Artificial Intelligence Predicts Severity of COVID-19 Based on Correlation of Exaggerated Monocyte Activation, Excessive Organ Damage and Hyperinflammatory Syndrome: A Prospective Clinical Study

Krysko, Olga; Kondakova, Elena; Vershinina, O. S.; Galova, Elena A.; Благонравова, А С; Gorshkova, E. N.; Bachert, Claus; Ivanchenko, Mikhail; Krysko, Dmitri V.; Vedunova, Maria

doi:10.3389/fimmu.2021.715072

Cited by 13 publications

(10 citation statements)

References 76 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…47.7 ±16.2 years, p < 0.02) with mildly increased body mass index (31.6 ±4.9 vs. 27.8 ±4.7, p=0.003) ( Table 1 ). No marked difference in the symptoms was seen on admission between patient groups while severely affected patients had several comorbidities as described in our earlier report ( 22 ). On admission, severely ill patients with COVID-19 had a higher breathing rate than moderately ill patients.…”

Section: Resultssupporting

confidence: 73%

See 1 more Smart Citation

Severity of SARS-CoV-2 infection is associated with high numbers of alveolar mast cells and their degranulation

et al. 2022

Self Cite

View full text Add to dashboard Cite

BackgroundThe systemic inflammatory response post-SARS-CoV-2 infection increases pro-inflammatory cytokine production, multi-organ damage, and mortality rates. Mast cells (MC) modulate thrombo-inflammatory disease progression (e.g., deep vein thrombosis) and the inflammatory response post-infection.ObjectiveTo enhance our understanding of the contribution of MC and their proteases in SARS-CoV-2 infection and the pathogenesis of the disease, which might help to identify novel therapeutic targets.MethodsMC proteases chymase (CMA1), carboxypeptidase A3 (CPA3), and tryptase beta 2 (TPSB2), as well as cytokine levels, were measured in the serum of 60 patients with SARS-CoV-2 infection (30 moderate and 30 severe; severity of the disease assessed by chest CT) and 17 healthy controls by ELISA. MC number and degranulation were quantified by immunofluorescent staining for tryptase in lung autopsies of patients deceased from either SARS-CoV-2 infection or unrelated reasons (control). Immortalized human FcεR1+c-Kit+ LUVA MC were infected with SARS-CoV-2, or treated with its viral proteins, to assess direct MC activation by flow cytometry.ResultsThe levels of all three proteases were increased in the serum of patients with COVID-19, and strongly correlated with clinical severity. The density of degranulated MC in COVID-19 lung autopsies was increased compared to control lungs. The total number of released granules and the number of granules per each MC were elevated and positively correlated with von Willebrand factor levels in the lung. SARS-CoV-2 or its viral proteins spike and nucleocapsid did not induce activation or degranulation of LUVA MC in vitro.ConclusionIn this study, we demonstrate that SARS-CoV-2 is strongly associated with activation of MC, which likely occurs indirectly, driven by the inflammatory response. The results suggest that plasma MC protease levels could predict the disease course, and that severe COVID-19 patients might benefit from including MC-stabilizing drugs in the treatment scheme.

show abstract

Section: Resultssupporting

confidence: 73%

“…The patients in the control group were age-matched healthy volunteers without any acute respiratory symptoms. A detailed description of the study cohort is provided in our earlier report ( 22 ). The patients were grouped into moderate and severe illness according to the COVID-19 treatment guidelines ( 23 ).…”

Section: Methodsmentioning

confidence: 99%

Severity of SARS-CoV-2 infection is associated with high numbers of alveolar mast cells and their degranulation

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Differences in pathogen-related cytokine profiles can be revealed by simple pairwise comparisons of absolute cytokine concentration across different diseases or stimulation conditions, but multidimensional data analysis often provides a more in-depth view ( 12 , 28 , 50 – 52 ). Among various statistical approaches, non-supervised Principal Component Analysis (PCA) has been the most extensively applied, while supervised approaches, namely, PLS-DA, Random Forest, and Support Vector Machine data analysis, have been less widely explored as a means to interpret cytokine profiling ( 45 , 53 – 56 ).…”

