Dataset Information

Can Lung Imaging Scores and Clinical Variables Predict Severe Course and Fatal Outcome in COVID-19 Pneumonia Patients? A Single-Center Observational Study.

ABSTRACT: COVID-19 prediction models mostly consist of combined clinical features, laboratory parameters, and, less often, chest X-ray (CXR) findings. Our main goal was to propose a prediction model involving imaging methods, specifically ultrasound. This was a single-center, retrospective cohort observational study of patients admitted to the University Hospital Split from November 2020 to May 2021. Imaging protocols were based on the assessment of 14 lung zones for both lung ultrasound (LUS) and computed tomography (CT), correlated to a CXR score assessing 6 lung zones. Prediction models for the necessity of mechanical ventilation (MV) or a lethal outcome were developed by combining imaging, biometric, and biochemical parameters. A total of 255 patients with COVID-19 pneumonia were included in the study. Four independent predictors were added to the regression model for the necessity of MV: LUS score, day of the illness, leukocyte count, and cardiovascular disease (χ2 = 29.16, p < 0.001). The model accurately classified 89.9% of cases. For the lethal outcome, only two independent predictors contributed to the regression model: LUS score and patient's age (χ2 = 48.56, p < 0.001, 93.2% correctly classified). The predictive model identified four key parameters at patient admission which could predict an adverse outcome.

SUBMITTER: Skopljanac I

PROVIDER: S-EPMC9147733 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Can Lung Imaging Scores and Clinical Variables Predict Severe Course and Fatal Outcome in COVID-19 Pneumonia Patients? A Single-Center Observational Study.

Skopljanac Ivan I Pavicic Ivelja Mirela M Budimir Mrsic Danijela D Barcot Ognjen O Jelicic Irena I Domjanovic Josipa J Dolic Kresimir K

Life (Basel, Switzerland) 20220515 5

COVID-19 prediction models mostly consist of combined clinical features, laboratory parameters, and, less often, chest X-ray (CXR) findings. Our main goal was to propose a prediction model involving imaging methods, specifically ultrasound. This was a single-center, retrospective cohort observational study of patients admitted to the University Hospital Split from November 2020 to May 2021. Imaging protocols were based on the assessment of 14 lung zones for both lung ultrasound (LUS) and compute ...[more]

PMID: 35629402

Similar Datasets

Project description:Background and Aims Chest X-ray (CXR) is indispensable to the assessment of severity, diagnosis, and management of pneumonia. Deep learning is an artificial intelligence (AI) technology that has been applied to the interpretation of medical images. This study investigated the feasibility of classifying fatal pneumonia based on CXR images using deep learning models on publicly available platforms. Methods CXR images of patients with pneumonia at diagnosis were labeled as fatal or nonfatal based on medical records. We applied CXR images from 1031 patients with nonfatal pneumonia and 243 patients with fatal pneumonia for training and self-evaluation of the deep learning models. All labeled CXR images were randomly allocated to the training, validation, and test datasets of deep learning models. Data augmentation techniques were not used in this study. We created two deep learning models using two publicly available platforms. Results The first model showed an area under the precision-recall curve of 0.929 with a sensitivity of 50.0% and a specificity of 92.4% for classifying fatal pneumonia. We evaluated the performance of our deep learning models using sensitivity, specificity, PPV, negative predictive value (NPV), accuracy, and F1 score. Using the external validation test dataset of 100 CXR images, the sensitivity, specificity, accuracy, and F1 score were 68.0%, 86.0%, 77.0%, and 74.7%, respectively. In the original dataset, the performance of the second model showed a sensitivity, specificity, and accuracy of 39.6%, 92.8%, and 82.7%, respectively, while external validation showed values of 38.0%, 92.0%, and 65.0%, respectively. The F1 score was 52.1%. These results were comparable to those obtained by respiratory physicians and residents. Conclusions The deep learning models yielded good accuracy in classifying fatal pneumonia. By further improving the performance, AI could assist physicians in the severity assessment of patients with pneumonia.

