Visual and software-based quantitative chest CT assessment of COVID-19: correlation with clinical findings

Comparing Visual and Software-Based Quantitative Assessment Scores of Lungs' Parenchymal Involvement Quantification in COVID-19 Patients

(1) Background: Computed tomography (CT) plays a paramount role in the characterization and follow-up of COVID-19. Several score systems have been implemented to properly assess the lung parenchyma involved in patients suffering from SARS-CoV-2 infection, such as the visual quantitative assessment score (VQAS) and software-based quantitative assessment score (SBQAS) to help in managing patients with SARS-CoV-2 infection. This study aims to investigate and compare the diagnostic accuracy of the VQAS and SBQAS with two different types of software based on artificial intelligence (AI) in patients affected by SARS-CoV-2. (2) Methods: This is a retrospective study; a total of 90 patients were enrolled with the following criteria: patients' age more than 18 years old, positive test for COVID-19 and unenhanced chest CT scan obtained between March and June 2021. The VQAS was independently assessed, and the SBQAS was performed with two different artificial intelligence-driven software programs (Icolung and CT-COPD). The Intraclass Correlation Coefficient (ICC) statistical index and Bland-Altman Plot were employed. (3) Results: The agreement scores between radiologists (R1 and R2) for the VQAS of the lung parenchyma involved in the CT images were good (ICC = 0.871). The agreement score between the two software types for the SBQAS was moderate (ICC = 0.584). The accordance between Icolung and the median of the visual evaluations (Median R1-R2) was good (ICC = 0.885). The correspondence between CT-COPD and the median of the VQAS (Median R1-R2) was moderate (ICC = 0.622). (4) Conclusions: This study showed moderate and good agreement upon the VQAS and the SBQAS; enhancing this approach as a valuable tool to manage COVID-19 patients and the combination of AI tools with physician expertise can lead to the most accurate diagnosis and treatment plans for patients.

Chest CT scan predictors of intensive care unit admission in hospitalized pregnant women with COVID-19: a case-control study

PURPOSE

To investigate the role of chest computed tomography (CT) scan in the prediction of admission of pregnant women with COVID-19 into intensive care unit (ICU).

METHODS

This was a single-center retrospective case-control study. We included pregnant women diagnosed with COVID-19 by reverse transcriptase polymerase chain reaction between February 2020 and July 2021, requiring hospital admission due to symptoms, who also had a CT chest scan at presentation. Patients admitted to the ICU (case group) were compared with patients who did not require ICU admission (control group). The CT scans were reported by an experienced radiologist, blinded to the patient's course and outcome, aided by an artificial intelligence software. Total CT scan score, chest CT severity score (CT-SS), total lung volume (TLV), infected lung volume (ILV), and infected-to-total lung volume ratio (ILV/TLV) were calculated. Receiver operating characteristic curves were constructed to test the sensitivity and specificity of each parameter.

RESULTS

8/28 patients (28.6%) required ICU admission. These also had lower TLV, higher ILV, and ILV/TLV. The area under the curve (AUC) for these three parameters was 0.789, 0.775, and 0.763, respectively. TLV, ILV, and ILV/TLV had good sensitivity (62.5%, 87.5%, and 87.5%, respectively) and specificity (84.2%, 70%, and 73.7%, respectively) for predicting ICU admission at the following selected thresholds: 2255 mL, 319 mL, and 14%, respectively. The performance of CT-SS, CT scan score, and ILV/TLV in predicting ICU admission was comparable.

CONCLUSION

TLV, ILV, and ILV/TLV as measured by an artificial intelligence software on chest CT, may predict ICU admission in hospitalized pregnant women, symptomatic for COVID-19.

