key: cord-0691722-sgyo1a5g
authors: Kanne, Jeffrey P.; Bai, Harrison; Bernheim, Adam; Chung, Michael; Haramati, Linda B; Kallmes, David F.; Little, Brent P.; Rubin, Geoffrey D.; Sverzellati, Nicola
title: COVID-19 Imaging: What We Know Now and What Remains Unknown
date: 2021-02-09
journal: Radiology
DOI: 10.1148/radiol.2021204522
sha: 1544a7622a79831367a6c1ee64dd95798f629c08
doc_id: 691722
cord_uid: sgyo1a5g

Infection by SARS-CoV-2 virus ranges from asymptomatic to severe and sometimes fatal disease, most frequently the result of acute lung injury. The role of imaging has evolved during the pandemic, initially with CT as alternative and possibly superior test compared to RT-PCR, to a more limited role based on specific indications. Several classification and reporting schemes were developed for chest imaging early during the pandemic for patients with suspected COVID-19 to aid in triage when the availability of RT-PCR testing was limited and its performance unclear. Interobserver agreement for categories with findings typical of COVID-19 and those suggesting an alternative diagnosis is high across multiple studies. Furthermore, some studies looking at the extent of lung involvement on chest radiography and CT showed correlations with critical illness and need for mechanical ventilation. In addition to pulmonary manifestations, cardiovascular complications such as thromboembolism and myocarditis have been ascribed to COVID-19, sometimes contributing to neurologic and abdominal manifestations. Finally, artificial intelligence has shown promise in both diagnosis and prognosis of COVID-19 pneumonia both with respect to radiography and CT.

The typical chest CT appearance of COVID-19 pneumonia is bilateral peripheral opacities with a lower lung distribution (Fig. 2) . The opacities are usually ground-glass opacity (GGO) sometimes with areas of consolidation and are often nodular or mass-like, thereby resembling an organizing pneumonia pattern (11, 12) . Additional imaging patterns resembling organizing pneumonia include a perilobular pattern of opacification and a "reverse halo" sign, defined as a focal, rounded area of GGO surrounded by a ring or arc of denser consolidation (Figs. 3 and 4) . Diffuse GGO which can mimic other infections, drug toxicities, and inhalational lung disease has also been reported (12) . Although prototypical CT features of COVID-19 pneumonia are well described, in clinical practice, many patients will have some but not all the imaging manifestations. For example, the opacities may be unilateral but have a rounded morphology. Alternatively, the opacities may have an upper lobe predominance but still retain a peripheral or subpleural distribution.

CT features which are indeterminate for COVID-19 have also been described and classified in guidelines such as the Radiological Society of North America (RSNA) consensus statement on CT reporting (12) . These include imaging findings which have been reported in COVID-19 but are not specific enough to arrive at a relatively confident radiological diagnosis. For example, diffuse or perihilar GGO with or without consolidation, or scattered non-rounded opacities can occur with a variety of other infectious and some noninfectious processes such as edema or alveolar hemorrhage (Fig. 3) . Certain CT features are uncommonly seen in COVID-19 pneumonia including lobar or segmental consolidation without GGO, discrete small pulmonary nodules, pulmonary cavitation, septal thickening, pleural effusion, and pneumothorax. Interestingly, the rate of barotrauma in mechanically ventilated patients with COVID-19 has been reported to be much more common that patients with other causes of acute respiratory distress syndrome (24% versus 11%) (13) . A comprehensive review of the various scoring and assessment systems developed for CT lung findings is discussed in detail in a later section.

While the presence of characteristic COVID-19 imaging findings is helpful in diagnosis and risk stratification, it is noteworthy that both CXR and chest CT may lack lung abnormalities in the earliest stages of infection, with rates of normal CT as high as 56% in patients imaged within two days of symptom onset (14). Therefore, a normal CXR and CT do not reliably exclude disease. Pan et al. described four temporal stages of acute and subacute COVID-19 on CT, including an initial phase where abnormalities manifest as GGO, may be unilateral, and tend to lack the characteristic peripheral lung distribution (15) . Patients often experience progression from day 5 to 8 when pulmonary opacities become more extensive and confluent with more common bilateral lung involvement (Fig. 4) . The peak stage occurs around 9 to 13 days and features more extensive consolidation, which parallels the evolution of acute lung injury (2, 14, 15) . This dovetails with investigators who have found that abnormalities on CXR are most extensive 10 to 12 days after symptom onset (8). There is variation among patients, but beginning at about two weeks, many enter the absorption stage (16) . During this period, consolidation may wane, and other manifestations absent in the earlier phases of acute infection such as linear opacities, a "reverse-halo" sign, and a "crazypaving" pattern may emerge. During the first several weeks of infection, pleural effusions are uncommon, cavitation is rare, and pulmonary fibrosis is not expected.