Project description:BackgroundBoth intracranial pressure (ICP) and the cerebrovascular pressure reactivity represent the dysregulation of pathways directly involved in traumatic brain injury (TBI) pathogenesis and have been used to inform clinical management. However, how these parameters evolve over time following injury and whether this evolution has any prognostic importance have not been studied.Methods and findingsWe analysed the temporal profile of ICP and pressure reactivity index (PRx), examined their relation to TBI-specific mortality, and determined if the prognostic relevance of these parameters was affected by their temporal profile using mixed models for repeated measures of ICP and PRx for the first 240 hours from the time of injury. A total of 601 adults with TBI, admitted between September 2002 to January 2016, and with high-resolution continuous monitoring from a single centre, were studied. At 6 months postinjury, 133 (19%) patients had a fatal outcome; of those, 88 (78%) died from nonsurvivable TBI or brain death. The difference in mean ICP between those with a fatal outcome and functional survivors was only significant for the first 168 hours after injury (all p < 0.05). For PRx, those patients with a fatal outcome also had a higher (more impaired) PRx throughout the first 120 hours after injury (all p < 0.05). The separation of ICP and PRx was greatest in the first 72 hours after injury. Mixed models demonstrated that the explanatory power of the PRx decreases over time; therefore, the prognostic weight assigned to PRx should similarly decrease. However, the ability of ICP to predict a fatal outcome remained relatively stable over time. As control of ICP is the central purpose of TBI management, it is likely that some of the information that is reflected in the natural history of ICP changes is no longer apparent because of therapeutic intervention.ConclusionsWe demonstrated the temporal evolution of ICP and PRx and their relationship with fatal outcome, indicating a potential early prognostic and therapeutic window. The combination of dynamic monitoring variables and their time profile improved prediction of outcome. Therefore, time-driven dynamic modelling of outcome in patients with severe TBI may allow for more accurate and clinically useful prediction models. Further research is needed to confirm and expand on these findings.

Project description:COVID-19 has been reported in over 40million people globally with variable clinical outcomes. In this systematic review and meta-analysis, we assessed demographic, laboratory and clinical indicators as predictors for severe courses of COVID-19. We systematically searched multiple databases (PubMed, Web of Science Core Collection, MedRvix and bioRvix) for publications from December 2019 to May 31 st 2020. Random-effects meta-analyses were used to calculate pooled odds ratios and differences of medians between (1) patients admitted to ICU versus non-ICU patients and (2) patients who died versus those who survived. We adapted an existing Cochrane risk-of-bias assessment tool for outcome studies. Of 6,702 unique citations, we included 88 articles with 69,762 patients. There was concern for bias across all articles included. Age was strongly associated with mortality with a difference of medians (DoM) of 13.15 years (95% confidence interval (CI) 11.37 to 14.94) between those who died and those who survived. We found a clinically relevant difference between non-survivors and survivors for C-reactive protein (CRP; DoM 69.10, CI 50.43 to 87.77), lactate dehydrogenase (LDH; DoM 189.49, CI 155.00 to 223.98), cardiac troponin I (cTnI; DoM 21.88, CI 9.78 to 33.99) and D-Dimer (DoM 1.29mg/L, CI 0.9 - 1.69). Furthermore, cerebrovascular disease was the co-morbidity most strongly associated with mortality (Odds Ratio 3.45, CI 2.42 to 4.91) and ICU admission (Odds Ratio 5.88, CI 2.35 to 14.73). This comprehensive meta-analysis found age, cerebrovascular disease, CRP, LDH and cTnI to be the most important risk-factors in predicting severe COVID-19 outcomes and will inform decision analytical tools to support clinical decision-making. In this systematic review we meta-analyzed 88 articles for risk factors of ICU admission and mortality in COVID-19. We found age, cerebrovascular disease, CRP, LDH and cTnI are the most important risk-factors for ICU admission or mortality.

Dataset Information

Can Lung Imaging Scores and Clinical Variables Predict Severe Course and Fatal Outcome in COVID-19 Pneumonia Patients? A Single-Center Observational Study.

Publications

Can Lung Imaging Scores and Clinical Variables Predict Severe Course and Fatal Outcome in COVID-19 Pneumonia Patients? A Single-Center Observational Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets