID-19) has become a threatening risk for global public health. It is of great importance to rapidly and accurately screen patients with COVID-19 from community acquired pneumonia (CAP). In this study, a total of 1658 patients with COVID-19 and 1027 CAP patients underwent thin-section CT. All images were preprocessed to obtain the segmentations of infections and lung fields. A set of handcrafted location-specific features was proposed to best capture the COVID-19 distribution pattern, in comparison to conventional CT severity score (CT-SS) and Radiomics features. An infection Size Aware Random Forest method (iSARF) was used for classification. Experimental results show that the proposed method yielded best performance when using the handcrafted features with sensitivity of 91.6%, specificity of 86.8%, and accuracy of 89.8% over state-of-the-art classifiers. Additional test on 734 subjects with thick slice images demonstrates great generalizability. It is anticipated that our proposed framework could assist clinical decision making. Furthermore, the data of extracted features will be made available after the review process.

Adaptive Feature Selection Guided Deep Forest for COVID-19 Classification With Chest CT.

Chest computed tomography (CT) becomes an effective tool to assist the diagnosis of coronavirus disease-19 (COVID-19). Due to the outbreak of COVID-19 worldwide, using the computed-aided diagnosis technique for COVID-19 classification based on CT images could largely alleviate the burden of clinicians. In this paper, we propose an Adaptive Feature Selection guided Deep Forest (AFS-DF) for COVID-19 classification based on chest CT images. Specifically, we first extract location-specific features from CT images. Then, in order to capture the high-level representation of these features with the relatively small-scale data, we leverage a deep forest model to learn high-level representation of the features. Moreover, we propose a feature selection method based on the trained deep forest model to reduce the redundancy of features, where the feature selection could be adaptively incorporated with the COVID-19 classification model. We evaluated our proposed AFS-DF on COVID-19 dataset with 1495 patients of COVID-19 and 1027 patients of community acquired pneumonia (CAP). The accuracy (ACC), sensitivity (SEN), specificity (SPE), AUC, precision and F1-score achieved by our method are 91.79%, 93.05%, 89.95%, 96.35%, 93.10% and 93.07%, respectively. Experimental results on the COVID-19 dataset suggest that the proposed AFS-DF achieves superior performance in COVID-19 vs. CAP classification, compared with 4 widely used machine learning methods.

Correlation of Chest CT and RT-PCR Testing for Coronavirus Disease 2019 (COVID-19) in China: A Report of 1014 Cases.

Background Chest CT is used in the diagnosis of coronavirus disease 2019 (COVID-19) and is an important complement to reverse-transcription polymerase chain reaction (RT-PCR) tests. Purpose To investigate the diagnostic value and consistency of chest CT as compared with RT-PCR assay in COVID-19. Materials and Methods This study included 1014 patients in Wuhan, China, who underwent both chest CT and RT-PCR tests between January 6 and February 6, 2020. With use of RT-PCR as the reference standard, the performance of chest CT in the diagnosis of COVID-19 was assessed. In addition, for patients with multiple RT-PCR assays, the dynamic conversion of RT-PCR results (negative to positive, positive to negative) was analyzed as compared with serial chest CT scans for those with a time interval between RT-PCR tests of 4 days or more. Results Of the 1014 patients, 601 of 1014 (59%) had positive RT-PCR results and 888 of 1014 (88%) had positive chest CT scans. The sensitivity of chest CT in suggesting COVID-19 was 97% (95% confidence interval: 95%, 98%; 580 of 601 patients) based on positive RT-PCR results. In the 413 patients with negative RT-PCR results, 308 of 413 (75%) had positive chest CT findings. Of those 308 patients, 48% (103 of 308) were considered as highly likely cases and 33% (103 of 308) as probable cases. At analysis of serial RT-PCR assays and CT scans, the mean interval between the initial negative to positive RT-PCR results was 5.1 days ± 1.5; the mean interval between initial positive to subsequent negative RT-PCR results was 6.9 days ± 2.3. Of the 1014 patients, 60% (34 of 57) to 93% (14 of 15) had initial positive CT scans consistent with COVID-19 before (or parallel to) the initial positive RT-PCR results. Twenty-four of 57 patients (42%) showed improvement on follow-up chest CT scans before the RT-PCR results turned negative. Conclusion Chest CT has a high sensitivity for diagnosis of coronavirus disease 2019 (COVID-19). Chest CT may be considered as a primary tool for the current COVID-19 detection in epidemic areas. © RSNA, 2020 Online supplemental material is available for this article. A translation of this abstract in Farsi is available in the supplement. ترجمه چکیده این مقاله به فارسی، در ضمیمه موجود است.