Quantitative analysis of chest computed tomography of COVID-19 pneumonia using a software widely used in Japan

This study aimed to determine the optimal conditions to measure the percentage of the area considered as pneumonia (pneumonia volume ratio [PVR]) and the computed tomography (CT) score due to coronavirus disease 2019 (COVID-19) using the Ziostation2 image analysis software (Z2; Ziosoft, Tokyo, Japan), which is popular in Japan, and to evaluate its usefulness for assessing the clinical severity. We included 53 patients (41 men and 12 women, mean age: 61.3 years) diagnosed with COVID-19 using polymerase chain reaction who had undergone chest CT and were hospitalized between January 2020 and January 2021. Based on the COVID-19 infection severity, the patients were classified as mild (n = 38) or severe (n = 15). For 10 randomly selected samples, the PVR and CT scores by Z2 under different conditions and the visual simple PVR and CT scores were compared. The conditions with the highest statistical agreement were determined. The usefulness of the clinical severity assessment based on the PVR and CT scores using Z2 under the determined conditions was statistically evaluated. The best agreement with the visual measurement was achieved by the Z2 measurement condition of ≥-600 HU. The areas under the receiver operating characteristic curves, Youden's index, and the sensitivity, specificity, and p-values of the PVR and CT scores by Z2 were as follows: PVR: 0.881, 18.69, 66.7, 94.7, and <0.001; CT score: 0.77, 7.5, 40, 74, and 0.002, respectively. We determined the optimal condition for assessing the PVR of COVID-19 pneumonia using Z2 and demonstrated that the AUC of the PVR was higher than that of CT scores in the assessment of clinical severity. The introduction of new technologies is time-consuming and expensive; our method has high clinical utility and can be promptly used in any facility where Z2 has been introduced.

COVID-19 pneumonia: lessons learned, challenges, and preparing for the future

Coronavirus disease 2019 (COVID-19) is a viral disease that causes life-threatening health problems during acute illness, causing a pandemic and millions of deaths. Although computed tomography (CT) was used as a diagnostic tool for COVID-19 in the early period of the pan demic due to the inaccessibility or long duration of the polymerase chain reaction tests, cur rent studies have revealed that CT scan should not be used to diagnose COVID-19. However, radiologic findings are vital in assessing pneumonia severity and investigating complications in patients with COVID-19. Long-term symptoms, also known as long COVID, in people recovering from COVID-19 affect patients' quality of life and cause global health problems. Herein, we aimed to summarize the lessons learned in COVID-19 pneumonia, the challenges in diagnosing the disease and complications, and the prospects for future studies.

Two in one: Overlapping CT findings of COVID-19 and underlying lung diseases

Coronavirus disease 2019 (COVID-19) is associated with pneumonia and has various pulmonary manifestations on computed tomography (CT). Although COVID-19 pneumonia is usually seen as bilateral predominantly peripheral ground-glass opacities with or without consolidation, it can present with atypical radiological findings and resemble the imaging findings of other lung diseases. Diagnosis of COVID-19 pneumonia is much more challenging for both clinicians and radiologists in the presence of pre-existing lung disease. The imaging features of COVID-19 and underlying lung disease can overlap and obscure the findings of each other. Knowledge of the radiological findings of both diseases and possible complications, correct diagnosis, and multidisciplinary consensus play key roles in the appropriate management of diseases. In this pictorial review, the chest CT findings are presented of patients with underlying lung diseases and overlapping COVID-19 pneumonia and the various reasons for radiological lung abnormalities in these patients are discussed.

Admission chest CT findings and risk assessment for stroke-associated pneumonia

Introduction

Stroke-associated pneumonia (SAP) is a significant cause of morbidity and mortality after stroke. Various factors, including dysphagia and stroke severity, are closely related to SAP risk; however, the contribution of the baseline pulmonary parenchymal status to this interplay is an understudied field. Herein, we evaluated the prognostic performance of admission chest computed tomography (CT) findings in predicting SAP.

Methods

We evaluated admission chest CT images, acquired as part of a COVID-19-related institutional policy, in a consecutive series of acute ischemic stroke patients. The pulmonary opacity load at baseline was quantified using automated volumetry and visual scoring algorithms. The relationship between pulmonary opacities with risk of pneumonia within 7 days of symptom onset (i.e., SAP) was evaluated by bivariate and multivariate analyses.

Results

Twenty-three percent of patients in our cohort (n = 100) were diagnosed with SAP. Patients with SAP were more likely to have atrial fibrillation, COPD, severe neurological deficits, and dysphagia. The visual opacity score on chest CT was significantly higher among patients who developed SAP (p = 0.014), while no such relationship was observed in terms of absolute or relative opacity volume. In multivariate analyses, admission stroke severity, presence of dysphagia and a visual opacity score of ≥ 3 (OR 6.37, 95% CI 1.61–25.16; p = 0.008) remained significantly associated with SAP risk.

