Deciphering Unexpected Vascular Locations of Scedosporium spp. and Lomentospora prolificans Fungal Infections, France.

Scedosporium spp. and Lomentospora prolificans are emerging non-Aspergillus filamentous fungi. The Scedosporiosis/lomentosporiosis Observational Study we previously conducted reported frequent fungal vascular involvement, including aortitis and peripheral arteritis. For this article, we reviewed 7 cases of Scedosporium spp. and L. prolificans arteritis from the Scedosporiosis/lomentosporiosis Observational Study and 13 cases from published literature. Underlying immunosuppression was reported in 70% (14/20) of case-patients, mainly those who had solid organ transplants (10/14). Osteoarticular localization of infection was observed in 50% (10/20) of cases; infections were frequently (7/10) contiguous with vascular infection sites. Scedosporium spp./Lomentospora prolificans infections were diagnosed in 9 of 20 patients ≈3 months after completing treatment for nonvascular scedosporiosis/lomentosporiosis. Aneurysms were found in 8/11 aortitis and 6/10 peripheral arteritis cases. Invasive fungal disease--related deaths were high (12/18 [67%]). The vascular tropism of Scedosporium spp. and L. prolificans indicates vascular imaging, such as computed tomography angiography, is needed to manage infections, especially for osteoarticular locations.

Primary immunodeficiency diseases of adults: a review of pulmonary complication imaging findings.

Our objective in this review is to familiarize radiologists with the spectrum of initial and progressive CT manifestations of pulmonary complications observed in adult patients with primary immunodeficiency diseases, including primary antibody deficiency (PAD), hyper-IgE syndrome (HIES), and chronic granulomatous disease (CGD). In patients with PAD, recurrent pulmonary infections may lead to airway remodeling with bronchial wall-thickening, bronchiectasis, mucus-plugging, mosaic perfusion, and expiratory air-trapping. Interstitial lung disease associates pulmonary lymphoid hyperplasia, granulomatous inflammation, and organizing pneumonia and is called granulomatous-lymphocytic interstitial lung disease (GLILD). The CT features of GLILD are solid and semi-solid pulmonary nodules and areas of air space consolidation, reticular opacities, and lymphadenopathy. These features may overlap those of mucosa-associated lymphoid tissue (MALT) lymphoma, justifying biopsies. In patients with HIES, particularly the autosomal dominant type (Job syndrome), recurrent pyogenic infections lead to permanent lung damage. Secondary infections with aspergillus species develop in pre-existing pneumatocele and bronchiectasis areas, leading to chronic airway infection. The complete spectrum of CT pulmonary aspergillosis may be seen including aspergillomas, chronic cavitary pulmonary aspergillosis, allergic bronchopulmonary aspergillosis (ABPA)-like pattern, mixed pattern, and invasive. Patients with CGD present with recurrent bacterial and fungal infections leading to parenchymal scarring, traction bronchiectasis, cicatricial emphysema, airway remodeling, and mosaicism. Invasive aspergillosis, the major cause of mortality, manifests as single or multiple nodules, areas of airspace consolidation that may be complicated by abscess, empyema, or contiguous extension to the pleura or chest wall. CLINICAL RELEVANCE STATEMENT: Awareness of the imaging findings spectrum of pulmonary complications that can occur in adult patients with primary immunodeficiency diseases is important to minimize diagnostic delay and improve patient outcomes. KEY POINTS: • Unexplained bronchiectasis, associated or not with CT findings of obliterative bronchiolitis, should evoke a potential diagnosis of primary autoantibody deficiency. • The CT evidence of various patterns of aspergillosis developed in severe bronchiectasis or pneumatocele in a young adult characterizes the pulmonary complications of hyper-IgE syndrome. • In patients with chronic granulomatous disease, invasive aspergillosis is relatively frequent, often asymptomatic, and sometimes mimicking or associated with non-infectious inflammatory pulmonary lesions.

Cement leakage during vertebroplasty: comparison between intact and wall disruption in spinal metastases.