Over weeks, COVID-19 pulmonary findings on both CXR and CT resolve or can evolve into a more structured and organized phase, in which case GGO and consolidation transform into more reticular opacities and may be associated with fibrosis, volume loss, architectural distortion, and traction bronchiectasis.

Assessment of disease severity by imaging in COVID-19 may inform clinical decisions related to need for hospital admission, timing of intubation, patient course and prognosis, and therapeutic efficacy. CT may enable reproducible quantitative severity scoring and can be particularly helpful in detecting mild disease, characterizing longitudinal change, and assessing the extent of disease in the setting of baseline pulmonary abnormalities.

A variety of methods have been used to assess lung involvement at CT in COVID-19. Qualitative methods classify parenchymal disease as mild, moderate, or severe. Semiquantitative methods estimate lobar or zonal involvement by quartiles (0-25%, 26-49%, 50-75%, 76%-100%) (17) , with <5% lobar involvement also sometimes used (18) . Software-based quantitative methods, including those using machine learning, can be used to calculate the total lung involvement as well as the percentage of GGO and consolidation and may have higher accuracy than human semiquantitative estimates (19) (20) (21) .

Several studies have shown correlations between extent of parenchymal involvement at CT and clinical assessment of COVID-19 disease severity as defined by parameters such as severity of symptoms, oxygenation status, and certain laboratory measures of infection and inflammation. Semiquantitative and quantitative studies have shown significantly higher CT severity scores for patients with severe and critical disease than for those with less severe disease (17, (22) (23) (24) (25) . For example, in one study of 189 inpatients, the average volume of lung involvement measured by semiautomated segmentation of parenchymal opacities on CT was higher in critically ill patients (38.5%) than in non-critically ill patients (5.8%), with a threshold of 23% distinguishing these two groups with 96% sensitivity and specificity (22). In another study of 78 patients, a semiquantitative total CT severity score ranging from 0 to 20 distinguished mild, moderate, and severe clinical disease with high accuracy (82.6% sensitivity and 100% specificity for a cutoff score of 7.5) and a high interclass correlation between readers (0.976) (17) .

CT severity scores also show correlations with serum markers of disease severity. A study of 84 hospitalized patients with COVID-19 showed significant correlations of with lymphocyte count and I n p r e s s percentage, neutrophil percentage, C-reactive protein and procalcitonin levels (all p<0.05) (26) . Another semiquantitative study of 106 inpatients with COVID-19 pneumonia showed significant positive correlations between CT severity and levels of inflammatory cytokines such as interleukin-6 and interleukin-2R (27) . Additional studies have observed similar correlations (22, 28, 29) CT severity scoring may show promise for clinical triage and assessment of prognosis, and higher CT severity scores predict clinical outcomes in COVID- 19 (29, 30) . In a study of 572 hospitalized patients, 70% of patients with total lung involvement greater than 50% were admitted to the intensive care unit or died within seven days of a CT performed at admission, while these rates were lower for lung involvement of 26-50% (41%) and <25% (23%) (18) . Semiquantitative CT severity scores of 18 or greater on a scale from 0-25 correlated with increased mortality risk in another study of 130 patients in the emergency department setting (HR 3.7, p=0.0348).

Semiquantitative CT severity scores of 18 or greater on a scale from 0-25 correlated with increased mortality risk in another study of 130 patients in the emergency department setting (HR 3.7, p=0.0348) (28). Higher semiquantitative total severity scores and multilobar involvement were associated with increased fatality risk in a study of 128 patients with COVID-19 hospitalized for observation; death was more common in patients with a CT severity score of 15 or greater (OR 53, p=0.003), and CT severity score was the only independent risk factor for mortality in a multivariate analysis that incorporated age and several inflammatory serum markers (30).