Cardiac involvement in patients 1 year after recovery from moderate and severe COVID-19 infections.

Background: Some patients suffered persistent cardiac symptoms after hospital discharge following COVID-19 infection, including chest tightness, chest pain, and palpitation. However, the cardiac involvement in these patients remains unknown. The purpose of this study was to investigate the effect of COVID-19 infection on the cardiovascular system after 1 year of recovery in patients hospitalized with persistent cardiac symptoms. Materials and methods: In this prospective observational study, a total of 32 patients who had COVID-19 (11 diagnosed as severe COVID-19 and 21 as moderate) with persistent cardiac symptoms after hospital discharge were enrolled. Contrast-enhanced cardiovascular magnetic resonance (CMR) imaging was performed on all patients. Comparisons were made with age- and sex-matched healthy controls (n = 13), and age-, sex- and risk factor-matched controls (n = 21). Further analysis was made between the severe and moderate COVID-19 cohorts. Results: The mean time interval between acute COVID-19 infection and CMR was 462 ± 18 days. Patients recovered from COVID-19 had reduced left ventricular ejection fraction (LVEF) (p = 0.003) and increased extracellular volumes (ECVs) (p = 0.023) compared with healthy controls. Focal late gadolinium enhancement (LGE) was found in 22 (68.8%) patients, mainly distributed linearly in the septal mid-wall or patchily in RV insertion point. The LGE extent in patients with severe COVID-19 was higher than that in patients with moderate COVID-19 (p = 0.009). Conclusion: This 1-year follow-up study revealed that patients with persistent cardiac symptoms, after recovering from COVID-19, had decreased cardiac function and increased ECV compared with healthy controls. Patients with COVID-19 predominately had a LGE pattern of septal mid-wall or RV insertion point. Patients with severe COVID-19 had greater LGE extent than patients with moderate COVID-19.

Relationship between Lung and Brain Injury in COVID-19 Patients: A Hyperpolarized 129Xe-MRI-based 8-Month Follow-Up.

Although the lungs are the primary organ involved, increasing evidence supports the neuroinvasive potential of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This study investigates the potential relationship between coronavirus disease (COVID-19)-related deterioration of brain structure and the degree of damage to lung function. Nine COVID-19 patients were recruited in critical condition from Jin Yin-tan Hospital (Wuhan, China) who had been discharged between 4 February and 27 February 2020. The demographic, clinical, treatment, and laboratory data were extracted from the electronic medical records. All patients underwent chest CT imaging, 129Xe gas lung MRI, and 1H brain MRI. Four of the patients were followed up for 8 months. After nearly 12 months of recovery, we found no significant difference in lung ventilation defect percentage (VDP) between the COVID-19 group and the healthy group (3.8 ± 2.1% versus 3.7 ± 2.2%) using 129Xe MRI, and several lung-function-related parameters-such as gas-blood exchange time (T)-showed improvement (42.2 ms versus 32.5 ms). Combined with 1H brain MRI, we found that the change in gray matter volume (GMV) was strongly related to the degree of pulmonary function recovery-the greater the increase in GMV, the higher degree of pulmonary function damage.

Relationship between Lung and Brain Injury in COVID-19 Patients: A Hyperpolarized 129Xe-MRI-based 8-Month Follow-Up

Although the lungs are the primary organ involved, increasing evidence supports the neuroinvasive potential of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This study investigates the potential relationship between coronavirus disease (COVID-19)-related deterioration of brain structure and the degree of damage to lung function. Nine COVID-19 patients were recruited in critical condition from Jin Yin-tan Hospital (Wuhan, China) who had been discharged between 4 February and 27 February 2020. The demographic, clinical, treatment, and laboratory data were extracted from the electronic medical records. All patients underwent chest CT imaging, 129Xe gas lung MRI, and 1H brain MRI. Four of the patients were followed up for 8 months. After nearly 12 months of recovery, we found no significant difference in lung ventilation defect percentage (VDP) between the COVID-19 group and the healthy group (3.8 ±2.1% versus 3.7 ±2.2%) using 129Xe MRI, and several lung-function-related parameters-such as gas–blood exchange time (T)-showed improvement (42.2 ms versus 32.5 ms). Combined with 1H brain MRI, we found that the change in gray matter volume (GMV) was strongly related to the degree of pulmonary function recovery-the greater the increase in GMV, the higher degree of pulmonary function damage.