BACKGROUND: Percutaneous vertebroplasty (PVP) is an effective measure for painful metastases or impending pathological fractures of the spine with cement leakages being the most frequent complication. Posterior extrusion of cement into the spinal canal may result in neurological symptoms and deficits. PURPOSE: To compare the occurrence of intraspinal canal cement leakage between vertebrae with posterior wall disruption and vertebrae without posterior wall disruption. MATERIAL AND METHODS: A single-center retrospective study was conducted of all PVP for spine metastases between June 2020 and November 2021. All leaks were analyzed by a postprocedural computed tomography scan or cone-beam computed tomography. RESULTS: A total of 77 patients with 143 vertebrae treated by PVP were included. Posterior wall disruption was observed in 64 (44.8%) vertebrae while 79 (55.2%) had a complete posterior wall. Spinal canal cement leakage occurred in 36 (25.2%) vertebrae and was comparable in both groups, occurring in 16 (25.0%) vertebrae with posterior wall disruption and 20 (25.3%) vertebrae without posterior wall disruption (P = 1). No risk factors for spinal canal leakage were found in the univariate and multivariate analyses. One spinal leak was symptomatic with intercostal neuralgia. CONCLUSION: Our results suggest that an incomplete vertebral posterior wall does not increase the rate of spinal canal cement leakage during PVP.

Bronchial artery embolization for hemoptysis in adult patients with cystic fibrosis: a single-center retrospective study.

BACKGROUND: Hemoptysis is a severe complication of cystic fibrosis (CF) for which bronchial artery embolization (BAE) is an efficient primary therapeutic option. However, recurrence is more frequent than for other etiologies of hemoptysis. PURPOSE: To assess the safety and efficacy of BAE in patients with CF and hemoptysis and predictive factors for recurrent hemoptysis. MATERIAL AND METHODS: This retrospective study reviewed all adult patients with CF treated by BAE for hemoptysis in our center from 2004 to 2021. The primary endpoint was the recurrence of hemoptysis after bronchial artery embolization. Secondary endpoints were overall survival and complications. We introduced the vascular burden (VB) defined as the sum of all bronchial artery diameters measured on pre-procedural enhanced computed tomography (CT) scans. RESULTS: A total of 48 BAE were performed in 31 patients. A total of 19 recurrences occurred with a median recurrence-free survival of 3.9 years. In univariate analyzes, percentage of unembolized VB (%UVB) (hazard ratio [HR] = 1.034, 95% confidence interval [CI=1.016-1.052; P < 0.001) and %UVB vascularizing the suspected bleeding lung (%UVB-lat) (HR = 1.024, 95% CI=1.012-1.037; P < 0.001) were associated with recurrence. In multivariate analyzes, only %UVB-lat remained significantly associated with recurrence (HR = 1.020, 95% CI=1.002-1.038; P = 0.030). One patient died during follow-up. No complication of grade 3 or higher was reported according to the CIRSE classification system for complications. CONCLUSION: When possible, unilateral BAE seems sufficient in patients with CF with hemoptysis even in such a diffuse disease involving both lungs. The efficiency of BAE could be improved by thoroughly targeting all arteries vascularizing the bleeding lung.

COVID-19 pneumonia: high diagnostic accuracy of chest CT in patients with intermediate clinical probability.

OBJECTIVES: To assess inter-reader agreements and diagnostic accuracy of chest CT to identify COVID-19 pneumonia in patients with intermediate clinical probability during an acute disease outbreak. METHODS: From March 20 to April 8, 319 patients (mean age 62.3 years old) consecutive patients with an intermediate clinical probability of COVID-19 pneumonia underwent a chest CT scan. Two independent chest radiologists blinded to clinical information and RT-PCR results retrospectively reviewed and classified images on a 1-5 confidence level scale for COVID-19 pneumonia. Agreements between radiologists were assessed with kappa statistics. Diagnostic accuracy of chest CT compared with RT-PCR assay and patient outcomes was measured using receiver operating characteristics (ROC). Positive predictive value (PPV) and negative predictive value (NPV) for COVID-19 pneumonia were calculated. RESULTS: Inter-observer agreement for highly probable (kappa: 0.83 [p < .001]) and highly probable or probable (kappa: 0.82 [p < .001]) diagnosis of COVID-19 pneumonia was very good. RT-PCR tests performed in 307 patients were positive in 174 and negative in 133. The areas under the curve (AUC) were 0.94 and 0.92 respectively. With a disease prevalence of 61.2%, PPV were 95.9% and 94.3%, and NPV 84.4% and 77.1%. CONCLUSION: During acute COVID-19 outbreak, chest CT scan may be used for triage of patients with intermediate clinical probability with very good inter-observer agreements and diagnostic accuracy. KEY POINTS: • Concordances between two chest radiologists to diagnose or exclude a COVID-19 pneumonia in 319 consecutive patients with intermediate clinical probability were very good (kappa: 0.82; p < .001). • When compared with RT-PCR results and patient outcomes, the diagnostic accuracy of CT to identify COVID-19 pneumonia was high for both radiologists (AUC: 0.94 and 0.92). • With a disease prevalence of 61.2% in the studied population, the positive predictive values of CT for diagnosing COVID-19 pneumonia were 95.9% and 94.3% with negative predictive values of 84.4% and 77.1%.