However, CT severity scores are just one of many clinical and laboratory parameters that correlate with patient prognosis (31). In addition, a significant percentage of patients with asymptomatic infection may have parenchymal involvement at CT that overlaps in severity with that of symptomatic patients (32), and CT severity scores of clinically severe cases of COVID-19 pneumonia may overlap with those of moderate clinical severity (30), underscoring limitations in drawing clinical conclusions from CT I n p r e s s severity alone. Although initial evidence is promising, clinical studies of the usefulness of CT severity scoring in management of patients with COVID-19 are still awaited.

What role does severity of disease on chest radiography play in disease evaluation?

CXR is commonly used as the initial diagnostic imaging test to evaluate patients with suspected or known COVID-19. Several studies investigating the relationship of severity of lung abnormalities on CXR with disease severity have shown scores reflecting increased extent and intensity of lung opacities to be associated with more severe clinical manifestations, higher rates of ICU admission, and death (10, 33, 34) . In one retrospective study of 338 young adults (median age 39 years), a CXR severity score ≥ 2 out of a maximum of 6 (OR 6.2) and obesity (OR 2.4) were independent predictors of hospital admission.

Of those admitted to hospital, a CXR severity score ≥ 3 was an independent predictor of endotracheal intubation (10) . Another study of 1157 subjects looked at a variety of clinical factors including radiologic assessment of lung edema (RALE) score found that with each unit increase of the RALE score, the hazard increased by 1.49 for ICU admission and 1.23 for death (33). A group from the Netherlands created a risk model from a retrospective study of 356 hospitalized patients with COVID-19 that included a 0-8 chest radiograph score based on 4 zones and severity of 0-2. Patients who required ICU admission or died had significantly higher radiograph scores (mean 4.4) than those who did not (mean 3.3), (p <0.01).

Furthermore, bilateral lung involvement at presentation was present in 86% of patients with critical illness as compared to 73% without (p=0.06) (34).

Patients with COVID-19 and normal or near normal CXRs typically have a benign clinical course.

In a retrospective study of 109 subjects with COVID-19 using a 72-point CXR severity score showed that a severity score < 5 between days 6-10 after onset of symptoms had a negative predictive value of 95.45 for supplemental oxygen requirement and 100.00 for ICU admission (35). A retrospective study of I n p r e s s normal results in 98%. Supplemental oxygen and inpatient treatment were only necessary for four (0.2%) (36) . These studies support recommendations that imaging is not usually indicated for patients with COVID-19 who have minimal or no symptoms. Routine radiographic monitoring of stable patients with COVID-19, including those requiring mechanical ventilation, is not recommended (6) .

Studies comparing the various scoring systems have not been performed, and each system has its own merits. In routine clinical practice, patients with COVID-19 and normal or minimal findings on CXR will likely have a benign clinical course whereas patients with more extensive lung opacities are much more likely to require supplemental oxygen, ICU admission, and mechanical ventilation.

Early during the COVID-19 pandemic when access to accurate RT-PCR testing was limited, several reporting systems were proposed for reporting CXR and CT scans of patients with suspected COVID-19 in a high disease prevalence setting. These systems provide standardized language and diagnostic categories aiming to convey the likelihood of lung abnormalities on CT representing COVID- 19 n p r e s s with a substantial agreement for CO-RADS 1 category (Fleiss' k = 0.61) and moderate agreement for CO-RADS 5 category (Fleiss' k = 0.60). CO-RADS score ≥ 4 was identified by ROC analysis as the optimal threshold with a cumulative area under the curve of 0.72, sensitivity 61%, and specificity 81% (42) .

Proposed CXR reporting language and categories include those of the British Society of Thoracic Imaging (43) and a multicenter US group (44) ( Table 2) . A retrospective study of the BSTI guidelines found substantial interobserver agreement (Fleiss' k = 0.61) for "classic" and "probable categories". Agreement was fair for the "Indeterminate for COVID-19" (k = 0.23), and "Non-COVID-19" (k = 0.37) categories (45) . The authors of this study suggest combining the latter two categories into a single "not classic of COVID-19" to improve interobserver agreement and to avoid labeling patients with COVID-19

as "Non-COVID-19".

Despite routine use of CT as a triage tool early in the outbreak in China, the role of CT has been more limited elsewhere in the world, where use has been mostly limited to specific indications such as pulmonary embolism (PE) (6) . With increased access to RT-PCR and faster results reporting, the prospective value of these classification systems in patients with COVID-19 is unclear. The positive predictive value of all these systems varies greatly with the prevalence of COVID-19 in the community.

