A deep learning-based application for COVID-19 diagnosis on CT: The Imaging COVID-19 AI initiative.

BACKGROUND: Recently, artificial intelligence (AI)-based applications for chest imaging have emerged as potential tools to assist clinicians in the diagnosis and management of patients with coronavirus disease 2019 (COVID-19). OBJECTIVES: To develop a deep learning-based clinical decision support system for automatic diagnosis of COVID-19 on chest CT scans. Secondarily, to develop a complementary segmentation tool to assess the extent of lung involvement and measure disease severity. METHODS: The Imaging COVID-19 AI initiative was formed to conduct a retrospective multicentre cohort study including 20 institutions from seven different European countries. Patients with suspected or known COVID-19 who underwent a chest CT were included. The dataset was split on the institution-level to allow external evaluation. Data annotation was performed by 34 radiologists/radiology residents and included quality control measures. A multi-class classification model was created using a custom 3D convolutional neural network. For the segmentation task, a UNET-like architecture with a backbone Residual Network (ResNet-34) was selected. RESULTS: A total of 2,802 CT scans were included (2,667 unique patients, mean [standard deviation] age = 64.6 [16.2] years, male/female ratio 1.3:1). The distribution of classes (COVID-19/Other type of pulmonary infection/No imaging signs of infection) was 1,490 (53.2%), 402 (14.3%), and 910 (32.5%), respectively. On the external test dataset, the diagnostic multiclassification model yielded high micro-average and macro-average AUC values (0.93 and 0.91, respectively). The model provided the likelihood of COVID-19 vs other cases with a sensitivity of 87% and a specificity of 94%. The segmentation performance was moderate with Dice similarity coefficient (DSC) of 0.59. An imaging analysis pipeline was developed that returned a quantitative report to the user. CONCLUSION: We developed a deep learning-based clinical decision support system that could become an efficient concurrent reading tool to assist clinicians, utilising a newly created European dataset including more than 2,800 CT scans.

Systematic review of overlapping microRNA patterns in COVID-19 and idiopathic pulmonary fibrosis.

BACKGROUND: Pulmonary fibrosis is an emerging complication of SARS-CoV-2 infection. In this study, we speculate that patients with COVID-19 and idiopathic pulmonary fibrosis (IPF) may share aberrant expressed microRNAs (miRNAs) associated to the progression of lung fibrosis. OBJECTIVE: To identify miRNAs presenting similar alteration in COVID-19 and IPF, and describe their impact on fibrogenesis. METHODS: A systematic review of the literature published between 2010 and January 2022 (PROSPERO, CRD42022341016) was conducted using the key words (COVID-19 OR SARS-CoV-2) AND (microRNA OR miRNA) or (idiopathic pulmonary fibrosis OR IPF) AND (microRNA OR miRNA) in Title/Abstract. RESULTS: Of the 1988 references considered, 70 original articles were appropriate for data extraction: 27 studies focused on miRNAs in COVID-19, and 43 on miRNAs in IPF. 34 miRNAs were overlapping in COVID-19 and IPF, 7 miRNAs presenting an upregulation (miR-19a-3p, miR-200c-3p, miR-21-5p, miR-145-5p, miR-199a-5p, miR-23b and miR-424) and 9 miRNAs a downregulation (miR-17-5p, miR-20a-5p, miR-92a-3p, miR-141-3p, miR-16-5p, miR-142-5p, miR-486-5p, miR-708-3p and miR-150-5p). CONCLUSION: Several studies reported elevated levels of profibrotic miRNAs in COVID-19 context. In addition, the balance of antifibrotic miRNAs responsible of the modulation of fibrotic processes is impaired in COVID-19. This evidence suggests that the deregulation of fibrotic-related miRNAs participates in the development of fibrotic lesions in the lung of post-COVID-19 patients.

