Comparison of chest CT severity scoring systems for COVID-19

Elmokadem, Ali H.; Mounir, Ahmad M.; Ramadan, Zainab A.; Elsedeiq, Mahmoud; Saleh, Gehad A.

doi:10.1007/s00330-021-08432-5

Cited by 45 publications

(42 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our results revealed a statistically significantly higher prevalence of crazy paving pattern in non-survivors compared to survivors, as this pattern was discovered in 17 individuals, all of whom were non-survivors. This prevalence is in agreement with previous literature [ 21 , 39 , 40 ]. However, there was no statistically significant difference in GGO among survivors and non-survivors, which aligns with the results of a previous study [ 41 ].…”

Section: Discussionsupporting

confidence: 94%

“…In this study, all patients had GGOs. This finding is similar to the results of multiple previous studies in which the majority of COVID-19 patients had GGO and mixed GGO with consolidation [ 21 , 37 , 38 ]. However, pulmonary consolidation was found in 71 individuals, with non-survivors having a higher rate of pulmonary consolidation (90% vs. 42.5%, p = 0.001).…”

Section: Discussionsupporting

confidence: 92%

“…The researchers found that a TSS threshold of 7.5 showed a sensitivity of 82.6% and a specificity of 100% for diagnosing severe-critical COVID-19 infection [24]. Another study revealed concordant results, with 77.3% sensitivity and 90.5% specificity using a relatively higher TSS cutoff value of 12 to differentiate severe critical cases from non-severe cases [21].The combination of both qualitative and quantitative assessment could help to distinguish severe critical cases from non-severe cases, and could aid in the rapid diagnosis of and care for critical cases, thereby decreasing mortality and improving patient outcomes.…”

Section: Discussionmentioning

confidence: 99%

“…A variety of chest CT grading systems have been proposed to assess the severity of pulmonary affection in COVID-19 infection. A recent study examined the diagnostic performance of five different CT chest severity scoring systems and found that CT total severity score (TSS) had the highest specificity and utilized the least amount of time when compared to other scoring systems [ 21 ].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Chest CT Total Severity Score on Admission to Predict In-Hospital Mortality in COVID-19 Patients with Acute and Chronic Renal Impairment

Tharwat

Saleh²,

Saleh³

et al. 2022

Diagnostics

Self Cite

View full text Add to dashboard Cite

Aim: To identify the predictors of in-hospital mortality in patients with coronavirus disease of 2019 (COVID-19) and acute renal impairment (ARI) or chronic kidney disease (CKD), and to evaluate the performance and inter-reader concordance of chest CT total severity scores (TSSs). Methods: This retrospective single-center study was conducted on symptomatic COVID-19 patients with renal impairment (either acute or chronic) and a serum creatinine of >2 mg/dL at the time of admission. The patients’ demographic characteristics, clinical data, and laboratory data were extracted from the clinical computerized medical records. All chest CT images obtained at the time of hospital admission were analyzed. Two radiologists independently assessed the pulmonary abnormalities and scored the severity using CT chest total severity score (TSS). Univariate logistic regression analysis was used to determine factors associated with in-hospital mortality. A receiver operating characteristic (ROC) curve analysis was performed for the TSS in order to identify the cut-off point that predicts mortality. Bland–Altman plots were used to evaluate agreement between the two radiologists assessing TSS. Results: A total of 100 patients were included, with a mean age of 60 years, 54 were males, 53 had ARI, and 47 had CKD. In terms of in-hospital mortality, 60 patients were classified in the non-survivor group and 40 were classified in the survivor group. The mortality rate was higher for those with ARI compared to those with CKD (p = 0.033). The univariate regression analysis showed an increasing odds of in-hospital mortality associated with higher respiratory rate (OR 1.149, 95% CI 1.057–1.248, p = 0.001), total bilirubin (OR 2.532, 95% CI 1.099–5.836, p = 0.029), lactate dehydrogenase (LDH) (OR 1.001, 95% CI 1.000–1.003, p = 0.018), CRP (OR 1.010, 95% CI 1.002–1.017, p = 0.012), invasive mechanical ventilation (MV) (OR 7.667, 95% CI 2.118–27.755, p = 0.002), a predominant pattern of pulmonary consolidation (OR 21.714, 95% CI 4.799–98.261, p < 0.001), and high TSS (OR 2.082, 95% CI 1.579–2.745, p < 0.001). The optimum cut-off value of TSS used to predict in-hospital mortality was 8.5 with a sensitivity of 86.7% and a specificity of 87.5%. There was excellent interobserver agreement (ICC > 0.9) between the two independent radiologists in their quantitative assessment of pulmonary changes using TSS. Conclusions: In-hospital mortality is high in COVID-19 patients with ARI/CKD, especially for those with ARI. High serum bilirubin, a predominant pattern of pulmonary consolidation, and TSS are the most significant predictors of mortality in these patients. Patients with a higher TSS may require more intensive hospital care. TSS is a reliable and helpful auxiliary tool for risk stratification among COVID-19 patients with ARI/CKD.

show abstract

Section: Discussionsupporting

confidence: 94%

Section: Discussionsupporting

confidence: 92%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Chest CT Total Severity Score on Admission to Predict In-Hospital Mortality in COVID-19 Patients with Acute and Chronic Renal Impairment