However, further investigation is warranted to evaluate the diagnostic performance of CT for patients not suspected clinically of having COVID-19 but who have highly suggestive CT findings (6, 46) . While each classification system is likely to be helpful in suggesting the presence or absence of COVID-19 when typical findings are present or absent, respectively, the value of any one system will vary depending on disease prevalence and access to and rapidity of RT-PCR. Radiologist experience may also play a role.

The diagnosis of viral infection, including SARS-CoV-2, relies on RT-PCR test to identify genetic material in biological (47, 48) . However, the availability of RT-PCR was limited in the first half of 2020 and for this reason CT was used for early triage and management of COVID-19 pneumonia. Because of this, the specific definition of SARS-CoV-2 infection in the early literature is unclear, with the CT definition and RT-PCR definitions complementary. Although RT-PCR is considered the reference standard for diagnosis SARS-CoV-2 infection, it is far from perfect.

RT-PCR is subject to false-negative results because its diagnostic performance can be influenced by multiple factors such as inadequate sampling and improper extraction of nucleic acids from biological materials, variations in the accuracies of different tests, or low initial or late viral load (49, 50) . False negative CT scans have also been reported in 3%-56% of RT-PCR positive patients (14, 51, 52) . CT signs of pneumonia represent on potential manifestation of COVID-19 severity and tend to develop later in the disease course, typically 6-11 days after infection (53) .

Studies in symptomatic subjects reported higher sensitivity for CT as compared to RT-PCR (51, 53) . It has been advocated that such findings could be due to several factors, particularly the inclusion of only patients with moderate to severe symptoms (54).

The interpretation of the RT-PCR/CT mismatch is difficult and is confounded by a variety of factors (55) . In a large meta-analysis, the pooled sensitivity for CT was 94% (95% CI: 91%, 96%) and for RT-PCR 89% (95% CI: 81%, 94%) (56) . In this study, lower sensitivity of CT was reported as a function of symptoms and disease severity, whereas these factors did not influence RT-PCR performance, underscoring that imaging is not meant for screening asymptomatic patients (6, 36) .

The specificity of RT-PCR is optimal whereas the CT findings of COVID-19 pneumonia are far from pathognomonic. The systematic review from Cochrane Database reported substantial reduction of sensitivity and specificity in studies that included suspected cases. The chances of a positive CT result I n p r e s s were 86% in patients with a SARS-CoV-2 infection and 82% in patients without. Therefore, the specificity of CT is too weak to justify its use for the diagnosis of COVID-19 pneumonia (57) .

The accuracy of imaging tests in diagnosing COVID-19 and in any clinical setting is influenced by the prevalence of both COVID-19 and comparable viral pneumonias as well as other clinico-radiological mimickers. This issue was witnessed in regions with low rate of COVID-19 (<10%) where the positive predictive value (PPV) of chest CT was trivial (56) .

Artificial intelligence (AI) based on imaging has a potentially important role in the diagnosis, disease quantification, severity assessment, and prognosis of COVID-19 pneumonia. AI has been proposed as a tool to reduce radiologists' workload, streamline workflow, improve diagnostic accuracy, and facilitate resource allocation.

Published studies have focused on using CT and CXR to distinguish COVID-19 from other types of pneumonia and predict disease severity ( In summary, AI based on chest CT and CXR have demonstrated potential in both diagnosis and prognosis of COVID-19 pneumonia. However, to integrate these AI algorithms into routine clinical care in the fight against pandemic, we advocate the following: first, open-source datasets and code are strongly advocated for the broader community to train, test, and evaluate the performance of the machine learning classifiers. This is exemplified by the recent publication of the RSNA International COVID-19 Open Annotated Radiology Database (RICORD) (64); second, true generalizability will need to be assessed in real-time in a prospective study design. If one or a few of these models can be validated prospectively, they could inform treatment algorithms and guidelines customized for patients along the spectrum of COVID-19 ranging from mild symptoms to death and pave the way for a bigger role of AIbased imaging in COVID-19 resurgence and future pandemics.

Pulmonary macrovascular and microvascular manifestations of COVID-19, initially underrecognized, have received increasing attention in the radiology, clinical, and pathology literature. Our current understanding reflects intensive study predominantly focused on patients with severe disease and limited by marked paucity of data regarding asymptomatic patients and those with mild infection. The highest level of evidence, prospective randomized trials with outcomes data, is also presently lacking.

Diagnostic imaging pathways have evolved, often radically, during the COVID-19

pandemic. These changes vary widely in different countries, regions, and institutions reflecting differences in guidelines, availability of resources, prevalence of COVID-19, and institutional expertise.