Abnormal dynamic ventilation function of COVID-19 survivors detected by pulmonary free-breathing proton MRI.

OBJECTIVES: To visualize and quantitatively assess regional lung function of survivors of COVID-19 who were hospitalized using pulmonary free-breathing 1H MRI. METHODS: A total of 12 healthy volunteers and 27 COVID-19 survivors (62.4 ± 8.1 days between infection and image acquisition) were recruited in this prospective study and performed chest 1H MRI acquisitions with free tidal breathing. Then, conventional Fourier decomposition ventilation (FD-V) and global fractional ventilation (FVGlobal) were analyzed. Besides, a modified PREFUL (mPREFUL) method was developed to adapt to COVID-19 survivors and generate dynamic ventilation maps and parameters. All the ventilation maps and parameters were analyzed using Student's t-test. Pearson's correlation and a Bland-Altman plot between FVGlobal and mPREFUL were analyzed. RESULTS: There was no significant difference between COVID-19 and healthy groups regarding a static FD-V map (0.47 ± 0.12 vs 0.42 ± 0.08; p = .233). However, mPREFUL demonstrated lots of regional high ventilation areas (high ventilation percentage (HVP): 23.7% ± 10.6%) existed in survivors. This regional heterogeneity (i.e., HVP) in survivors was significantly higher than in healthy volunteers (p = .003). The survivors breathed deeper (flow-volume loop: 5375 ± 3978 vs 1688 ± 789; p = .005), and breathed more air in respiratory cycle (total amount: 62.6 ± 19.3 vs 37.3 ± 9.9; p < .001). Besides, mPREFUL showed both good Pearson's correlation (r = 0.74; p < .001) and Bland-Altman consistency (mean bias = -0.01) with FVGlobal. CONCLUSIONS: Dynamic ventilation imaging using pulmonary free-breathing 1H MRI found regional abnormity of dynamic ventilation function in COVID-19 survivors. KEY POINTS: • Pulmonary free-breathing1H MRI was used to visualize and quantitatively assess regional lung ventilation function of COVID-19 survivors. • Dynamic ventilation maps generated from 1H MRI were more sensitive to distinguish the COVID-19 and healthy groups (total air amount: 62.6 ± 19.3 vs 37.3 ± 9.9; p < .001), compared with static ventilation maps (FD-V value: 0.47 ± 0.12 vs 0.42 ± 0.08; p = .233). • COVID-19 survivors had larger regional heterogeneity (high ventilation percentage: 23.7% ± 10.6% vs 13.1% ± 7.9%; p = .003), and breathed deeper (flow-volume loop: 5375 ± 3978 vs 1688 ± 789; p = .005) than healthy volunteers.

Reversible Bronchiectasis in COVID-19 Survivors With Acute Respiratory Distress Syndrome: Pseudobronchiectasis.

To retrospectively analyze whether traction bronchiectasis was reversible in coronavirus disease 2019 (COVID-19) survivors with acute respiratory distress syndrome (ARDS), and whether computed tomography (CT) findings were associated with the reversibility, 41 COVID-19 survivors with ARDS were followed-up for more than 4 months. Demographics, clinical data, and all chest CT images were collected. The follow-up CT images were compared with the previous CT scans. There were 28 (68%) patients with traction bronchiectasis (Group I) and 13 (32%) patients without traction bronchiectasis (Group II) on CT images. Traction bronchiectasis disappeared completely in 21 of the 28 (75%) patients (Group IA), but did not completely disappear in seven of the 28 (25%) patients (Group IB). In the second week after onset, the evaluation score on CT images in Group I was significantly higher than that in Group II (p = 0.001). The proportion of reticulation on the last CT images in Group IB was found higher than that in Group IA (p < 0.05). COVID-19 survivors with ARDS might develop traction bronchiectasis, which can be absorbed completely in most patients. Traction bronchiectasis in a few patients did not disappear completely, but bronchiectasis was significantly relieved. The long-term follow-up is necessary to further assess whether traction bronchiectasis represents irreversible fibrosis.

CT Features of Coronavirus Disease 2019 (COVID-19) Pneumonia in 62 Patients in Wuhan, China.

OBJECTIVE. The purpose of this study was to investigate 62 subjects in Wuhan, China, with laboratory-confirmed coronavirus disease (COVID-19) pneumonia and describe the CT features of this epidemic disease. MATERIALS AND METHODS. A retrospective study of 62 consecutive patients with laboratory-confirmed COVID-19 pneumonia was performed. CT images and clinical data were reviewed. Two thoracic radiologists evaluated the distribution and CT signs of the lesions and also scored the extent of involvement of the CT signs. The Mann-Whitney U test was used to compare lesion distribution and CT scores. The chi-square test was used to compare the CT signs of early-phase versus advanced-phase COVID-19 pneumonia. RESULTS. A total of 62 patients (39 men and 23 women; mean [± SD] age, 52.8 ± 12.2 years; range, 30-77 years) with COVID-19 pneumonia were evaluated. Twenty-four of 30 patients who underwent routine blood tests (80.0%) had a decreased lymphocyte count. Of 27 patients who had their erythrocyte sedimentation rate and high-sensitivity C-reactive protein level assessed, 18 (66.7%) had an increased erythrocyte sedimentation rate, and all 27 (100.0%) had an elevated high-sensitivity C-reactive protein level. Multiple lesions were seen on the initial CT scan of 52 of 62 patients (83.9%). Forty-eight of 62 patients (77.4%) had predominantly peripheral distribution of lesions. The mean CT score for the upper zone (3.0 ± 3.4) was significantly lower than that for the middle (4.5 ± 3.8) and lower (4.5 ± 3.7) zones (p = 0.022 and p = 0.020, respectively), and there was no significant difference in the mean CT score of the middle and lower zones (p = 1.00). The mean CT score for the anterior area (4.4 ± 4.1) was significantly lower than that for the posterior area (7.7 ± 6.3) (p = 0.003). CT findings for the patients were as follows: 25 patients (40.3%) had ground-glass opacities (GGO), 21 (33.9%), consolidation; 39 (62.9%), GGO plus a reticular pattern; 34 (54.8%), vacuolar sign; 28 (45.2%), microvascular dilation sign; 35 (56.5%), fibrotic streaks; 21 (33.9%), a subpleural line; and 33 (53.2%), a subpleural transparent line. With regard to bronchial changes seen on CT, 45 patients (72.6%) had air bronchogram, and 11 (17.7%) had bronchus distortion. In terms of pleural changes, CT showed that 30 patients (48.4%) had pleural thickening, 35 (56.5%) had pleural retraction sign, and six (9.7%) had pleural effusion. Compared with early-phase disease (≤ 7 days after the onset of symptoms), advanced-phase disease (8-14 days after the onset of symptoms) was characterized by significantly increased frequencies of GGO plus a reticular pattern, vacuolar sign, fibrotic streaks, a subpleural line, a subpleural transparent line, air bronchogram, bronchus distortion, and pleural effusion; however, GGO significantly decreased in advanced-phase disease. CONCLUSION. CT examination of patients with COVID-19 pneumonia showed a mixed and diverse pattern with both lung parenchyma and the interstitium involved. Identification of GGO and a single lesion on the initial CT scan suggested early-phase disease. CT signs of aggravation and repair coexisted in advanced-phase disease. Lesions presented with a characteristic multifocal distribution in the middle and lower lung regions and in the posterior lung area. A decreased lymphocyte count and an increased high-sensitivity C-reactive protein level were the most common laboratory findings.

Long-term chest CT follow-up in COVID-19 Survivors: 102-361 days after onset.