Characteristics of Mid-Term Post-Intensive Care Syndrome in Patients Attending a Follow-Up Clinic: A Prospective Comparison Between COVID-19 and Non-COVID-19 Survivors.

At present, it is not clear if critically ill COVID-19 survivors have different needs in terms of follow-up compared with other critically ill survivors, and thus if duplicated post-ICU trajectories are mandatory. OBJECTIVES: To compare the post-intensive care syndrome (PICS) of COVID-19 acute respiratory distress syndrome and non-COVID-19 (NC) survivors referred to a follow-up clinic at 3 months (M3) after ICU discharge. DESIGN SETTING AND PARTICIPANTS: Adults who survived an ICU stay greater than or equal to 7 days and attended the M3 consultation were included in this observational study performed in a post-ICU follow-up clinic of a single tertiary hospital. MAIN OUTCOMES AND MEASURES: Patients underwent a standardized assessment, addressing health-related quality of life (3-level version of EQ-5D), sleep disorders (Pittsburgh Sleep Quality Index [PSQI]), physical status (Barthel index, handgrip and quadriceps strengths), mental health disorders (Hospital Anxiety and Depression Scale and Impact of Event Scale-Revised [IES-R]), and cognitive impairment (Montreal Cognitive Assessment [MoCA]). RESULTS: A total of 143 survivors (86 COVID and 57 NC) attended the M3 consultation. Their median age and severity scores were similar. NC patients had a shorter ICU stay (10 d [8-17.2 d]) compared with COVID group (18 d [10.8-30 d]) (p = 0.001). M3 outcomes were similar in the two groups, except for a higher PSQI (p = 0.038) in the COVID group (6 [3-9.5]) versus NC group (4 [2-7]), and a slightly lower Barthel index in the NC group (100 [100-100]) than in the COVID group (100 [85-100]) (p = 0.026). However, the proportion of patients with abnormal values at each score was similar in the two groups. Health-related quality of life was similar in the two groups. The three MoCA (≥ 26), IES-R (<33), and Barthel (=100) were normal in 58 of 143 patients (40.6%). In contrast, 68.5% (98/143) had not returned to their baseline level of daily activities. CONCLUSIONS AND RELEVANCE: In our follow-up clinic at 3 months after discharge, the proportion of patients presenting alterations in the main PICS domains was similar whether they survived a COVID-19 or another critical illness, despite longer ICU stay in COVID group. Cognition and sleep were the two most affected PICS domains.

Characteristics of Mid-Term Post-Intensive Care Syndrome in Patients Attending a Follow-Up Clinic: A Prospective Comparison Between COVID-19 and Non-COVID-19 Survivors

IMPORTANCE: At present, it is not clear if critically ill COVID-19 survivors have different needs in terms of follow-up compared with other critically ill survivors, and thus if duplicated post-ICU trajectories are mandatory. OBJECTIVES: To compare the post-intensive care syndrome (PICS) of COVID-19 acute respiratory distress syndrome and non-COVID-19 (NC) survivors referred to a follow-up clinic at 3 months (M3) after ICU discharge. DESIGN, SETTING, AND PARTICIPANTS: Adults who survived an ICU stay greater than or equal to 7 days and attended the M3 consultation were included in this observational study performed in a post-ICU follow-up clinic of a single tertiary hospital. MAIN OUTCOMES AND MEASURES: Patients underwent a standardized assessment, addressing health-related quality of life (3-level version of EQ-5D), sleep disorders (Pittsburgh Sleep Quality Index [PSQI]), physical status (Barthel index, handgrip and quadriceps strengths), mental health disorders (Hospital Anxiety and Depression Scale and Impact of Event Scale-Revised [IES-R]), and cognitive impairment (Montreal Cognitive Assessment [MoCA]). RESULTS: A total of 143 survivors (86 COVID and 57 NC) attended the M3 consultation. Their median age and severity scores were similar. NC patients had a shorter ICU stay (10 d [8–17.2 d]) compared with COVID group (18 d [10.8–30 d]) (p = 0.001). M3 outcomes were similar in the two groups, except for a higher PSQI (p = 0.038) in the COVID group (6 [3–9.5]) versus NC group (4 [2–7]), and a slightly lower Barthel index in the NC group (100 [100–100]) than in the COVID group (100 [85–100]) (p = 0.026). However, the proportion of patients with abnormal values at each score was similar in the two groups. Health-related quality of life was similar in the two groups. The three MoCA (≥ 26), IES-R (<33), and Barthel (=100) were normal in 58 of 143 patients (40.6%). In contrast, 68.5% (98/143) had not returned to their baseline level of daily activities. CONCLUSIONS AND RELEVANCE: In our follow-up clinic at 3 months after discharge, the proportion of patients presenting alterations in the main PICS domains was similar whether they survived a COVID-19 or another critical illness, despite longer ICU stay in COVID group. Cognition and sleep were the two most affected PICS domains.