Variability in data resulting from differences in diagnostic imaging use in different populations during I n p r e s s the pandemic and compared with pre-pandemic practice remains an analytic dilemma. This is particularly true for acute PE imaging, with performance characteristics that have been shown to be greatly impacted by disease prevalence (pretest probability) and diagnostic modality (65) . with COVID-19 who underwent CTPA and 62 matched patients from the pre-COVID-19 era from a single New York institution (69). They found a higher CTPA positivity rate for patients with COVID-19 as compared to the non-COVID-19 cohort (37.1% vs. 14.5%). One observational cohort study of 3334 subjects (60% men, median age 64 years) hospitalized COVID-19 patients from the same New York institution found a 16% rate of thrombotic complications, including acute PE in 3.2% (n=106) (70). Other series have described acute PE prevalence on CTPA in hospitalized COVID-19 patients ranging from I n p r e s s The dilated vessel sign was reported in the lungs early during the pandemic on unenhanced chest CT with some debate regarding its etiology. Possible explanations include small PEs, in situ pulmonary vascular thrombosis, and increased pulmonary blood flow (Fig. 5 ). There are imaging and pathological data to support both processes, and additional study will be useful for further elucidation (75, 76) . 

lung opacities in COVID-19 patients without PE (83) . They proposed that regional vasodilation rather than the normal vasoconstriction due to a dysfunctional diffuse inflammatory process. This paradoxical shunting to the hypoventilated regions may, in part, explain the poorly understood clinical phenomenon of the "happy hypoxic"-a well appearing profoundly hypoxic COVID-19 patient.

In severe COVID- 19 , it has been difficult to assess the contribution of PE to mortality in those In multisystem inflammatory system in children (MIS-C), coronary artery aneurysms have been described in close to 10% of hospitalized children (91) . CCTA has the capacity to exquisitely I n p r e s s demonstrate coronary artery aneurysms, particularly useful for children with limited echocardiographic windows.

While diagnostic imaging in COVID-19 has centered on the lungs and pulmonary manifestations of the disease, abdominal organs are also affected, most notably the liver, biliary tree, gastrointestinal tract, and abdominal vasculature (92). The affinity of SARS-CoV-2 for cells with surface expression of receptors for angiotensin-converting enzyme 2 (ACE2) forms a basis for preferential COVID-19 involvement of abdominal organs as well as the lungs (93) . Within the abdomen, ACE2 expression predominates in small bowl enterocytes, vascular endothelium, and biliary epithelium (92, 94).

Thromboembolic complications of COVID-19 have been reported across a variety of vascular beds and attributed to both microvascular inflammation and coagulopathy, leading to thrombosis and embolization (94) (95) (96) . Most abdominopelvic manifestations are associated with microvascular involvement and thus manifest as end-organ ischemia or infarction involving bowel, spleen, kidneys, and liver (92, 96) with rare reports of COVID-19 associated superior mesenteric artery or portal vein thrombosis (94, 97) .

Because the use of abdominal imaging in association with COVID-19 is substantially less frequent than thoracic manifestations, the evidence base is limited to small case series and case reports with limited data to support SARS-CoV-2 causation versus association.

Abdominal pain and sepsis are the most common indication for abdominal imaging in COVID-19.

Among patients hospitalized with COVID-19, cross-sectional abdominal imaging has been reported in 17% with equal use of ultrasonography (US) and CT and is significantly more likely among patients admitted to the ICU (92).

Common findings observed on abdominal CT scans include colorectal and small bowel wall thickening, fluid-filled colon, and infarction of the kidney, spleen, or liver. Pneumatosis and portal I n p r e s s venous gas without vascular involvement may be associated with ischemic enteritis and necrosis or pneumatosis cystoides intestinalis (98) . Direct viral infection, small vessel thrombosis, and nonocclusive mesenteric ischemia have been proposed as causes for the spectrum of bowel findings in COVID-19 (92).

