Evaluation of novel coronavirus disease (COVID-19) using quantitative lung CT and clinical data: prediction of short-term outcome.
Ontology highlight
ABSTRACT: Background
Computed tomography (CT) enables quantification of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, helping in outcome prediction.Methods
From 1 to 22 March 2020, patients with pneumonia symptoms, positive lung CT scan, and confirmed SARS-CoV-2 on reverse transcription-polymerase chain reaction (RT-PCR) were consecutively enrolled. Clinical data was collected. Outcome was defined as favourable or adverse (i.e., need for mechanical ventilation or death) and registered over a period of 10 days following CT. Volume of disease (VoD) on CT was calculated semi-automatically. Multiple linear regression was used to predict VoD by clinical/laboratory data. To predict outcome, important features were selected using a priori analysis and subsequently used to train 4 different models.Results
A total of 106 consecutive patients were enrolled (median age 63.5 years, range 26-95 years; 41/106 women, 38.7%). Median duration of symptoms and C-reactive protein (CRP) was 5 days (range 1-30) and 4.94 mg/L (range 0.1-28.3), respectively. Median VoD was 249.5 cm3 (range 9.9-1505) and was predicted by lymphocyte percentage (p = 0.008) and CRP (p < 0.001). Important variables for outcome prediction included CRP (area under the curve [AUC] 0.77), VoD (AUC 0.75), age (AUC 0.72), lymphocyte percentage (AUC 0.70), coronary calcification (AUC 0.68), and presence of comorbidities (AUC 0.66). Support vector machine had the best performance in outcome prediction, yielding an AUC of 0.92.Conclusions
Measuring the VoD using a simple CT post-processing tool estimates SARS-CoV-2 burden. CT and clinical data together enable accurate prediction of short-term clinical outcome.
SUBMITTER: Matos J
PROVIDER: S-EPMC7318726 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:PURPOSE:To identify and quantify lung changes associated with coronavirus disease-2019 (COVID-19) with quantitative lung CT during the disease. METHODS:This retrospective study reviewed COVID-19 patients who underwent multiple chest CT scans during their disease course. Quantitative lung CT was used to determine the nature and volume of lung involvement. A semi-quantitative scoring system was also used to evaluate lung lesions. RESULTS:This study included eighteen cases (4 cases in mild type, 10 cases in moderate type, 4 cases in severe type, and without critical type cases) with confirmed COVID-19. Patients had a mean hospitalized period of 24.1 ± 7.1 days (range: 14-38 days) and underwent an average CT scans of 3.9 ± 1.6 (range: 2-8). The total volumes of lung abnormalities reached a peak of 8.8 ± 4.1 days (range: 2-14 days). The ground-glass opacity (GGO) volume percentage was higher than the consolidative opacity (CO) volume percentage on the first CT examination (Z = 2.229, P = 0.026), and there was no significant difference between the GGO volume percentage and that of CO at the peak stage (Z = - 0.628, P = 0.53). The volume percentage of lung involvement identified by AI demonstrated a strong correlation with the total CT scores at each stage (r = 0.873, P = 0.0001). CONCLUSIONS:Quantitative lung CT can automatically identify the nature of lung involvement and quantify the dynamic changes of lung lesions on CT during COVID-19. For patients who recovered from COVID-19, GGO was the predominant imaging feature on the initial CT scan, while GGO and CO were the main appearances at peak stage.
| S-EPMC7547301 | biostudies-literature
Project description:BackgroundPrediction of the dynamics of new SARS-CoV-2 infections during the current COVID-19 pandemic is critical for public health planning of efficient health care allocation and monitoring the effects of policy interventions. We describe a new approach that forecasts the number of incident cases in the near future given past occurrences using only a small number of assumptions.MethodsOur approach to forecasting future COVID-19 cases involves 1) modeling the observed incidence cases using a Poisson distribution for the daily incidence number, and a gamma distribution for the series interval; 2) estimating the effective reproduction number assuming its value stays constant during a short time interval; and 3) drawing future incidence cases from their posterior distributions, assuming that the current transmission rate will stay the same, or change by a certain degree.ResultsWe apply our method to predicting the number of new COVID-19 cases in a single state in the U.S. and for a subset of counties within the state to demonstrate the utility of this method at varying scales of prediction. Our method produces reasonably accurate results when the effective reproduction number is distributed similarly in the future as in the past. Large deviations from the predicted results can imply that a change in policy or some other factors have occurred that have dramatically altered the disease transmission over time.ConclusionWe presented a modelling approach that we believe can be easily adopted by others, and immediately useful for local or state planning.