Increased KL-6 levels in moderate to severe COVID-19 infection.

BACKGROUND: The global coronavirus disease 2019 (COVID-19) has presented significant challenges and created concerns worldwide. Besides, patients who have experienced a SARS-CoV-2 infection could present post-viral complications that can ultimately lead to pulmonary fibrosis. Serum levels of Krebs von den Lungen 6 (KL-6), high molecular weight human MUC1 mucin, are increased in the most patients with various interstitial lung damage. Since its production is raised during epithelial damages, KL-6 could be a helpful non-invasive marker to monitor COVID-19 infection and predict post-infection sequelae. METHODS: We retrospectively evaluated KL-6 levels of 222 COVID-19 infected patients and 70 healthy control. Serum KL-6, fibrinogen, lactate dehydrogenase (LDH), platelet-lymphocytes ratio (PLR) levels and other biological parameters were analyzed. This retrospective study also characterized the relationships between serum KL-6 levels and pulmonary function variables. RESULTS: Our results showed that serum KL-6 levels in COVID-19 patients were increased compared to healthy subjects (470 U/ml vs 254 U/ml, P <0.00001). ROC curve analysis enabled us to identify that KL-6 > 453.5 U/ml was associated with COVID-19 (AUC = 0.8415, P < 0.0001). KL-6 level was positively correlated with other indicators of disease severity such as fibrinogen level (r = 0.1475, P = 0.0287), LDH level (r = 0,31, P = 0,004) and PLR level (r = 0.23, P = 0.0005). However, KL-6 levels were not correlated with pulmonary function tests (r = 0.04, P = 0.69). CONCLUSIONS: KL-6 expression was correlated with several disease severity indicators. However, the association between mortality and long-term follow-up outcomes needs further investigation. More extensive trials are required to prove that KL-6 could be a marker of disease severity in COVID-19 infection.

Automatized lung disease quantification in patients with COVID-19 as a predictive tool to assess hospitalization severity.

The pandemic of COVID-19 led to a dramatic situation in hospitals, where staff had to deal with a huge number of patients in respiratory distress. To alleviate the workload of radiologists, we implemented an artificial intelligence (AI) - based analysis named CACOVID-CT, to automatically assess disease severity on chest CT scans obtained from those patients. We retrospectively studied CT scans obtained from 476 patients admitted at the University Hospital of Liege with a COVID-19 disease. We quantified the percentage of COVID-19 affected lung area (% AA) and the CT severity score (total CT-SS). These quantitative measurements were used to investigate the overall prognosis and patient outcome: hospital length of stay (LOS), ICU admission, ICU LOS, mechanical ventilation, and in-hospital death. Both CT-SS and % AA were highly correlated with the hospital LOS, the risk of ICU admission, the risk of mechanical ventilation and the risk of in-hospital death. Thus, CAD4COVID-CT analysis proved to be a useful tool in detecting patients with higher hospitalization severity risk. It will help for management of the patients flow. The software measured the extent of lung damage with great efficiency, thus relieving the workload of radiologists.