Right upper abdominal US most commonly reveals signs of cholestasis, particularly gallbladder distension and sludge, and fatty liver (92). Acute cholecystitis, which can be acalculous, including a case of ischemic, gangrenous cholecystitis, has been reported in . In the latter instance, the gallbladder specimen lacked direct evidence of viral shedding, but medium vessel thrombosis, 

Approximately one half of patients that have undergone neuroimaging in the setting of COVID-19 have abnormal findings (106) . The primary patterns of abnormalities include ischemic and hemorrhagic stroke, leptomeningeal enhancement, encephalomyelitis, and widespread white matter hyperintensities, which may be associated with multiple microhemorrhages (107) (108) (109) (110) . Unfortunately, for many or most of these imaging findings, it remains unclear whether they are related to the underlying infection or to the multiple confounding comorbidities, including prolonged hypoxia and deranged coagulation parameters (111) . For example, the widespread microhemorrhages are striking but have been previously reported in seriously ill intensive care unit patients without COVID-19 (112) . A recent postmortem report noted that hemorrhagic transformation was rare with "pronounced neuroinflammatory changes" representing the dominant pathologic findings (113) . Rarity of hemorrhagic findings was also observed in a recent systematic review (114) . Large vessel occlusions have been seen in patients with COVID-19, which is unsurprising given the prothrombotic effects of the underlying infection. Anosmia, a common presenting symptom, was shown to be associated with T2 hyperintensity and volume loss in the olfactory bulb on MRI, but it remains unclear whether these imaging findings relate to direct viral involvement or post-inflammatory changes (115, 116) . A recent report of olfactory epithelial biopsy highlighted its disruption, suggesting that direct neural invasion was not the primary culprit in anosmia (117) .

The virus itself has been isolated from the CSF in very few reports (118) 

Although our knowledge of COVID-19 and the role imaging plays in diagnosis and management has greatly increased since the infection became a global pandemic, many questions remain unanswered:

• What are the long-term sequelae of COVID-19 in the lungs and cardiovascular system?

• Are there long-term sequelae of COVID-19 outside of the cardiovascular system that have yet to be determined?

• Do classification systems of CT and CXR findings sufficiently predict prognosis and is extent of disease enough?

• Can imaging be used to reduce hospital admissions?

• How will the role of imaging change with the winter season of respiratory illnesses in the Ground-glass opacity (GGO) I n p r e s s I n p r e s s 

Chest CT Findings in 2019 Novel Coronavirus (2019-nCoV) Infections from Wuhan, China: Key Points for the Radiologist

Essentials for Radiologists on COVID-19: An Update

Case 25-2020: A 47-Year-Old Woman with a Lung Mass

ACR recommendations for the use of chest radiography and computed tomography (CT) for suspected COVID-19 infection Reston, VA2020

Society of Thoracic Radiology. STR/ASER COVID-19 position statement. Society of Thoracic Radiology and

The Role of Chest Imaging in Patient Management during the COVID-19 Pandemic: A Multinational Consensus Statement from the Fleischner Society

Computed Tomography Features of Coronavirus Disease 2019 (COVID-19): A Review for Radiologists

Frequency and Distribution of

Chest Radiographic Findings in Patients Positive for COVID-19

Portable chest X-ray in coronavirus disease-19 (COVID-19): A pictorial review

Clinical and Chest Radiography Features Determine Patient Outcomes In Young and Middle Age Adults with COVID-19

Novel Coronavirus (2019-nCoV)

America Expert Consensus Statement on Reporting Chest CT Findings Related to COVID-19. Endorsed by the Society of Thoracic Radiology, the American College of Radiology, and RSNA -Secondary Publication

Increased Incidence of Barotrauma in Patients with COVID-19 on Invasive Mechanical Ventilation

Chest CT Findings in Coronavirus Disease-19 (COVID-19): Relationship to Duration of Infection

Time Course of Lung Changes On Chest CT During Recovery From 2019 Novel Coronavirus (COVID-19) Pneumonia. Radiology

Initial CT findings and temporal changes in patients with the novel coronavirus pneumonia (2019-nCoV): a study of 63 patients in Wuhan

CT image visual quantitative evaluation and clinical classification of coronavirus disease (COVID-19)

CT lung lesions as predictors of early death or ICU admission in COVID-19 patients

Assessment of the Severity of Coronavirus Disease: Quantitative Computed Tomography Parameters versus Semiquantitative Visual Score

severity and progression using chest CT images

Dynamic changes of Chest CT follow

COVID-19) pneumonia: relationship to clinical typing

Temporal radiographic changes in COVID-19 patients: relationship to disease severity and viral clearance

Relation Between Chest CT Findings and Clinical Conditions of Coronavirus Disease (COVID-19) Pneumonia: A Multicenter Study