Tharwat

Saleh²,

Saleh³

et al. 2022

Diagnostics

Self Cite

View full text Add to dashboard Cite

show abstract

“…AO was defined as the presence of ground-glass or consolidative opacities, or both. Each segment was also graded semi-quantitatively on a severity score for opacities ranging from 0 to 2 (0 = none, 1 = 1–50%, 2 = 51–100% of alveolar opacification), a previously established rating scheme that proved accurate and time-efficient [ 21 ]. Consequently, because each lung segment was assigned a semi-quantitative score ranging from 0 to 2, AO was measured according to 40 discrete severity levels (minimum severity score, 0; maximum score, 40 points/patient) [ 22 ].…”

Section: Methodsmentioning

confidence: 99%

Acute Pulmonary Embolism in COVID-19: A Potential Connection between Venous Congestion and Thrombus Distribution

et al. 2022

View full text Add to dashboard Cite

Background: Vascular abnormalities, including venous congestion (VC) and pulmonary embolism (PE), have been recognized as frequent COVID-19 imaging patterns and proposed as severity markers. However, the underlying pathophysiological mechanisms remain unclear. In this study, we aimed to characterize the relationship between VC, PE distribution, and alveolar opacities (AO). Methods: This multicenter observational registry (clinicaltrials.gov identifier NCT04824313) included 268 patients diagnosed with SARS-CoV-2 infection and subjected to contrast-enhanced CT between March and June 2020. Acute PE was diagnosed in 61 (22.8%) patients, including 17 females (27.9%), at a mean age of 61.7 ± 14.2 years. Demographic, laboratory, and outcome data were retrieved. We analyzed CT images at the segmental level regarding VC (qualitatively and quantitatively [diameter]), AO (semi-quantitatively as absent, <50%, or >50% involvement), clot location, and distribution related to VC and AO. Segments with vs. without PE were compared. Results: Out of 411 emboli, 82 (20%) were lobar or more proximal and 329 (80%) were segmental or subsegmental. Venous diameters were significantly higher in segments with AO (p = 0.031), unlike arteries (p = 0.138). At the segmental level, 77% of emboli were associated with VC. Overall, PE occurred in 28.2% of segments with AO vs. 21.8% without (p = 0.047). In the absence of VC, however, AO did not affect PE rates (p = 0.94). Conclusions: Vascular changes predominantly affected veins, and most PEs were located in segments with VC. In the absence of VC, AOs were not associated with the PE rate. VC might result from increased flow supported by the hypothesis of pulmonary arteriovenous anastomosis dysregulation as a relevant contributing factor.

show abstract

Differential privacy preserved federated learning for prognostic modeling in COVID‐19 patients using large multi‐institutional chest CT dataset

Shiri,

Salimi,

Sirjani

et al. 2024

Medical Physics

View full text Add to dashboard Cite

BackgroundNotwithstanding the encouraging results of previous studies reporting on the efficiency of deep learning (DL) in COVID‐19 prognostication, clinical adoption of the developed methodology still needs to be improved. To overcome this limitation, we set out to predict the prognosis of a large multi‐institutional cohort of patients with COVID‐19 using a DL‐based model.PurposeThis study aimed to evaluate the performance of deep privacy‐preserving federated learning (DPFL) in predicting COVID‐19 outcomes using chest CT images.MethodsAfter applying inclusion and exclusion criteria, 3055 patients from 19 centers, including 1599 alive and 1456 deceased, were enrolled in this study. Data from all centers were split (randomly with stratification respective to each center and class) into a training/validation set (70%/10%) and a hold‐out test set (20%). For the DL model, feature extraction was performed on 2D slices, and averaging was performed at the final layer to construct a 3D model for each scan. The DensNet model was used for feature extraction. The model was developed using centralized and FL approaches. For FL, we employed DPFL approaches. Membership inference attack was also evaluated in the FL strategy. For model evaluation, different metrics were reported in the hold‐out test sets. In addition, models trained in two scenarios, centralized and FL, were compared using the DeLong test for statistical differences.ResultsThe centralized model achieved an accuracy of 0.76, while the DPFL model had an accuracy of 0.75. Both the centralized and DPFL models achieved a specificity of 0.77. The centralized model achieved a sensitivity of 0.74, while the DPFL model had a sensitivity of 0.73. A mean AUC of 0.82 and 0.81 with 95% confidence intervals of (95% CI: 0.79–0.85) and (95% CI: 0.77–0.84) were achieved by the centralized model and the DPFL model, respectively. The DeLong test did not prove statistically significant differences between the two models (p‐value = 0.98). The AUC values for the inference attacks fluctuate between 0.49 and 0.51, with an average of 0.50 ± 0.003 and 95% CI for the mean AUC of 0.500 to 0.501.ConclusionThe performance of the proposed model was comparable to centralized models while operating on large and heterogeneous multi‐institutional datasets. In addition, the model was resistant to inference attacks, ensuring the privacy of shared data during the training process.

show abstract

Comparison of chest CT severity scoring systems for COVID-19

Cited by 45 publications

References 34 publications

Chest CT Total Severity Score on Admission to Predict In-Hospital Mortality in COVID-19 Patients with Acute and Chronic Renal Impairment

Chest CT Total Severity Score on Admission to Predict In-Hospital Mortality in COVID-19 Patients with Acute and Chronic Renal Impairment

Acute Pulmonary Embolism in COVID-19: A Potential Connection between Venous Congestion and Thrombus Distribution

Differential privacy preserved federated learning for prognostic modeling in COVID‐19 patients using large multi‐institutional chest CT dataset

Contact Info

Product

Resources

About