BACKGROUND: The aim of this study was to evaluate long-term longitudinal changes in chest computed tomography (CT) findings in coronavirus disease 2019 (COVID-19) survivors and their correlations with dyspnea after discharge. METHODS: A total of 337 COVID-19 survivors who underwent CT scan during hospitalization and between 102 and 361 days after onset were retrospectively included. Subjective CT findings, lesion volume, therapeutic measures and laboratory parameters were collected. The severity of the survivors' dyspnea was determined by follow-up questionnaire. The evolution of the CT findings from the peak period to discharge and throughout follow-up and the abilities of CT findings and clinical parameters to predict survival with and without dyspnea were analyzed. RESULTS: Ninety-one COVID-19 survivors still had dyspnea at follow-up. The age, comorbidity score, duration of hospital stays, receipt of hormone administration, receipt of immunoglobulin injections, intensive care unit (ICU) admission, receipt of mechanical ventilation, laboratory parameters, clinical classifications and parameters associated with lesion volume of the survivors with dyspnea were significantly different from those of survivors without dyspnea. Among the clinical parameters and CT parameters used to identify dyspnea, parameters associated with lesion volume showed the largest area under the curve (AUC) values, with lesion volume at discharge showing the largest AUC (0.820). Lesion volume decreased gradually from the peak period to discharge and through follow-up, with a notable decrease observed after discharge. Absorption of lesions continued 6 months after discharge. CONCLUSIONS: Among the clinical parameters and subjective CT findings, CT findings associated with lesion volume were the best predictors of post-discharge dyspnea in COVID-19 survivors.

Early prediction of lung lesion progression in COVID-19 patients with extended CT ventilation imaging.

PURPOSE: In the prediction of COVID-19 disease progression, a clear illustration and early determination of an area that will be affected by pneumonia remain great challenges. In this study, we aimed to predict and visualize the progression of lung lesions in COVID-19 patients in the early stage of illness by using chest CT. METHODS: COVID-19 patients who underwent three chest CT scans in the progressive phase were retrospectively enrolled. An extended CT ventilation imaging (CTVI) method was proposed in this work that was adapted to use two chest CT scans acquired on different days, and then lung ventilation maps were generated. The prediction maps were obtained according to the fractional ventilation values, which were related to pulmonary regional function and tissue property changes. The third CT scan was used to validate whether the prediction maps could be used to distinguish healthy regions and potential lesions. RESULTS: A total of 30 patients (mean age ± SD, 43 ± 10 years, 19 females, and 2-12 days between the second and third CT scans) were included in this study. The predicted lesion locations and sizes were almost the same as the true ones visualized in third CT scan. Quantitatively, the predicted lesion volumes and true lesion volumes showed both a good Pearson correlation (R2 = 0.80; P < 0.001) and good consistency in the Bland-Altman plot (mean bias = 0.04 cm3). Regarding the enlargements of the existing lesions, prediction results also exhibited a good Pearson correlation (R2 = 0.76; P < 0.001) with true lesion enlargements. CONCLUSION: The present findings demonstrated that the extended CTVI method could accurately predict and visualize the progression of lung lesions in COVID-19 patients in the early stage of illness, which is helpful for physicians to predetermine the severity of COVID-19 pneumonia and make effective treatment plans in advance.

Deep learning for predicting COVID-19 malignant progression.

As COVID-19 is highly infectious, many patients can simultaneously flood into hospitals for diagnosis and treatment, which has greatly challenged public medical systems. Treatment priority is often determined by the symptom severity based on first assessment. However, clinical observation suggests that some patients with mild symptoms may quickly deteriorate. Hence, it is crucial to identify patient early deterioration to optimize treatment strategy. To this end, we develop an early-warning system with deep learning techniques to predict COVID-19 malignant progression. Our method leverages CT scans and the clinical data of outpatients and achieves an AUC of 0.920 in the single-center study. We also propose a domain adaptation approach to improve the generalization of our model and achieve an average AUC of 0.874 in the multicenter study. Moreover, our model automatically identifies crucial indicators that contribute to the malignant progression, including Troponin, Brain natriuretic peptide, White cell count, Aspartate aminotransferase, Creatinine, and Hypersensitive C-reactive protein.

Serial Quantitative Chest CT Assessment of COVID-19: A Deep Learning Approach.

PURPOSE: To quantitatively evaluate lung burden changes in patients with coronavirus disease 2019 (COVID-19) by using serial CT scan by an automated deep learning method. MATERIALS AND METHODS: Patients with COVID-19, who underwent chest CT between January 1 and February 3, 2020, were retrospectively evaluated. The patients were divided into mild, moderate, severe, and critical types, according to their baseline clinical, laboratory, and CT findings. CT lung opacification percentages of the whole lung and five lobes were automatically quantified by a commercial deep learning software and compared with those at follow-up CT scans. Longitudinal changes of the CT quantitative parameter were also compared among the four clinical types. RESULTS: A total of 126 patients with COVID-19 (mean age, 52 years ± 15 [standard deviation]; 53.2% males) were evaluated, including six mild, 94 moderate, 20 severe, and six critical cases. CT-derived opacification percentage was significantly different among clinical groups at baseline, gradually progressing from mild to critical type (all P < .01). Overall, the whole-lung opacification percentage significantly increased from baseline CT to first follow-up CT (median [interquartile range]: 3.6% [0.5%, 12.1%] vs 8.7% [2.7%, 21.2%]; P < .01). No significant progression of the opacification percentages was noted from the first follow-up to second follow-up CT (8.7% [2.7%, 21.2%] vs 6.0% [1.9%, 24.3%]; P = .655). CONCLUSION: The quantification of lung opacification in COVID-19 measured at chest CT by using a commercially available deep learning-based tool was significantly different among groups with different clinical severity. This approach could potentially eliminate the subjectivity in the initial assessment and follow-up of pulmonary findings in COVID-19.Supplemental material is available for this article.© RSNA, 2020.

Computing infection distributions and longitudinal evolution patterns in lung CT images.

BACKGROUND: Spatial and temporal lung infection distributions of coronavirus disease 2019 (COVID-19) and their changes could reveal important patterns to better understand the disease and its time course. This paper presents a pipeline to analyze statistically these patterns by automatically segmenting the infection regions and registering them onto a common template. METHODS: A VB-Net is designed to automatically segment infection regions in CT images. After training and validating the model, we segmented all the CT images in the study. The segmentation results are then warped onto a pre-defined template CT image using deformable registration based on lung fields. Then, the spatial distributions of infection regions and those during the course of the disease are calculated at the voxel level. Visualization and quantitative comparison can be performed between different groups. We compared the distribution maps between COVID-19 and community acquired pneumonia (CAP), between severe and critical COVID-19, and across the time course of the disease. RESULTS: For the performance of infection segmentation, comparing the segmentation results with manually annotated ground-truth, the average Dice is 91.6% ± 10.0%, which is close to the inter-rater difference between two radiologists (the Dice is 96.1% ± 3.5%). The distribution map of infection regions shows that high probability regions are in the peripheral subpleural (up to 35.1% in probability). COVID-19 GGO lesions are more widely spread than consolidations, and the latter are located more peripherally. Onset images of severe COVID-19 (inpatients) show similar lesion distributions but with smaller areas of significant difference in the right lower lobe compared to critical COVID-19 (intensive care unit patients). About the disease course, critical COVID-19 patients showed four subsequent patterns (progression, absorption, enlargement, and further absorption) in our collected dataset, with remarkable concurrent HU patterns for GGO and consolidations. CONCLUSIONS: By segmenting the infection regions with a VB-Net and registering all the CT images and the segmentation results onto a template, spatial distribution patterns of infections can be computed automatically. The algorithm provides an effective tool to visualize and quantify the spatial patterns of lung infection diseases and their changes during the disease course. Our results demonstrate different patterns between COVID-19 and CAP, between severe and critical COVID-19, as well as four subsequent disease course patterns of the severe COVID-19 patients studied, with remarkable concurrent HU patterns for GGO and consolidations.

Early fibroproliferative signs on high-resolution CT are associated with mortality in COVID-19 pneumonia patients with ARDS: a retrospective study.