Positive Effects of Vitamin D Supplementation in Patients Hospitalized for COVID-19: A Randomized, Double-Blind, Placebo-Controlled Trial.

Retrospective studies showed a relationship between vitamin D status and COVID-19 severity and mortality, with an inverse relation between SARS-CoV-2 positivity and circulating calcifediol levels. The objective of this pilot study was to investigate the effect of vitamin D supplementation on the length of hospital stay and clinical improvement in patients with vitamin D deficiency hospitalized with COVID-19. The study was randomized, double blind and placebo controlled. A total of 50 subjects were enrolled and received, in addition to the best available COVID therapy, either vitamin D (25,000 IU per day over 4 consecutive days, followed by 25,000 IU per week up to 6 weeks) or placebo. The length of hospital stay decreased significantly in the vitamin D group compared to the placebo group (4 days vs. 8 days; p = 0.003). At Day 7, a significantly lower percentage of patients were still hospitalized in the vitamin D group compared to the placebo group (19% vs. 54%; p = 0.0161), and none of the patients treated with vitamin D were hospitalized after 21 days compared to 14% of the patients treated with placebo. Vitamin D significantly reduced the duration of supplemental oxygen among the patients who needed it (4 days vs. 7 days in the placebo group; p = 0.012) and significantly improved the clinical recovery of the patients, as assessed by the WHO scale (p = 0.0048). In conclusion, this study demonstrated that the clinical outcome of COVID-19 patients requiring hospitalization was improved by administration of vitamin D.

Exercise Limitation after Critical Versus Mild COVID-19 Infection: A Metabolic Perspective.

Exercise limitation in COVID-19 survivors is poorly explained. In this retrospective study, cardiopulmonary exercise testing (CPET) was coupled with an oxidative stress assessment in COVID-19 critically ill survivors (ICU group). Thirty-one patients were included in this group. At rest, their oxygen uptake (VO2) was elevated (8 [5.6-9.7] mL/min/kg). The maximum effort was reached at low values of workload and VO2 (66 [40.9-79.2]% and 74.5 [62.6-102.8]% of the respective predicted values). The ventilatory equivalent for carbon dioxide remained within normal ranges. Their metabolic efficiency was low: 15.2 [12.9-17.8]%. The 50% decrease in VO2 after maximum effort was delayed, at 130 [120-170] s, with a still-high respiratory exchange ratio (1.13 [1-1.2]). The blood myeloperoxidase was elevated (92 [75.5-106.5] ng/mL), and the OSS was altered. The CPET profile of the ICU group was compared with long COVID patients after mid-disease (MLC group) and obese patients (OB group). The MLC patients (n = 23) reached peak workload and predicted VO2 values, but their resting VO2, metabolic efficiency, and recovery profiles were similar to the ICU group to a lesser extent. In the OB group (n = 15), no hypermetabolism at rest was observed. In conclusion, the exercise limitation after a critical COVID-19 bout resulted from an altered metabolic profile in the context of persistent inflammation and oxidative stress. Altered exercise and metabolic profiles were also observed in the MLC group. The contribution of obesity on the physiopathology of exercise limitation after a critical bout of COVID-19 did not seem relevant.

AI-Based Chest CT Analysis for Rapid COVID-19 Diagnosis and Prognosis: A Practical Tool to Flag High-Risk Patients and Lower Healthcare Costs.