Conclusions

Pulmonary opacity burden, as evaluated on admission chest CT, is significantly associated with development of pneumonia within initial days of stroke. This association is independent of other well-known predisposing factors for SAP, including age, stroke severity, and presence of dysphagia.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13760-022-02043-7.

Quantitative Chest CT Analysis to Measure Short-Term Sequelae of COVID-19 Pneumonia: A Monocentric Prospective Study

(1) Background: Quantitative CT analysis (QCT) has demonstrated promising results in the prognosis prediction of patients affected by COVID-19. We implemented QCT not only at diagnosis but also at short-term follow-up, pairing it with a clinical examination in search of a correlation between residual respiratory symptoms and abnormal QCT results. (2) Methods: In this prospective monocentric trial performed during the “first wave” of the Italian pandemic, i.e., from March to May 2020, we aimed to test the relationship between %deltaCL (variation of %CL-compromised lung volume) and variations of symptoms-dyspnea, cough and chest pain-at follow-up clinical assessment after hospitalization. (3) Results: 282 patients (95 females, 34%) with a median age of 60 years (IQR, 51–69) were included. We reported a correlation between changing lung abnormalities measured by QCT, and residual symptoms at short-term follow up after COVID-19 pneumonia. Independently from age, a low percentage of surviving patients (1–4%) may present residual respiratory symptoms at approximately two months after discharge. QCT was able to quantify the extent of residual lung damage underlying such symptoms, as the reduction of both %PAL (poorly aerated lung) and %CL volumes was correlated to their disappearance. (4) Conclusions QCT may be used as an objective metric for the measurement of COVID-19 sequelae.

CT-based severity assessment for COVID-19 using weakly supervised non-local CNN

Evaluating patient criticality is the foremost step in administering appropriate COVID-19 treatment protocols. Learning an Artificial Intelligence (AI) model from clinical data for automatic risk-stratification enables accelerated response to patients displaying critical indicators. Chest CT manifestations including ground-glass opacities and consolidations are a reliable indicator for prognostic studies and show variability with patient condition. To this end, we propose a novel attention framework to estimate COVID-19 severity as a regression score from a weakly annotated CT scan dataset. It takes a non-locality approach that correlates features across different parts and spatial scales of the 3D scan. An explicit guidance mechanism from limited infection labeling drives attention refinement and feature modulation. The resulting encoded representation is further enriched through cross-channel attention. The attention model also infuses global contextual awareness into the deep voxel features by querying the base CT scan to mine relevant features. Consequently, it learns to effectively localize its focus region and chisel out the infection precisely. Experimental validation on the MosMed dataset shows that the proposed architecture has significant potential in augmenting existing methods as it achieved a 0.84 R-squared score and 0.133 mean absolute difference.

Prognostic findings for ICU admission in patients with COVID-19 pneumonia: baseline and follow-up chest CT and the added value of artificial intelligence

Infection with SARS-CoV-2 has dominated discussion and caused global healthcare and economic crisis over the past 18 months. Coronavirus disease 19 (COVID-19) causes mild-to-moderate symptoms in most individuals. However, rapid deterioration to severe disease with or without acute respiratory distress syndrome (ARDS) can occur within 1-2 weeks from the onset of symptoms in a proportion of patients. Early identification by risk stratifying such patients who are at risk of severe complications of COVID-19 is of great clinical importance. Computed tomography (CT) is widely available and offers the potential for fast triage, robust, rapid, and minimally invasive diagnosis: Ground glass opacities (GGO), crazy-paving pattern (GGO with superimposed septal thickening), and consolidation are the most common chest CT findings in COVID pneumonia. There is growing interest in the prognostic value of baseline chest CT since an early risk stratification of patients with COVID-19 would allow for better resource allocation and could help improve outcomes. Recent studies have demonstrated the utility of baseline chest CT to predict intensive care unit (ICU) admission in patients with COVID-19. Furthermore, developments and progress integrating artificial intelligence (AI) with computer-aided design (CAD) software for diagnostic imaging allow for objective, unbiased, and rapid assessment of CT images.