CT Quantitative Analysis and Its Relationship with

Clinical Features for Assessing the Severity of Patients with COVID-19

Association between cytokine profiles and lung injury in COVID-19 pneumonia

Utility of Screening Chest Radiographs in Patients with

Asymptomatic or Minimally Symptomatic COVID-19 in Singapore. Radiology. 2020:203496

CO-RADS: A Categorical CT Assessment Scheme for Patients Suspected of Having COVID-19-Definition and Evaluation

Guidance for the Reporting Radiologist version 2.0 London: BSTI; 2020

imaging reporting and data system (COVID-RADS) and common lexicon: a proposal based on the imaging data of 37 studies

COVID-19 S: A new proposal for diagnosis and structured reporting of COVID-19 on computed tomography imaging

RSNA Expert Consensus Statement on Reporting Chest CT Findings Related to COVID-19: Interobserver Agreement Between Chest Radiologists

Diagnostic accuracy and interobserver variability of CO-RADS in patients with suspected coronavirus disease-2019: a multireader validation study

The continuing evolution of COVID-19 imaging pathways in the UK: a British Society of Thoracic Imaging expert reference group update

Review of Chest Radiograph Findings of COVID-19 Pneumonia and Suggested Reporting Language

Validation of the British Society of Thoracic Imaging guidelines for COVID-19 chest radiograph reporting

Detection of Unsuspected Coronavirus Disease 2019 Cases by Computed Tomography and Retrospective Implementation of the

Consensus Guidelines

Sensitivity of Chest CT for COVID-19: Comparison to RT-PCR

Diagnostic Testing for the Novel Coronavirus

Diagnosis of SARS-CoV-2 infection based on CT scan vs RT-PCR: reflecting on experience from MERS-CoV

Can Chest CT Features Distinguish Patients With Negative From Those With Positive Initial RT-PCR Results for Coronavirus Disease (COVID-19)?

Correlation of Chest CT and RT-PCR Testing for

COVID-19) in China: A Report of 1014 Cases

Patients with RT-PCR-confirmed COVID-19 and Normal Chest CT

Temporal Changes of CT Findings in 90 Patients with COVID-19 Pneumonia: A Longitudinal Study

Diagnostic Tools for Coronavirus Disease (COVID-19): Comparing CT and RT-PCR Viral Nucleic Acid Testing

Epub 2020/05/15

Imaging Publications in the COVID-19 Pandemic: Applying New Research Results to Clinical Practice

Diagnostic Performance of CT and Reverse Transcriptase Polymerase Chain Reaction for Coronavirus Disease 2019: A Meta-Analysis

Thoracic imaging tests for the diagnosis of COVID-19

Using Artificial Intelligence to Detect COVID-19 and Community-acquired Pneumonia Based on Pulmonary CT: Evaluation of the Diagnostic Accuracy

Radiologist Performance in Distinguishing COVID-19 from Pneumonia of Other Origin at Chest CT

Artificial intelligence-enabled rapid diagnosis of COVID-19 patients

Radiographs: A Multireader Evaluation of an Artificial Intelligence System

Automated assessment of COVID

The RSNA International COVID-19 Open Annotated Radiology Database (RICORD)

American Society of Hematology 2018 guidelines for management of venous thromboembolism: diagnosis of venous thromboembolism

Point-of-care lung ultrasound in patients with COVID-19 -a narrative review

Diagnostic Evaluation of Pulmonary

Embolism During the COVID-19 Pandemic

High risk of thrombosis in patients with severe SARS-CoV-2 infection: a multicenter prospective cohort study

Pulmonary Embolism on CTPA in COVID-19 Patients

Thrombosis in Hospitalized Patients With COVID-19 in a New York City Health System

Incidence of pulmonary embolism in non-critically ill COVID-19 patients. Predicting factors for a challenging diagnosis

Acute pulmonary embolism in non-hospitalized COVID-19 patients referred to CTPA by emergency department

Acute Pulmonary Embolism Associated with COVID-19 Pneumonia Detected with Pulmonary CT Angiography

Pulmonary Embolism in Patients with COVID-19 at CT Angiography and Relationship to d-Dimer Levels

Is COVID Evolution Due to Occurrence of Pulmonary Vascular Thrombosis?

Epub 2020/04/28

Clinical and computed tomography characteristics of COVID-19 associated acute pulmonary embolism: A different phenotype of thrombotic disease?