OBJECTIVES: To investigate the chest high-resolution computed tomography (HRCT) findings in coronavirus disease 2019 (COVID-19) pneumonia patients with acute respiratory distress syndrome (ARDS) and to evaluate its relationship with clinical outcome. MATERIALS AND METHODS: In this retrospective study, 79 COVID-19 patients with ARDS were recruited. Clinical data were extracted from electronic medical records and analyzed. HRCT scans, obtained within 3 days before clinical ARDS onset, were evaluated by three independent observers and graded into six findings according to the extent of fibroproliferation. Multivariable Cox proportional hazard regression analysis was used to assess the independent predictive value of the computed tomography (CT) score and radiological fibroproliferation. Patient survival was determined by Kaplan-Meier analysis. RESULTS: Compared with survivors, non-survivors showed higher rates of lung fibroproliferation, whereas there were no significant differences in the area of increased attenuation without traction bronchiolectasis or bronchiectasis. A HRCT score <230 enabled the prediction of survival with 73.5% sensitivity and 93.3% specificity, 100% negative predictive value (NPP), 83.3% positive predictive value (PPV) and 88.6% accuracy (Area Under the Curve [AUC] = 0.9; 95% confidence Interval [CI] 0.831-0.968). A multivariate Cox proportional hazards model showed that the HRCT score is a significant independent risk factor for mortality (Hazard Ratio [HR] 9.94; 95% CI 4.10-24.12). Kaplan-Meier analysis revealed that a HRCT score ⩾230 was associated with a higher fatality rate. Organ injury occurred less frequently in patients with a HRCT score <230 compared to those with a HRCT score ⩾230. CONCLUSION: Early pulmonary fibroproliferative signs on HRCT are associated with increased mortality and susceptibility to organ injury in COVID-19 pneumonia patients with early ARDS.

A rapid screening classifier for diagnosing COVID-19.

Rationale: Coronavirus disease 2019 (COVID-19) has caused a global pandemic. A classifier combining chest X-ray (CXR) with clinical features may serve as a rapid screening approach. Methods: The study included 512 patients with COVID-19 and 106 with influenza A/B pneumonia. A deep neural network (DNN) was applied, and deep features derived from CXR and clinical findings formed fused features for diagnosis prediction. Results: The clinical features of COVID-19 and influenza showed different patterns. Patients with COVID-19 experienced less fever, more diarrhea, and more salient hypercoagulability. Classifiers constructed using the clinical features or CXR had an area under the receiver operating curve (AUC) of 0.909 and 0.919, respectively. The diagnostic efficacy of the classifier combining the clinical features and CXR was dramatically improved and the AUC was 0.952 with 91.5% sensitivity and 81.2% specificity. Moreover, combined classifier was functional in both severe and non-serve COVID-19, with an AUC of 0.971 with 96.9% sensitivity in non-severe cases, which was on par with the computed tomography (CT)-based classifier, but had relatively inferior efficacy in severe cases compared to CT. In extension, we performed a reader study involving three experienced pulmonary physicians, artificial intelligence (AI) system demonstrated superiority in turn-around time and diagnostic accuracy compared with experienced pulmonary physicians. Conclusions: The classifier constructed using clinical and CXR features is efficient, economical, and radiation safe for distinguishing COVID-19 from influenza A/B pneumonia, serving as an ideal rapid screening tool during the COVID-19 pandemic.

Damaged lung gas exchange function of discharged COVID-19 patients detected by hyperpolarized 129Xe MRI.

The recovery process of COVID-19 patients is unclear. Some recovered patients complain of continued shortness of breath. Vasculopathy has been reported in COVID-19, stressing the importance of probing pulmonary microstructure and function at the alveolar-capillary interface. While computed tomography (CT) detects structural abnormalities, little is known about the impact of disease on lung function. 129Xe magnetic resonance imaging (MRI) is a technique uniquely capable of assessing ventilation, microstructure, and gas exchange. Using 129Xe MRI, we found that COVID-19 patients show a higher rate of ventilation defects (5.9% versus 3.7%), unchanged microstructure, and longer gas-blood exchange time (43.5 ms versus 32.5 ms) compared with healthy individuals. These findings suggest that regional ventilation and alveolar airspace dimensions are relatively normal around the time of discharge, while gas-blood exchange function is diminished. This study establishes the feasibility of localized lung function measurements in COVID-19 patients and their potential usefulness as a supplement to structural imaging.

Lung volume reduction and infection localization revealed in Big data CT imaging of COVID-19.