Biomarker evaluation for prognostic stratification of patients with COVID-19: the added value of quantitative chest CT

Aim: Pulmonary disease burden and biomarkers are possible predictors of outcomes in patients with COVID-19 and provide complementary information. In this study, the prognostic value of adding quantitative chest computed tomography (CT) to a multiple biomarker approach was evaluated among 148 hospitalized patients with confirmed COVID-19. Materials & methods: Patients admitted between March and July 2020 who were submitted to chest CT and biomarker measurement (troponin I, D-dimer and C-reactive protein) were retrospectively analyzed. Biomarker and tomographic data were compared and associated with death and intensive care unit admission. Results: The number of elevated biomarkers was significantly associated with greater opacification percentages, lower lung volumes and higher death and intensive care unit admission rates. Total lung volume <3.0 l provided further stratification for mortality when combined with biomarker evaluation. Conclusion: Adding automated CT data to a multiple biomarker approach may provide a simple strategy for enhancing risk stratification of patients with COVID-19.

Comparison of semiquantitative chest CT scoring systems to estimate severity in coronavirus disease 2019 (COVID-19) pneumonia

OBJECTIVES

To compare the clinical usefulness among three different semiquantitative computed tomography (CT) severity scoring systems for coronavirus disease 2019 (COVID-19) pneumonia.

METHODS

Two radiologists independently reviewed chest CT images in 108 patients to rate three CT scoring systems (total CT score [TSS], chest CT score [CCTS], and CT severity score [CTSS]). We made a minor modification to CTSS. Quantitative dense area ratio (QDAR: the ratio of lung involvement to lung parenchyma) was calculated using the U-net model. Clinical severity at admission was classified as severe (n = 14) or mild (n = 94). Interobserver agreement, interpretation time, and degree of correlation with clinical severity as well as QDAR were evaluated.

RESULTS

Interobserver agreement was excellent (intraclass correlation coefficient: 0.952-0.970, p < 0.001). Mean interpretation time was significantly longer in CTSS (48.9-80.0 s) than in TSS (25.7-41.7 s, p < 0.001) and CCTS (27.7-39.5 s, p < 0.001). Area under the curve for differentiating clinical severity at admission was 0.855-0.842 in TSS, 0.853-0.850 in CCTS, and 0.853-0.836 in CTSS. All scoring systems correlated with QDAR in the order of CCTS (ρ = 0.443-0.448), TSS (ρ = 0.435-0.437), and CTSS (ρ = 0.415-0.426).

CONCLUSIONS

All semiquantitative scoring systems demonstrated substantial diagnostic performance for clinical severity at admission with excellent interobserver agreement. Interpretation time was significantly shorter in TSS and CCTS than in CTSS. The correlation between the scoring system and QDAR was highest in CCTS, followed by TSS and CTSS. CCTS appeared to be the most appropriate CT scoring system for clinical practice.

KEY POINTS

• Three semiquantitative scoring systems demonstrate substantial accuracy (area under the curve: 0.836-0.855) for diagnosing clinical severity at admission and (area under the curve: 0.786-0.802) for risk of developing critical illness. • Total CT score (TSS) and chest CT score (CCTS) were considered to be more appropriate in terms of clinical usefulness as compared with CT severity score (CTSS), given the shorter interpretation time in TSS and CCTS, and the lowest correlation with quantitative dense area ratio in CTSS. • CCTS is assumed to distinguish subtle from mild lung involvement better than TSS by adopting a 5% threshold in scoring the degree of severity.

AI-Based Quantitative CT Analysis of Temporal Changes According to Disease Severity in COVID-19 Pneumonia

Objective

To quantitatively evaluate computed tomography (CT) parameters of coronavirus disease 2019 (COVID-19) pneumonia an artificial intelligence (AI)-based software in different clinical severity groups during the disease course.