Pulmonary Arterial Thrombosis in COVID-19 With Fatal Outcome : Results From a Prospective

Pulmonary and cardiac pathology in African American patients with COVID-19: an autopsy series from New Orleans

Splitting the matrix": intussusceptive angiogenesis meets MT1-MMP

Hypoxaemia related to COVID-19: vascular and perfusion abnormalities on dual-energy CT

Diagnosis, Prevention, and Treatment of Thromboembolic Complications in COVID-19: Report of the National Institute for Public Health of the Netherlands

The association between treatment with heparin and survival in patients with Covid-19

Outcomes of

Cardiovascular Magnetic Resonance Imaging in Patients Recently Recovered From Coronavirus Disease

JAMA Cardiol

Cardiac Involvement in Patients Recovered From COVID-2019 Identified Using Magnetic Resonance Imaging

Epub 2020/05/12

Cardiovascular Magnetic Resonance Findings in Competitive Athletes Recovering From COVID-19 Infection

Epub 2020/09/11

Pandemic on ST-Segment-Elevation Myocardial Infarction Presentations and In-Hospital Outcomes

ST

Segment Elevation Cardiac Catheterization Laboratory Activations in the United States During

Statement from the North American Society for Cardiovascular Imaging on imaging strategies to reduce the scarcity of healthcare resources during the COVID-19 outbreak

Abdominal Imaging Findings in COVID-19: Preliminary Observations

Evidence for Gastrointestinal Infection of SARS-CoV

Hypercoagulability in COVID-19: Identification of Arterial and Venous Thromboembolism in the Abdomen, Pelvis, and Lower Extremities

Multisystem assessment of the imaging manifestations of coagulopathy in hospitalized patients with COVID-19

Update: Venous Thrombosis and Hypercoagulability in the Abdomen and Pelvis-Findings in COVID-19

Superior Mesenteric Artery Thrombosis and Acute Intestinal Ischemia as a Consequence of COVID-19 Infection

Pneumatosis Intestinalis in COVID-19: Case Series

COVID-19 with acute cholecystitis: a case report

Acalcolous Hemorrhagic Cholecystitis and SARS-CoV-2 Infection

Histopathological findings in a COVID-19 patient affected by ischemic gangrenous cholecystitis

Pancreatic Injury Patterns in Patients With Coronavirus Disease 19 Pneumonia

COVID-19 Induced Acute Pancreatitis: A Case Report and Literature Review

COVID-19-associated acute pancreatitis: a rare cause of acute abdomen

Brain MRI Findings in Patients in the Intensive Care Unit with COVID-19 Infection

Neurologic and neuroimaging findings in COVID-19 patients: A retrospective multicenter study

Nervous System Involvement in COVID-19: Results from a Retrospective Consecutive Neuroimaging Cohort

MRI Brain Findings in 126 Patients with COVID-19: Initial Observations from a Descriptive Literature Review

Neuroimaging findings of brain MRI and CT in patients with COVID-19: A systematic review and meta-analysis

Insights Into Neuroimaging Findings of Patients With

Coronavirus Disease

Neuroimaging Findings in Patients with COVID-19

Neuropathology of patients with COVID-19 in Germany: a post-mortem case series

Epub 2020/10/05

Brain abnormalities in COVID-19 acute/subacute phase: A rapid systematic review

Olfactory Bulb Signal Abnormality in Patients with COVID-19 Who Present with Neurologic Symptoms

Olfactory Bulb MRI and Paranasal Sinus CT Findings in Persistent COVID-19 Anosmia

Olfactory epithelium histopathological findings in long-term coronavirus disease 2019 related anosmia

Neurologic Involvement in COVID-19: Cause or Coincidence? A Neuroimaging Perspective

Automated Assessment of CO-RADS and Chest CT Severity Scores in Patients with

CT quantification of pneumonia lesions in early days predicts progression to severe illness in a cohort of COVID-19 patients

A fully automatic deep learning system for COVID-19 diagnostic and prognostic analysis

Artificial intelligence for the detection of COVID-19 pneumonia on chest CT using multinational datasets

Epub 2020/08/14

Open resource of clinical data from patients with pneumonia for the prediction of COVID-19 outcomes via deep learning

Development and evaluation of an artificial intelligence system for COVID-19 diagnosis

Diagnosis of COVID-19 Pneumonia Using Chest Radiography: Value of Artificial Intelligence

The authors would like thank Adina Haramati, M.D., for contributing to the pulmonary vascular and cardiovascular content. I n p r e s s