The ongoing worldwide COVID-19 pandemic has become a huge threat to global public health. Using CT image, 3389 COVID-19 patients, 1593 community-acquired pneumonia (CAP) patients, and 1707 nonpneumonia subjects were included to explore the different patterns of lung and lung infection. We found that COVID-19 patients have a significant reduced lung volume with increased density and mass, and the infections tend to present as bilateral lower lobes. The findings provide imaging evidence to improve our understanding of COVID-19.

Mobile chest X-ray manifestations of 54 deceased patients with coronavirus disease 2019: Retrospective study.

To describe the mobile chest X-ray manifestations of deceased patients with coronavirus disease 2019 (COVID-19).In this retrospective study, we analyzed in patients with COVID-19 from Tongji Hospital (Wuhan, China), who had been died between February 18 and March 25, 2020. Two radiologists analyzed the radiologic characteristics of mobile chest X-ray, and analyzed the serial X-ray changes.Fifty-four deceased patients with COVID-19 were included in the study. We found that 50 (93%) patients with lesions occurred in the bilateral lung, 4 (7%) patients occurred in the right lung, 54 (100%) patients were multifocal involvement. The number of lung fields involved was 42 (78%) patients in 6 fields, 3 (6%) patients in 5 lung fields, 4 (7%) patients in 4 lung fields, and 5 (9%) patients in 3 lung fields. Fifty-three (98%) patients had patchy opacities, 3 (6%) patients had round or oval solid nodules, 9 (17%) patients had fibrous stripes, 13 (24%) patients had pleural effusion, 8 (15%) patients had pleural thickening, 6 (11%) patients had pneumothorax, 3 (6%) patients had subcutaneous emphysema. Among the 24 patients who had serial mobile chest X-rays, 16 (67%) patients had the progression of the lesions, 8 (33%) patients had no significant change of the lesions, and there was no case of reduction of the lesions.The mobile chest X-ray manifestations of deceased patients with COVID-19 were mostly bilateral lung, multifocal involvement, and extensive lung field, and pleural effusion, pleural thickening, and pneumothorax probably could be observed. The serial mobile chest X-ray showed that the chest lesions were progressive with a high probability.

Chest CT imaging features and severity scores as biomarkers for prognostic prediction in patients with COVID-19.

BACKGROUND: Coronavirus disease 2019 (COVID-19) has become a pandemic. Few studies have explored the role of chest computed tomography (CT) features and severity scores for prognostic prediction. In this study, we aimed to investigate the role of chest CT severity score and imaging features in the prediction of the prognosis of COVID-19 patients. METHODS: A total of 134 patients (62 recovered and 72 deceased patients) with confirmed COVID-19 were enrolled. The clinical, laboratory, and chest CT (316 scans) data were retrospectively reviewed. Demographics, symptoms, comorbidities, and temporal changes of laboratory results, CT features, and severity scores were compared between recovered and deceased groups using the Mann-Whitney U test and logistic regression to identify the risk factors for poor prognosis. RESULTS: Median age was 48 and 58 years for recovered and deceased patients, respectively. More patients had at least one comorbidity in the deceased group than the recovered group (60% vs. 29%). Leukocytes, neutrophil, high-sensitivity C-reactive protein (hsCRP), prothrombin, D-dimer, serum ferritin, interleukin (IL)-2, and IL-6 were significantly elevated in the deceased group than the recovered group at different stages. The total CT score at the peak stage was significantly greater in the deceased group than the recovered group (20 vs. 11 points). The optimal cutoff value of the total CT scores was 16.5 points, achieving 69.4% sensitivity and 82.2% specificity for the prognostic prediction. The crazy-paving pattern and consolidation were more common in the deceased patients than those in the recovered patients. Linear opacities significantly increased with the disease course in the recovered patients. Sex, age, neutrophil, IL-2, IL-6, and total CT scores were independent risk factors for the prognosis with odds ratios of 3.8 to 8.7. CONCLUSIONS: Sex (male), older age (>60 years), elevated neutrophil, IL-2, IL-6 level, and total CT scores (≥16) were independent risk factors for poor prognosis in patients with COVID-19. Temporal changes of chest CT features and severity scores could be valuable for early identification of severe cases and eventually reducing the mortality rate of COVID-19.