Methods

From March 11 to April 15, 2020, 51 patients (age, 18–84 years; 28 men) diagnosed and hospitalized with COVID-19 pneumonia with a total of 116 CT scans were enrolled in the study. Patients were divided into mild (n = 12), moderate (n = 31), and severe (n = 8) groups based on clinical severity. An AI-based quantitative CT analysis, including lung volume, opacity score, opacity volume, percentage of opacity, and mean lung density, was performed in initial and follow-up CTs obtained at different time points. Receiver operating characteristic analysis was performed to find the diagnostic ability of quantitative CT parameters for discriminating severe from nonsevere pneumonia.

Results

In baseline assessment, the severe group had significantly higher opacity score, opacity volume, higher percentage of opacity, and higher mean lung density than the moderate group (all P ≤ 0.001). Through consecutive time points, the severe group had a significant decrease in lung volume (P = 0.006), a significant increase in total opacity score (P = 0.003), and percentage of opacity (P = 0.007). A significant increase in total opacity score was also observed for the mild group (P = 0.011). Residual opacities were observed in all groups. The involvement of more than 4 lobes (sensitivity, 100%; specificity, 65.26%), total opacity score greater than 4 (sensitivity, 100%; specificity, 64.21), total opacity volume greater than 337.4 mL (sensitivity, 80.95%; specificity, 84.21%), percentage of opacity greater than 11% (sensitivity, 80.95%; specificity, 88.42%), total high opacity volume greater than 10.5 mL (sensitivity, 95.24%; specificity, 66.32%), percentage of high opacity greater than 0.8% (sensitivity, 85.71%; specificity, 80.00%) and mean lung density HU greater than −705 HU (sensitivity, 57.14%; specificity, 90.53%) were related to severe pneumonia.

Conclusions

An AI-based quantitative CT analysis is an objective tool in demonstrating disease severity and can also assist the clinician in follow-up by providing information about the disease course and prognosis according to different clinical severity groups.

The Applications of Artificial Intelligence in Chest Imaging of COVID-19 Patients: A Literature Review

Diagnostic imaging is regarded as fundamental in the clinical work-up of patients with a suspected or confirmed COVID-19 infection. Recent progress has been made in diagnostic imaging with the integration of artificial intelligence (AI) and machine learning (ML) algorisms leading to an increase in the accuracy of exam interpretation and to the extraction of prognostic information useful in the decision-making process. Considering the ever expanding imaging data generated amid this pandemic, COVID-19 has catalyzed the rapid expansion in the application of AI to combat disease. In this context, many recent studies have explored the role of AI in each of the presumed applications for COVID-19 infection chest imaging, suggesting that implementing AI applications for chest imaging can be a great asset for fast and precise disease screening, identification and characterization. However, various biases should be overcome in the development of further ML-based algorithms to give them sufficient robustness and reproducibility for their integration into clinical practice. As a result, in this literature review, we will focus on the application of AI in chest imaging, in particular, deep learning, radiomics and advanced imaging as quantitative CT.

Routine Hematological Parameters May Be Predictors of COVID-19 Severity

To date, coronavirus disease 2019 (COVID-19) has affected over 100 million people globally. COVID-19 can present with a variety of different symptoms leading to manifestation of disease ranging from mild cases to a life-threatening condition requiring critical care-level support. At present, a rapid prediction of disease severity and critical care requirement in COVID-19 patients, in early stages of disease, remains an unmet challenge. Therefore, we assessed whether parameters from a routine clinical hematology workup, at the time of hospital admission, can be valuable predictors of COVID-19 severity and the requirement for critical care. Hematological data from the day of hospital admission (day of positive COVID-19 test) for patients with severe COVID-19 disease (requiring critical care during illness) and patients with non-severe disease (not requiring critical care) were acquired. The data were amalgamated and cleaned and modeling was performed. Using a decision tree model, we demonstrated that routine clinical hematology parameters are important predictors of COVID-19 severity. This proof-of-concept study shows that a combination of activated partial thromboplastin time, white cell count-to-neutrophil ratio, and platelet count can predict subsequent severity of COVID-19 with high sensitivity and specificity (area under ROC 0.9956) at the time of the patient's hospital admission. These data, pending further validation, indicate that a decision tree model with hematological parameters could potentially form the basis for a rapid risk stratification tool that predicts COVID-19 severity in hospitalized patients.