Early diagnosis of COVID-19 is required to provide the best treatment to our patients, to prevent the epidemic from spreading in the community, and to reduce costs associated with the aggravation of the disease. We developed a decision tree model to evaluate the impact of using an artificial intelligence-based chest computed tomography (CT) analysis software (icolung, icometrix) to analyze CT scans for the detection and prognosis of COVID-19 cases. The model compared routine practice where patients receiving a chest CT scan were not screened for COVID-19, with a scenario where icolung was introduced to enable COVID-19 diagnosis. The primary outcome was to evaluate the impact of icolung on the transmission of COVID-19 infection, and the secondary outcome was the in-hospital length of stay. Using EUR 20000 as a willingness-to-pay threshold, icolung is cost-effective in reducing the risk of transmission, with a low prevalence of COVID-19 infections. Concerning the hospitalization cost, icolung is cost-effective at a higher value of COVID-19 prevalence and risk of hospitalization. This model provides a framework for the evaluation of AI-based tools for the early detection of COVID-19 cases. It allows for making decisions regarding their implementation in routine practice, considering both costs and effects.

The Role of Imaging in the Detection of Non-COVID-19 Pathologies during the Massive Screening of the First Pandemic Wave.

During the COVID-19 pandemic induced by the SARS-CoV-2, numerous chest scans were carried out in order to establish the diagnosis, quantify the extension of lesions but also identify the occurrence of potential pulmonary embolisms. In this perspective, the performed chest scans provided a varied database for a retrospective analysis of non-COVID-19 chest pathologies discovered de novo. The fortuitous discovery of de novo non-COVID-19 lesions was generally not detected by the automated systems for COVID-19 pneumonia developed in parallel during the pandemic and was thus identified on chest CT by the radiologist. The objective is to use the study of the occurrence of non-COVID-19-related chest abnormalities (known and unknown) in a large cohort of patients having suffered from confirmed COVID-19 infection and statistically correlate the clinical data and the occurrence of these abnormalities in order to assess the potential of increased early detection of lesions/alterations. This study was performed on a group of 362 COVID-19-positive patients who were prescribed a CT scan in order to diagnose and predict COVID-19-associated lung disease. Statistical analysis using mean, standard deviation (SD) or median and interquartile range (IQR), logistic regression models and linear regression models were used for data analysis. Results were considered significant at the 5% critical level (p < 0.05). These de novo non-COVID-19 thoracic lesions detected on chest CT showed a significant prevalence in cardiovascular pathologies, with calcifying atheromatous anomalies approaching nearly 35.4% in patients over 65 years of age. The detection of non-COVID-19 pathologies was mostly already known, except for suspicious nodule, thyroid goiter and the ascending thoracic aortic aneurysm. The presence of vertebral compression or signs of pulmonary fibrosis has shown a significant impact on inpatient length of stay. The characteristics of the patients in this sample, both from a demographic and a tomodensitometric point of view on non-COVID-19 pathologies, influenced the length of hospital stay as well as the risk of intra-hospital death. This retrospective study showed that the potential importance of the detection of these non-COVID-19 lesions by the radiologist was essential in the management and the intra-hospital course of the patients.

An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography.

Purpose: In this study, we propose an artificial intelligence (AI) framework based on three-dimensional convolutional neural networks to classify computed tomography (CT) scans of patients with coronavirus disease 2019 (COVID-19), influenza/community-acquired pneumonia (CAP), and no infection, after automatic segmentation of the lungs and lung abnormalities. Methods: The AI classification model is based on inflated three-dimensional Inception architecture and was trained and validated on retrospective data of CT images of 667 adult patients (no infection n=188, COVID-19 n=230, influenza/CAP n=249) and 210 adult patients (no infection n=70, COVID-19 n=70, influenza/CAP n=70), respectively. The model's performance was independently evaluated on an internal test set of 273 adult patients (no infection n=55, COVID-19 n= 94, influenza/CAP n=124) and an external validation set from a different centre (305 adult patients: COVID-19 n=169, no infection n=76, influenza/CAP n=60). Results: The model showed excellent performance in the external validation set with area under the curve of 0.90, 0.92 and 0.92 for COVID-19, influenza/CAP and no infection, respectively. The selection of the input slices based on automatic segmentation of the abnormalities in the lung reduces analysis time (56 s per scan) and computational burden of the model. The Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) score of the proposed model is 47% (15 out of 32 TRIPOD items). Conclusion: This AI solution provides rapid and accurate diagnosis in patients suspected of COVID-19 infection and influenza.