Machine learning automatically detects COVID-19 using chest CTs in a large multicenter cohort.

OBJECTIVES: To investigate machine learning classifiers and interpretable models using chest CT for detection of COVID-19 and differentiation from other pneumonias, interstitial lung disease (ILD) and normal CTs. METHODS: Our retrospective multi-institutional study obtained 2446 chest CTs from 16 institutions (including 1161 COVID-19 patients). Training/validation/testing cohorts included 1011/50/100 COVID-19, 388/16/33 ILD, 189/16/33 other pneumonias, and 559/17/34 normal (no pathologies) CTs. A metric-based approach for the classification of COVID-19 used interpretable features, relying on logistic regression and random forests. A deep learning-based classifier differentiated COVID-19 via 3D features extracted directly from CT attenuation and probability distribution of airspace opacities. RESULTS: Most discriminative features of COVID-19 are the percentage of airspace opacity and peripheral and basal predominant opacities, concordant with the typical characterization of COVID-19 in the literature. Unsupervised hierarchical clustering compares feature distribution across COVID-19 and control cohorts. The metrics-based classifier achieved AUC = 0.83, sensitivity = 0.74, and specificity = 0.79 versus respectively 0.93, 0.90, and 0.83 for the DL-based classifier. Most of ambiguity comes from non-COVID-19 pneumonia with manifestations that overlap with COVID-19, as well as mild COVID-19 cases. Non-COVID-19 classification performance is 91% for ILD, 64% for other pneumonias, and 94% for no pathologies, which demonstrates the robustness of our method against different compositions of control groups. CONCLUSIONS: Our new method accurately discriminates COVID-19 from other types of pneumonia, ILD, and CTs with no pathologies, using quantitative imaging features derived from chest CT, while balancing interpretability of results and classification performance and, therefore, may be useful to facilitate diagnosis of COVID-19. KEY POINTS: • Unsupervised clustering reveals the key tomographic features including percent airspace opacity and peripheral and basal opacities most typical of COVID-19 relative to control groups. • COVID-19-positive CTs were compared with COVID-19-negative chest CTs (including a balanced distribution of non-COVID-19 pneumonia, ILD, and no pathologies). Classification accuracies for COVID-19, pneumonia, ILD, and CT scans with no pathologies are respectively 90%, 64%, 91%, and 94%. • Our deep learning (DL)-based classification method demonstrates an AUC of 0.93 (sensitivity 90%, specificity 83%). Machine learning methods applied to quantitative chest CT metrics can therefore improve diagnostic accuracy in suspected COVID-19, particularly in resource-constrained environments.

Diagnostic Yield of Venous Thrombosis and Pulmonary Embolism by Combined CT Venography and Pulmonary Angiography in Patients with Cryptogenic Stroke and Patent Foramen Ovale.

BACKGROUND: Paradoxical embolism via a patent foramen ovale (PFO) has been suggested as a potential stroke mechanism. Combined CT venography and pulmonary angiography (CVPA) is a simple, validated and accurate technique to diagnose deep venous thrombosis (DVT) or pulmonary embolism (PE). We sought to assess the prevalence of DVT or PE among patients with PFO and cryptogenic stroke (CS) by CVPA. METHODS: Patients were identified retrospectively from a clinical registry of consecutive patients with stroke admitted to our Stroke Unit. The following criteria were required for inclusion in this study: CS, PFO identified by transthoracic echography using contrast medium and CVPA performed during the hospitalization following stroke. RESULTS: A total of 114 patients with PFO underwent a CVPA within 7 days (interquartile range 4-9) from stroke symptom onset. On cerebral imaging, 11% had multiple infarcts. CVPA documented deep vein thrombosis (DVT) in 10 patients (8.8%) and PE in 5 patients (4.4%), that is, a total of 12 patients with prevalence of 10.5% (95% CI 5.5-17.7). Patients with PE-DVT had higher D-dimers and C reactive protein level than patients without PE-DVT (p < 0.05). CONCLUSION: CVPA may be used by the stroke team in the work-up of suspected paradoxical embolism among cryptogenic ischemic stroke patients with PFO.