Artificial Intelligence Applications for COVID-19 in Intensive Care and Emergency Settings: A Systematic Review

Background: Little is known about the role of artificial intelligence (AI) as a decisive technology in the clinical management of COVID-19 patients. We aimed to systematically review and critically appraise the current evidence on AI applications for COVID-19 in intensive care and emergency settings. Methods: We systematically searched PubMed, Embase, Scopus, CINAHL, IEEE Xplore, and ACM Digital Library databases from inception to 1 October 2020, without language restrictions. We included peer-reviewed original studies that applied AI for COVID-19 patients, healthcare workers, or health systems in intensive care, emergency, or prehospital settings. We assessed predictive modelling studies and critically appraised the methodology and key findings of all other studies. Results: Of fourteen eligible studies, eleven developed prognostic or diagnostic AI predictive models, all of which were assessed to be at high risk of bias. Common pitfalls included inadequate sample sizes, poor handling of missing data, failure to account for censored participants, and weak validation of models. Conclusions: Current AI applications for COVID-19 are not ready for deployment in acute care settings, given their limited scope and poor quality. Our findings underscore the need for improvements to facilitate safe and effective clinical adoption of AI applications, for and beyond the COVID-19 pandemic.

Prognostic value of CT integrated with clinical and laboratory data during the first peak of the COVID-19 pandemic in Northern Italy: A nomogram to predict unfavorable outcome

Purpose

To evaluate the prognostic role of chest computed tomography (CT), alone or in combination with clinical and laboratory parameters, in COVID-19 patients during the first peak of the pandemic.

Methods

A retrospective single-center study of 301 COVID-19 patients referred to our Emergency Department (ED) from February 25 to March 29, 2020. At presentation, patients underwent chest CT and clinical and laboratory examinations. Outcomes included discharge from the ED after improvement/recovery (positive outcome), or admission to the intensive care unit or death (poor prognosis). A visual quantitative analysis was formed using two scores: the Pulmonary Involvement (PI) score based on the extension of lung involvement, and the Pulmonary Consolidation (PC) score based on lung consolidation. The prognostic value of CT alone or integrated with other parameters was studied by logistic regression and ROC analysis.

Results

The impact of the CT PI score [≥15 vs. ≤ 6] on predicting poor prognosis (OR 5.71 95 % CI 1.93−16.92, P = 0.002) was demonstrated; no significant association was found for the PC score. Chest CT had a prognostic role considering the PI score alone (AUC 0.722) and when evaluated with demographic characteristics, comorbidities, and laboratory data (AUC 0.841). We, therefore, developed a nomogram as an easy tool for immediate clinical application.

Conclusions

Visual analysis of CT gives useful information to physicians for prognostic evaluations, even in conditions of COVID-19 emergency. The predictive value is increased by evaluating CT in combination with clinical and laboratory data.

Visual and quantitative assessment of COVID-19 pneumonia on chest CT: The relationship with disease severity and clinical findings

BACKGROUND

Lungs are the primary organ of involvement of COVID-19, and the severity of pneumonia in COVID-19 patients is an important cause of morbidity and mortality.

AIM

We aimed to evaluate the visual and quantitative pneumonia severity on chest computed tomography (CT) in patients with coronavirus disease 2019 (COVID-19) and compare the CT findings with clinical and laboratory findings.

METHODS

We retrospectively evaluated adult COVID-19 patients who underwent chest CT, clinical scores, laboratory findings, and length of hospital stay. Two independent radiologists visually evaluated the pneumonia severity on chest CT (VSQS). Quantitative CT (QCT) assessment was performed using a free DICOM viewer, and the percentage of the well-aerated lung (%WAL), high-attenuation areas (%HAA) at different threshold values, and mean lung attenuation (MLA) values were calculated. The relationship between CT scores and the clinical, laboratory data, and length of hospital stay were evaluated in this cross-sectional study. The student's t-test and chi-square test were used to analyze the differences between variables. The Pearson correlation test analyzed the correlation between variables. The diagnostic performance of the variables was assessed using receiver operating characteristic (ROC) analysis was used.