Vital Signs Prediction for COVID-19 Patients in ICU.

This study introduces machine learning predictive models to predict the future values of the monitored vital signs of COVID-19 ICU patients. The main vital sign predictors include heart rate, respiration rate, and oxygen saturation. We investigated the performances of the developed predictive models by considering different approaches. The first predictive model was developed by considering the following vital signs: heart rate, blood pressure (systolic, diastolic and mean arterial, pulse pressure), respiration rate, and oxygen saturation. Similar to the first approach, the second model was developed using the same vital signs, but it was trained and tested based on a leave-one-subject-out approach. The third predictive model was developed by considering three vital signs: heart rate (HR), respiration rate (RR), and oxygen saturation (SpO2). The fourth model was a leave-one-subject-out model for the three vital signs. Finally, the fifth predictive model was developed based on the same three vital signs, but with a five-minute observation rate, in contrast with the aforementioned four models, where the observation rate was hourly to bi-hourly. For the five models, the predicted measurements were those of the three upcoming observations (on average, three hours ahead). Based on the obtained results, we observed that by limiting the number of vital sign predictors (i.e., three vital signs), the prediction performance was still acceptable, with the average mean absolute percentage error (MAPE) being 12%,5%, and 21.4% for heart rate, oxygen saturation, and respiration rate, respectively. Moreover, increasing the observation rate could enhance the prediction performance to be, on average, 8%,4.8%, and 17.8% for heart rate, oxygen saturation, and respiration rate, respectively. It is envisioned that such models could be integrated with monitoring systems that could, using a limited number of vital signs, predict the health conditions of COVID-19 ICU patients in real-time.

Added Value of Dual-Energy CT in COVID-19 Pneumopathy.

Teaching point: The use of dual-energy instead of conventional single-energy computed tomography pulmonary angiogram can provide additional value concerning the diagnosis of COVID-19 and its complications, especially in the detection of small pulmonary embolism.

Cardiopulmonary Exercise Testing in Critically Ill Coronavirus Disease 2019 Survivors: Evidence of a Sustained Exercise Intolerance and Hypermetabolism.

To investigate exercise capacity at 3 and 6 months after a prolonged ICU stay. DESIGN: Observational monocentric study. SETTING: A post-ICU follow-up clinic in a tertiary university hospital in Liège, Belgium. PATIENTS: Patients surviving an ICU stay greater than or equal to 7 days for a severe coronavirus disease 2019 pneumonia and attending our post-ICU follow-up clinic. MEASUREMENTS AND MAIN RESULTS: Cardiopulmonary and metabolic variables provided by a cardiopulmonary exercise testing on a cycle ergometer were collected at rest, at peak exercise, and during recovery. Fourteen patients (10 males, 59 yr [52-62 yr], all obese with body mass index > 27 kg/m2) were included after a hospital stay of 40 days (35-53 d). At rest, respiratory quotient was abnormally high at both 3 and 6 months (0.9 [0.83-0.96] and 0.94 [0.86-0.97], respectively). Oxygen uptake was also abnormally increased at 3 months (8.24 mL/min/kg [5.38-10.54 mL/min/kg]) but significantly decreased at 6 months (p = 0.013). At 3 months, at the maximum workload (67% [55-89%] of predicted workload), oxygen uptake peaked at 81% (64-104%) of predicted maximum oxygen uptake, with oxygen pulse and heart rate reaching respectively 110% (76-140%) and 71% (64-81%) of predicted maximum values. Ventilatory equivalent for carbon dioxide remains within normal ranges. The 50% decrease in oxygen uptake after maximum effort was delayed, at 130 seconds (115-142 s). Recovery was incomplete with a persistent anaerobic metabolism. At 6 months, no significant improvement was observed, excepting an increase in heart rate reaching 79% (72-95%) (p = 0.008). CONCLUSIONS: Prolonged reduced exercise capacity was observed up to 6 months in critically ill coronavirus disease 2019 survivors. This disability did not result from residual pulmonary or cardiac dysfunction but rather from a metabolic disorder characterized by a sustained hypermetabolism and an impaired oxygen utilization.

Long-term clinical follow-up of patients suffering from moderate-to-severe COVID-19 infection: a monocentric prospective observational cohort study.

OBJECTIVES: Various symptoms and considerable organ dysfunction persist following infection with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). Uncertainty remains about the potential mid- and long-term health sequelae. This prospective study of patients hospitalized with coronavirus disease 2019 (COVID-19) in Liège University Hospital, Belgium aimed to determine the persistent consequences of COVID-19. METHODS: Patients admitted to the University Hospital of Liège with moderate-to-severe confirmed COVID-19, discharged between 2 March and 1 October 2020, were recruited prospectively. Follow-up at 3 and 6 months after hospital discharge included demographic and clinical data, biological data, pulmonary function tests (PFTs) and high-resolution computed tomography (CT) scans of the chest. RESULTS: In total, 199 individuals were included in the analysis. Most patients received oxygen supplementation (80.4%). Six months after discharge, 47% and 32% of patients still had exertional dyspnoea and fatigue. PFTs at 3-month follow-up revealed a reduced diffusion capacity of carbon monoxide (mean 71.6 ± 18.6%), and this increased significantly at 6-month follow-up (P<0.0001). Chest CT scans showed a high prevalence (68.9% of the cohort) of persistent abnormalities, mainly ground glass opacities. Duration of hospitalization, intensive care unit admission and mechanical ventilation were not associated with the persistence of symptoms 3 months after discharge. CONCLUSION: The prevalence of persistent symptoms following hospitalization with COVID-19 is high and stable for up to 6 months after discharge. However, biological, functional and iconographic abnormalities improved significantly over time.

The SARS-CoV-2 viral load in COVID-19 patients is lower on face mask filters than on nasopharyngeal swabs.

Face masks and personal respirators are used to curb the transmission of SARS-CoV-2 in respiratory droplets; filters embedded in some personal protective equipment could be used as a non-invasive sample source for applications, including at-home testing, but information is needed about whether filters are suited to capture viral particles for SARS-CoV-2 detection. In this study, we generated inactivated virus-laden aerosols of 0.3-2 microns in diameter (0.9 µm mean diameter by mass) and dispersed the aerosolized viral particles onto electrostatic face mask filters. The limit of detection for inactivated coronaviruses SARS-CoV-2 and HCoV-NL63 extracted from filters was between 10 to 100 copies/filter for both viruses. Testing for SARS-CoV-2, using face mask filters and nasopharyngeal swabs collected from hospitalized COVID-19-patients, showed that filter samples offered reduced sensitivity (8.5% compared to nasopharyngeal swabs). The low concordance of SARS-CoV-2 detection between filters and nasopharyngeal swabs indicated that number of viral particles collected on the face mask filter was below the limit of detection for all patients but those with the highest viral loads. This indicated face masks are unsuitable to replace diagnostic nasopharyngeal swabs in COVID-19 diagnosis. The ability to detect nucleic acids on face mask filters may, however, find other uses worth future investigation.

Can predicting COVID-19 mortality in a European cohort using only demographic and comorbidity data surpass age-based prediction: An externally validated study.