A deep learning-based algorithm improves radiology residents' diagnoses of acute pulmonary embolism on CT pulmonary angiograms.

PURPOSE: To compare radiology residents' diagnostic performances to detect pulmonary emboli (PEs) on CT pulmonary angiographies (CTPAs) with deep-learning (DL)-based algorithm support and without. METHODS: Fully anonymized CTPAs (n = 207) of patients suspected of having acute PE served as input for PE detection using a previously trained and validated DL-based algorithm. Three residents in their first three years of training, blinded to the index report and clinical history, read the CTPAs first without, and 2 months later with the help of artificial intelligence (AI) output, to diagnose PE as present, absent or indeterminate. We evaluated concordances and discordances with the consensus-reading results of two experts in chest imaging. RESULTS: Because the AI algorithm failed to analyze 11 CTPAs, 196 CTPAs were analyzed; 31 (15.8 %) were PE-positive. Good-classification performance was higher for residents with AI-algorithm support than without (AUROCs: 0.958 [95 % CI: 0.921-0.979] vs. 0.894 [95 % CI: 0.850-0.931], p < 0.001, respectively). The main finding was the increased sensitivity of residents' diagnoses using the AI algorithm (92.5 % vs. 81.7 %, respectively). Concordance between residents (kappa: 0.77 [95 % CI: 0.76-0.78]; p < 0.001) improved with AI-algorithm use (kappa: 0.88 [95 % CI: 0.87-0.89]; p < 0.001). CONCLUSION: The AI algorithm we used improved between-resident agreements to interpret CTPAs for suspected PE and, hence, their diagnostic performances.

Prognostic value of cellular population data in patients with COVID-19.

Background and aims: Beckman Coulter hematology analysers identify leukocytes by their volume (V), conductivity (C) and scatter (S) of a laser beam at different angles. Each leukocyte sub-population [neutrophils (NE), lymphocytes (LY), monocytes (MO)] is characterized by the mean (MN) and the standard deviation (SD) of 7 measurements called "cellular population data" (@CPD), corresponding to morphological analysis of the leukocytes. As severe forms of infections to SARS-CoV-2 are characterized by a functional activation of mononuclear cells, leading to a cytokine storm, we evaluated whether CPD variations are correlated to the inflammation state, oxygen requirement and lung damage and whether CPD analysis could be useful for a triage of patients with COVID-19 in the Emergency Department (ED) and could help to identify patients with a high risk of worsening. Materials and method: The CPD of 825 consecutive patients with proven COVID-19 presenting to the ED were recorded and compared to classical biochemical parameters, the need for hospitalization in the ward or ICU, the need for oxygen, or lung injury on CT-scan. Results: 40 of the 42 CPD were significantly modified in COVID-19 patients in comparison to 245 controls. @MN-V-MO and @SD-V-MO were highly correlated with C-reactive protein, procalcitonin, ferritin and D-dimers. SD-UMALS-LY > 21.45 and > 23.92 identified, respectively, patients with critical lung injuries (>75%) and requiring tracheal intubation. @SD-V-MO > 25.03 and @SD-V-NE > 19.4 identified patients required immediate ICU admission, whereas a @MN-V-MO < 183 suggested that the patient could be immediately discharged. Using logistic regression, the combination of 8 CPD with platelet and basophil counts and the existence of diabetes or obesity could identify patients requiring ICU after a first stay in conventional wards (area under the curve = 0.843). Conclusion: CPD analysis constitutes an easy and inexpensive tool for triage and prognosis of COVID-19 patients in the ED.

Stratification of COVID-19 Severity Using SeptiCyte RAPID, a Novel Host Immune Response Test.