RESULTS

The VSQS and QCT scores were significantly correlated with procalcitonin, d-dimer, ferritin, and C-reactive protein levels. Both VSQ and QCT scores were significantly correlated with disease severity (p<0.001). Among the QCT parameters, the %HAA-600 value showed the best correlation with the VSQS (r=730,p<0.001). VSQS and QCT scores had high sensitivity and specificity in distinguishing disease severity and predicting prolonged hospitalization.

CONCLUSION

The VSQS and QCT scores can help manage the COVID-19 and predict the duration of hospitalization.

Independent role of CT chest scan in COVID-19 prognosis: Evidence from the machine learning classification

Background: The current coronavirus disease-19 (COVID-19) pandemic call attention to the key role informatics play in healthcare. The present study discovers an independent role of computerised tomography chest (CT) scans in prognosis of COVID-19 using classification learning algorithms. Methods: In this retrospective study, 57 RT PCR positive COVID-19 patients were enrolled from SMS Medical College, Jaipur (Rajasthan, India) after approval from the Institutional Ethics Committee. A set of 21 features including clinical findings and laboratory parameters and chest CT severity score were recorded. The CT score with mild, moderate and severe categories was chosen as response variable. The dimensionality reduction of feature space was performed and classifiers including, decision trees, K-nearest neighbours, support vector machine and ensemble learning were trained with principal components. The model with highest accuracy and area under the ROC curve (AUC) was selected. Results: The median age of patients was 55 years (range: 20-99 years) with 37 males. The feature space was reduced from 21 to 7 predictors, that included fever, cough, fibrin degradation products, haemoglobin, neutrophil-lymphocyte ratio, ferritin and procalcitonin. The linear support vector machine was chosen as the best classifier with 73.7 % and 0.69 accuracy and AUC for severe CT chest score, respectively. The variance contributed by first three principal components were 97.5 %, 2.4 % and 0.0 %, respectively. Conclusion: In view of low degree of relationships between predictors and chest CT scan severity score category as interpreted from accuracy and AUC it can be concluded that chest CT scan has an independent role in the prognosis of COVID-19 patients.

COVID-19: A qualitative chest CT model to identify severe form of the disease

•

Chest CT helps identify patients with severe COVID-19 using only three qualitative features.

•

A qualitative model based on three qualitative variables can avoid calculating semi-quantitative total CT score.

•

New Early Warning Score 2 is comparable to the CT score for identification of severe forms of COVID-19.

Purpose

The purpose of this study was to identify clinical and chest computed tomography (CT) features associated with a severe form of coronavirus disease 2019 (COVID-19) and to propose a quick and easy to use model to identify patients at risk of a severe form.

Materials and methods

A total of 158 patients with biologically confirmed COVID-19 who underwent a chest CT after the onset of the symptoms were included. There were 84 men and 74 women with a mean age of 68 ± 14 (SD) years (range: 24–96 years). There were 100 non-severe and 58 severe cases. Their clinical data were recorded and the first chest CT examination was reviewed using a computerized standardized report. Univariate and multivariate analyses were performed in order to identify the risk factors associated with disease severity. Two models were built: one was based only on qualitative CT features and the other one included a semi-quantitative total CT score to replace the variable representing the extent of the disease. Areas under the ROC curves (AUC) of the two models were compared with DeLong's method.

Results

Central involvement of lung parenchyma (P < 0.001), area of consolidation (P < 0.008), air bronchogram sign (P < 0.001), bronchiectasis (P < 0.001), traction bronchiectasis (P < 0.011), pleural effusion (P < 0.026), large involvement of either one of the upper lobes or of the middle lobe (P < 0.001) and total CT score ≥ 15 (P < 0.001) were more often observed in the severe group than in the non-severe group. No significant differences were found between the qualitative model (large involvement of either upper lobes or middle lobe [odd ratio (OR) = 2.473], central involvement [OR = 2.760], pleural effusion [OR = 2.699]) and the semi-quantitative model (total CT score ≥ 15 [OR = 3.342], central involvement [OR = 2.344], pleural effusion [OR = 2.754]) with AUC of 0.722 (95% CI: 0.638–0.806) vs. 0.739 (95% CI: 0.656–0.823), respectively (P = 0.209).

Conclusion

We have developed a new qualitative chest CT-based multivariate model that provides independent risk factors associated with severe form of COVID-19.