OBJECTIVE: To establish whether one can build a mortality prediction model for COVID-19 patients based solely on demographics and comorbidity data that outperforms age alone. Such a model could be a precursor to implementing smart lockdowns and vaccine distribution strategies. METHODS: The training cohort comprised 2337 COVID-19 inpatients from nine hospitals in The Netherlands. The clinical outcome was death within 21 days of being discharged. The features were derived from electronic health records collected during admission. Three feature selection methods were used: LASSO, univariate using a novel metric, and pairwise (age being half of each pair). 478 patients from Belgium were used to test the model. All modeling attempts were compared against an age-only model. RESULTS: In the training cohort, the mortality group's median age was 77 years (interquartile range = 70-83), higher than the non-mortality group (median = 65, IQR = 55-75). The incidence of former/active smokers, male gender, hypertension, diabetes, dementia, cancer, chronic obstructive pulmonary disease, chronic cardiac disease, chronic neurological disease, and chronic kidney disease was higher in the mortality group. All stated differences were statistically significant after Bonferroni correction. LASSO selected eight features, novel univariate chose five, and pairwise chose none. No model was able to surpass an age-only model in the external validation set, where age had an AUC of 0.85 and a balanced accuracy of 0.77. CONCLUSION: When applied to an external validation set, we found that an age-only mortality model outperformed all modeling attempts (curated on www.covid19risk.ai) using three feature selection methods on 22 demographic and comorbid features.

Circulating Nucleosomes as Potential Markers to Monitor COVID-19 Disease Progression.

The severity of coronavirus disease 2019 (COVID-19) varies significantly with cases spanning from asymptomatic to lethal with a subset of individuals developing Severe Acute Respiratory Syndrome (SARS) and death from respiratory failure. To determine whether global nucleosome and citrullinated nucleosome levels were elevated in COVID-19 patients, we tested two independent cohorts of COVID-19 positive patients with quantitative nucleosome immunoassays and found that nucleosomes were highly elevated in plasma of COVID-19 patients with a severe course of the disease relative to healthy controls and that both histone 3.1 variant and citrullinated nucleosomes increase with disease severity. Elevated citrullination of circulating nucleosomes is indicative of neutrophil extracellular trap formation, neutrophil activation and NETosis in severely affected individuals. Importantly, using hospital setting (outpatient, inpatient or ICU) as a proxy for disease severity, nucleosome levels increased with disease severity and may serve as a guiding biomarker for treatment. Owing to the limited availability of mechanical ventilators and extracorporal membrane oxygenation (ECMO) equipment, there is an urgent need for effective tools to rapidly assess disease severity and guide treatment selection. Based on our studies of two independent cohorts of COVID-19 patients from Belgium and Germany, we suggest further investigation of circulating nucleosomes and citrullination as biomarkers for clinical triage, treatment allocation and clinical drug discovery.

Development and Validation of an Automated Radiomic CT Signature for Detecting COVID-19.

The coronavirus disease 2019 (COVID-19) outbreak has reached pandemic status. Drastic measures of social distancing are enforced in society and healthcare systems are being pushed to and beyond their limits. To help in the fight against this threat on human health, a fully automated AI framework was developed to extract radiomics features from volumetric chest computed tomography (CT) exams. The detection model was developed on a dataset of 1381 patients (181 COVID-19 patients plus 1200 non COVID control patients). A second, independent dataset of 197 RT-PCR confirmed COVID-19 patients and 500 control patients was used to assess the performance of the model. Diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC). The model had an AUC of 0.882 (95% CI: 0.851-0.913) in the independent test dataset (641 patients). The optimal decision threshold, considering the cost of false negatives twice as high as the cost of false positives, resulted in an accuracy of 85.18%, a sensitivity of 69.52%, a specificity of 91.63%, a negative predictive value (NPV) of 94.46% and a positive predictive value (PPV) of 59.44%. Benchmarked against RT-PCR confirmed cases of COVID-19, our AI framework can accurately differentiate COVID-19 from routine clinical conditions in a fully automated fashion. Thus, providing rapid accurate diagnosis in patients suspected of COVID-19 infection, facilitating the timely implementation of isolation procedures and early intervention.