SeptiCyte® RAPID is a gene expression assay measuring the relative expression levels of host response genes PLA2G7 and PLAC8, indicative of a dysregulated immune response during sepsis. As severe forms of COVID-19 may be considered viral sepsis, we evaluated SeptiCyte RAPID in a series of 94 patients admitted to Foch Hospital (Suresnes, France) with proven SARS-CoV-2 infection. EDTA blood was collected in the emergency department (ED) in 67 cases, in the intensive care unit (ICU) in 23 cases and in conventional units in 4 cases. SeptiScore (0-15 scale) increased with COVID-19 severity. Patients in ICU had the highest SeptiScores, producing values comparable to 8 patients with culture-confirmed bacterial sepsis. Receiver operating characteristic (ROC) curve analysis had an area under the curve (AUC) of 0.81 for discriminating patients requiring ICU admission from patients who were immediately discharged or from patients requiring hospitalization in conventional units. SeptiScores increased with the extent of the lung injury. For 68 patients, a chest computed tomography (CT) scan was performed within 24 h of COVID-19 diagnosis. SeptiScore >7 suggested lung injury ≥50% (AUC = 0.86). SeptiCyte RAPID was compared to other biomarkers for discriminating Critical + Severe COVID-19 in ICU, versus Moderate + Mild COVID-19 not in ICU. The mean AUC for SeptiCyte RAPID was superior to that of any individual biomarker or combination thereof. In contrast to C-reactive protein (CRP), correlation of SeptiScore with lung injury was not impacted by treatment with anti-inflammatory agents. SeptiCyte RAPID can be a useful tool to identify patients with severe forms of COVID-19 in ED, as well as during follow-up.

Development of a Decision Support Tool for Anticoagulation in Critically Ill Patients Admitted for SARS-CoV-2 Infection: The CALT Protocol.

Severe COVID-19 infections are at high risk of causing thromboembolic events (TEEs). However, the usual exams may be unavailable or unreliable in predicting the risk of TEEs at admission or during hospitalization. We performed a retrospective analysis of two centers (n = 124 patients) including severe COVID-19 patients to determine the specific risk factors of TEEs in SARS-CoV-2 infection at admission and during stays at the intensive care unit (ICU). We used stepwise regression to create two composite scores in order to predict TEEs in the first 48 h (H0-H48) and during the first 15 days (D1-D15) in ICU. We then evaluated the performance of our scores in our cohort. During the period H0-H48, patients with a TEE diagnosis had higher D-Dimers and ferritin values at day 1 (D1) and day 3 (D3) and a greater drop in fibrinogen between D1 and D3 compared with patients without TEEs. Over the period D1-D15, patients with a diagnosis of a TEE showed a more marked drop in fibrinogen and had higher D-Dimers and lactate dehydrogenase (LDH) values at D1 and D3. Based on ROC analysis, the COVID-related acute lung and deep vein thrombosis (CALT) 1 score, calculated at D1, had a diagnostic performance for TEEs at H0-H48, estimated using an area under the curve (AUC) of 0.85 (CI95%: 0.76-0.93, p < 10-3). The CALT 2 score, calculated at D3, predicted the occurrence of TEEs over the period D1-D15 with an estimated AUC of 0.85 (CI95%: 0.77-0.93, p < 10-3). These two scores were used as the basis for the development of the CALT protocol, a tool to assist in the decision to use anticoagulation during severe SARS-CoV-2 infections. The CALT scores showed good performances in predicting the risk of TEEs in severe COVID-19 patients at admission and during ICU stays. They could, therefore, be used as a decision support protocol on whether or not to initiate therapeutic anticoagulation.

The Potential Role of Artificial Intelligence in Lung Cancer Screening Using Low-Dose Computed Tomography.

Two large randomized controlled trials of low-dose CT (LDCT)-based lung cancer screening (LCS) in high-risk smoker populations have shown a reduction in the number of lung cancer deaths in the screening group compared to a control group. Even if various countries are currently considering the implementation of LCS programs, recurring doubts and fears persist about the potentially high false positive rates, cost-effectiveness, and the availability of radiologists for scan interpretation. Artificial intelligence (AI) can potentially increase the efficiency of LCS. The objective of this article is to review the performances of AI algorithms developed for different tasks that make up the interpretation of LCS CT scans, and to estimate how these AI algorithms may be used as a second reader. Despite the reduction in lung cancer mortality due to LCS with LDCT, many smokers die of comorbid smoking-related diseases. The identification of CT features associated with these comorbidities could increase the value of screening with minimal impact on LCS programs. Because these smoking-related conditions are not systematically assessed in current LCS programs, AI can identify individuals with evidence of previously undiagnosed cardiovascular disease, emphysema or osteoporosis and offer an opportunity for treatment and prevention.