Section: Discussionmentioning

confidence: 99%

Standardized Whole Blood Assay and Bead-Based Cytokine Profiling Reveal Commonalities and Diversity of the Response to Bacteria and TLR Ligands in Cattle

et al. 2022

View full text Add to dashboard Cite

Recent developments in multiplex technologies enable the determination of a large nu\mber of soluble proteins such as cytokines in various biological samples. More than a one-by-one determination of the concentration of immune mediators, they permit the establishment of secretion profiles for a more accurate description of conditions related to infectious diseases or vaccination. Cytokine profiling has recently been made available for bovine species with the development of a Luminex® technology-based 15-plex assay. Independently from the manufacturer, we evaluated the bovine cytokine/chemokine multiplex assay for limits of detection, recovery rate, and reproducibility. Furthermore, we assessed cytokine secretion in blood samples from 107 cows upon stimulation with heat-killed bacteria and TLR2/4 ligands compared to a null condition. Secretion patterns were analyzed either using the absolute concentration of cytokines or using their relative concentration with respect to the overall secretion level induced by each stimulus. Using Partial Least Square-Discriminant Analysis, we show that the 15-cytokine profile is different under Escherichia coli, Staphylococcus aureus, and Streptococcus uberis conditions, and that IFN-γ, IL-1β, and TNF-α contribute the most to differentiate these conditions. LPS and E. coli induced largely overlapping biological responses, but S. aureus and S. uberis were associated with distinct cytokine profiles than their respective TLR ligands. Finally, results based on adjusted or absolute cytokine levels yielded similar discriminative power, but led to different stimuli-related signatures.

show abstract

“…Since the emergence of the COVID-19 pandemic, many predictive models concerning the diagnosis and prognosis have emerged [ 5 , 9 , 16 , 29 – 31 ]. In the development of COVID-19 prediction models, artificial intelligence, including machine learning and deep learning, has been widely used to improve the accuracy and expansibility of the prediction models [ 32 ].…”

Section: Discussionmentioning

confidence: 99%

Incubation period, clinical and lung CT features for early prediction of COVID-19 deterioration: development and internal verification of a risk model

Peng

Deng

et al. 2022

BMC Pulm Med

View full text Add to dashboard Cite

Background Most severe, critical, or mortal COVID-19 cases often had a relatively stable period before their status worsened. We developed a deterioration risk model of COVID-19 (DRM-COVID-19) to predict exacerbation risk and optimize disease management on admission. Method We conducted a multicenter retrospective cohort study with 239 confirmed symptomatic COVID-19 patients. A combination of the least absolute shrinkage and selection operator (LASSO), change-in-estimate (CIE) screened out independent risk factors for the multivariate logistic regression model (DRM-COVID-19) from 44 variables, including epidemiological, demographic, clinical, and lung CT features. The compound study endpoint was progression to severe, critical, or mortal status. Additionally, the model's performance was evaluated for discrimination, accuracy, calibration, and clinical utility, through internal validation using bootstrap resampling (1000 times). We used a nomogram and a network platform for model visualization. Results In the cohort study, 62 cases reached the compound endpoint, including 42 severe, 18 critical, and two mortal cases. DRM-COVID-19 included six factors: dyspnea [odds ratio (OR) 4.89;confidence interval (95% CI) 1.53–15.80], incubation period (OR 0.83; 95% CI 0.68–0.99), number of comorbidities (OR 1.76; 95% CI 1.03–3.05), D-dimer (OR 7.05; 95% CI, 1.35–45.7), C-reactive protein (OR 1.06; 95% CI 1.02–1.1), and semi-quantitative CT score (OR 1.50; 95% CI 1.27–1.82). The model showed good fitting (Hosmer–Lemeshow goodness, X2(8) = 7.0194, P = 0.53), high discrimination (the area under the receiver operating characteristic curve, AUROC, 0.971; 95% CI, 0.949–0.992), precision (Brier score = 0.051) as well as excellent calibration and clinical benefits. The precision-recall (PR) curve showed excellent classification performance of the model (AUCPR = 0.934). We prepared a nomogram and a freely available online prediction platform (https://deterioration-risk-model-of-covid-19.shinyapps.io/DRMapp/). Conclusion We developed a predictive model, which includes the including incubation period along with clinical and lung CT features. The model presented satisfactory prediction and discrimination performance for COVID-19 patients who might progress from mild or moderate to severe or critical on admission, improving the clinical prognosis and optimizing the medical resources.

show abstract

Artificial Intelligence Predicts Severity of COVID-19 Based on Correlation of Exaggerated Monocyte Activation, Excessive Organ Damage and Hyperinflammatory Syndrome: A Prospective Clinical Study

Cited by 13 publications

References 76 publications

Severity of SARS-CoV-2 infection is associated with high numbers of alveolar mast cells and their degranulation

Severity of SARS-CoV-2 infection is associated with high numbers of alveolar mast cells and their degranulation

Standardized Whole Blood Assay and Bead-Based Cytokine Profiling Reveal Commonalities and Diversity of the Response to Bacteria and TLR Ligands in Cattle

Incubation period, clinical and lung CT features for early prediction of COVID-19 deterioration: development and internal verification of a risk model

Contact Info

Product

Resources

About