Anterior epidural carbon dioxide dissection during spine cryoablation.

PURPOSE: Spine cryoablation (SC) of posterior vertebral lesions exposes to neuronal damages and incomplete treatment due to the proximity of the spinal canal. Carbon dioxide (CO2) dissection is a nerve protective method that can be used during spine cryoablation that tends to distribute in non-dependent areas. The purpose of this technical note was to expose the feasibility of anterior epidural CO2 dissection during SC in prone decubitus. MATERIALS AND METHODS: Three consecutives patients underwent SC of metastases abutting the posterior wall of the vertebra with anterior epidural CO2 dissection. A post-ablation MRI was performed after each cryoablation to state if the treatment was complete or incomplete. Complications were reported using the Common Terminology Criteria for Adverse Events v5.0 (CTCAE). RESULTS: Peri-procedural anterior epidural injection of CO2 was successful in all 3 procedures. Treatment was considered complete on all post-ablation MRI with ablation margins encompassing the targeted metastasis. No complication according to the CTCAE was reported. CONCLUSION: CO2 dissection of the anterior epidural space was successful in all 3 procedures allowing complete treatment on all post-ablation MRI.

Trans-arterial embolization for hemoptysis in lung transplant recipients.

INTRODUCTION: Hemoptysis isn't rare in lung transplant recipients (LTR). Yet, trans-arterial embolization (TAE) in LTR has been rarely reported in the literature. The aim of the study was to present the feasibility and outcomes of TAE for hemoptysis in LTR. MATERIALS AND METHODS: Retrospective study of all LTR who underwent TAE for hemoptysis in our single institution between 2005 and 2020. RESULTS: A total of 787 patients underwent lung transplantation between 2005 and 2020. Fifteen LTR underwent 21 TAE for hemoptysis in a median delay of 42 days after LT. TAE was performed within a year after LT in 13 patients (86.7%) with 12 of those patients having concomitant severe ischemic airway injury with necrosis and anastomotic dehiscence. Bronchoscopy confirmed bronchial anastomoses has being the source of the bleeding in 11 LTR (84.6%). Restoration of bronchial vascularization was highlighted in 13 patients (87%). Despite TAE, bronchial anastomosis healing was observed in all surviving patients with anastomotic dehiscence in a median delay of 43 days. CONCLUSION: In our experience, hemoptysis requiring TAE in LTR was rare, frequently occurring in the first weeks after LT, and seemed associated with anastomotic ischemia and dehiscence. Bleeding mainly originated from ischemic bronchial anastomosis through the restoration of the bronchial artery circulation. Our results suggest that bronchial arteriography should be routinely proposed in such patients in the event of hemoptysis.

Lung Transplantation: CT Assessment of Chronic Lung Allograft Dysfunction (CLAD).

Chronic lung allograft rejection remains one of the major causes of morbi-mortality after lung transplantation. The term Chronic Lung Allograft Dysfunction (CLAD) has been proposed to describe the different processes that lead to a significant and persistent deterioration in lung function without identifiable causes. The two main phenotypes of CLAD are Bronchiolitis Obliterans Syndrome (BOS) and Restrictive Allograft Syndrome (RAS), each of them characterized by particular functional and imaging features. These entities can be associated (mixed phenotype) or switched from one to the other. If CLAD remains a clinical diagnosis based on spirometry, computed tomography (CT) scan plays an important role in the diagnosis and follow-up of CLAD patients, to exclude identifiable causes of functional decline when CLAD is first suspected, to detect early abnormalities that can precede the diagnosis of CLAD (particularly RAS), to differentiate between the obstructive and restrictive phenotypes, and to detect exacerbations and evolution from one phenotype to the other. Recognition of early signs of rejection is crucial for better understanding of physiopathologic pathways and optimal management of patients.

Automated AI-Driven CT Quantification of Lung Disease Predicts Adverse Outcomes in Patients Hospitalized for COVID-19 Pneumonia.

The purpose of our work was to assess the independent and incremental value of AI-derived quantitative determination of lung lesions extent on initial CT scan for the prediction of clinical deterioration or death in patients hospitalized with COVID-19 pneumonia. 323 consecutive patients (mean age 65 ± 15 years, 192 men), with laboratory-confirmed COVID-19 and an abnormal chest CT scan, were admitted to the hospital between March and December 2020. The extent of consolidation and all lung opacities were quantified on an initial CT scan using a 3D automatic AI-based software. The outcome was known for all these patients. 85 (26.3%) patients died or experienced clinical deterioration, defined as intensive care unit admission. In multivariate regression based on clinical, biological and CT parameters, the extent of all opacities, and extent of consolidation were independent predictors of adverse outcomes, as were diabetes, heart disease, C-reactive protein, and neutrophils/lymphocytes ratio. The association of CT-derived measures with clinical and biological parameters significantly improved the risk prediction (p = 0.049). Automated quantification of lung disease at CT in COVID-19 pneumonia is useful to predict clinical deterioration or in-hospital death. Its combination with clinical and biological data improves risk prediction.

Automated Quantification of CT Patterns Associated with COVID-19 from Chest CT.

PURPOSE: To present a method that automatically segments and quantifies abnormal CT patterns commonly present in coronavirus disease 2019 (COVID-19), namely ground glass opacities and consolidations. MATERIALS AND METHODS: In this retrospective study, the proposed method takes as input a non-contrasted chest CT and segments the lesions, lungs, and lobes in three dimensions, based on a dataset of 9749 chest CT volumes. The method outputs two combined measures of the severity of lung and lobe involvement, quantifying both the extent of COVID-19 abnormalities and presence of high opacities, based on deep learning and deep reinforcement learning. The first measure of (PO, PHO) is global, while the second of (LSS, LHOS) is lobe-wise. Evaluation of the algorithm is reported on CTs of 200 participants (100 COVID-19 confirmed patients and 100 healthy controls) from institutions from Canada, Europe and the United States collected between 2002-Present (April 2020). Ground truth is established by manual annotations of lesions, lungs, and lobes. Correlation and regression analyses were performed to compare the prediction to the ground truth. RESULTS: Pearson correlation coefficient between method prediction and ground truth for COVID-19 cases was calculated as 0.92 for PO (P < .001), 0.97 for PHO (P < .001), 0.91 for LSS (P < .001), 0.90 for LHOS (P < .001). 98 of 100 healthy controls had a predicted PO of less than 1%, 2 had between 1-2%. Automated processing time to compute the severity scores was 10 seconds per case compared to 30 minutes required for manual annotations. CONCLUSION: A new method segments regions of CT abnormalities associated with COVID-19 and computes (PO, PHO), as well as (LSS, LHOS) severity scores.

Automated Quantification of CT Patterns Associated with COVID-19 from Chest CT.

PURPOSE: To present a method that automatically segments and quantifies abnormal CT patterns commonly present in coronavirus disease 2019 (COVID-19), namely ground glass opacities and consolidations. MATERIALS AND METHODS: In this retrospective study, the proposed method takes as input a non-contrasted chest CT and segments the lesions, lungs, and lobes in three dimensions, based on a dataset of 9749 chest CT volumes. The method outputs two combined measures of the severity of lung and lobe involvement, quantifying both the extent of COVID-19 abnormalities and presence of high opacities, based on deep learning and deep reinforcement learning. The first measure of (PO, PHO) is global, while the second of (LSS, LHOS) is lobewise. Evaluation of the algorithm is reported on CTs of 200 participants (100 COVID-19 confirmed patients and 100 healthy controls) from institutions from Canada, Europe and the United States collected between 2002-Present (April, 2020). Ground truth is established by manual annotations of lesions, lungs, and lobes. Correlation and regression analyses were performed to compare the prediction to the ground truth. RESULTS: Pearson correlation coefficient between method prediction and ground truth for COVID-19 cases was calculated as 0.92 for PO (P < .001), 0.97 for PHO(P < .001), 0.91 for LSS (P < .001), 0.90 for LHOS (P < .001). 98 of 100 healthy controls had a predicted PO of less than 1%, 2 had between 1-2%. Automated processing time to compute the severity scores was 10 seconds per case compared to 30 minutes required for manual annotations. CONCLUSION: A new method segments regions of CT abnormalities associated with COVID-19 and computes (PO, PHO), as well as (LSS, LHOS) severity scores.