| S-EPMC8046206 | biostudies-literature
Project description:IntroductionWe aimed to assess the feasibility of SPECT and PET Y-90 imaging, and to compare these modalities by visualizing hot and cold foci in phantoms for varying isotope concentrations.Materials and methodsThe data was acquired from the Jaszczak and NEMA phantoms. In the Jaszczak phantom Y-90 concentrations of 0.1 MBq/ml and 0.2 MBq/ml were used, while higher concentrations, up to 1.0 MBq/ml, were simulated by acquisition time extension with respect to the standard clinical protocol of 30 sec/projection for SPECT and 30 min/bed position for PET imaging. For NEMA phantom, the hot foci had concentrations of about 4 MB/ml and the background 0.1 or 0.0 MBq/ml. All of the acquired data was analysed both qualitatively and quantitatively. Qualitative assessment was conducted by six observers asked to identify the number of visible cold or hot foci. Inter-observer agreement was assessed. Quantitative analysis included calculations of contrast and contrast-to-noise ratio (CNR), and comparisons with the qualitative results.ResultsFor SPECT data up to two cold foci were discernible, while for PET four foci were visible. We have shown that CNR (with Rose criterion) is a good measure of foci visibility for both modalities. We also found good concordance of qualitative results for the Jaszczak phantom studies between the observers (corresponding Krippendorf's alpha coefficients of 0.76 to 0.84). In the NEMA phantom without background activity all foci were visible in SPECT/CT images. With isotope in the background, 5 of 6 spheres were discernible (CNR of 3.0 for the smallest foci). For PET studies all hot spheres were visible, regardless of the background activity.ConclusionsPET Y-90 imaging provided better results than Bremsstrahlung based SPECT imaging. This indicates that PET/CT might become the method of choice in Y-90 post radioembolization imaging for visualisation of both necrotic and hot lesions in the liver.
| S-EPMC7875374 | biostudies-literature
Project description:PURPOSE:[18F]-2-Fluoro-2-deoxy-D-glucose PET/CT (FDG PET/CT) is a sensitive and quantitative technic for detecting inflammatory process. Glucose uptake is correlated with an increased anaerobic glycolysis seen in activated inflammatory cells such as monocytes, lymphocytes, and granulocytes. The aim of the study was to assess the inflammatory status at the presumed peak of the inflammatory phase in non-critically ill patients requiring admission for COVID-19. METHODS:Patients admitted with COVID-19 were prospectively enrolled. FDG PET/CT was performed from day 6 to day 14 of the onset of symptoms. Depending on FDG PET/CT findings, patients' profiles were classified as "inflammatory" or "low inflammatory." FDG PET/CT data were compared with chest CT evolution and short-term clinical outcome. All inflammatory sites were reported to screen potential extra-pulmonary tropism. RESULTS:Thirteen patients were included. Maximum standardized uptake values ranged from 4.7 to 16.3 in lungs. All patients demonstrated increased mediastinal lymph nodes glucose uptake. Three patients (23%) presented mild nasopharyngeal, two patients (15%) bone marrow, and five patients (38%) splenic mild increase in glucose uptake. No patient had significant digestive focal or segmental glucose uptake. There was no significant physiological myocardial glucose uptake in all patients except one. There was no correlation between PET lung inflammatory status and chest CT evolution or short-term clinical outcome. CONCLUSION:Inflammatory process at the presumed peak of the inflammatory phase in COVID-19 patients is obvious in FDG PET/CT scans. Glucose uptake is heterogeneous and typically focused on lungs. TRIAL REGISTRATION:NCT04441489. Registered 22 June 2020 (retrospectively registered).
| S-EPMC7382557 | biostudies-literature
Project description:BackgroundTraumatic hemorrhagic contusions are associated with iodine leak; however, quantification of leakage and its importance to outcome is unclear.PurposeTo identify iodine-based dual-energy CT variables that correlate with in-hospital mortality and short-term outcomes for contusions at hospital discharge.Materials and MethodsIn this retrospective study, consecutive patients with contusions from May 2016 through January 2017 were analyzed. Two radiologists evaluated CT variables from unenhanced admission head CT and follow-up head dual-energy CT scans obtained after contrast material-enhanced whole-body CT. The outcomes evaluated were in-hospital mortality, Rancho Los Amigos scale (RLAS) score, and disability rating scale (DRS) score. Logistic regression and linear regression were used to develop prediction models for categorical and continuous outcomes, respectively.ResultsThe study included 65 patients (median age, 48 years; interquartile range, 25-65.5 years); 50 were men. Dual-energy CT variables that correlated with mortality, RLAS score, and DRS score were iodine concentration, pseudohematoma volume, iodine quantity in pseudohematoma, and iodine quantity in contusion. The single-energy CT variable that correlated with mortality, RLAS score, and DRS score was hematoma volume at follow-up CT. Multiple logistic regression analysis after inclusion of clinical variables identified two predictors that enabled determination of mortality: postresuscitation Glasgow coma scale (P-GCS) (adjusted odds ratio, 0.42; 95% confidence interval [CI]: 0.2, 0.86; P = 0.01) and iodine quantity in pseudohematoma (adjusted odds ratio, 1.4 per milligram; 95% CI: 1.02 per milligram, 1.9 per milligram; P = 0.03), with a mean area under the receiver operating characteristic curve of 0.96 ± 0.05 (standard error). For RLAS, the predictors were P-GCS (mean coefficient, 0.32 ± 0.06; P < .001) and iodine quantity in contusion (mean coefficient, -0.04 per milligram ± 0.02; P = 0.01). Predictors for DRS were P-GCS (mean coefficient, -1.15 ± 0.27; P < .001), age (mean coefficient, 0.13 per year ± 0.04; P = .002), and iodine quantity in contusion (mean coefficient, 0.19 per milligram ± 0.07; P = .02).ConclusionIodine-based dual-energy CT variables correlate with in-hospital mortality and short-term outcomes for contusions at hospital discharge.© RSNA, 2019Online supplemental material is available for this article.See also the editorial by Talbott and Hess in this issue.
| S-EPMC7705608 | biostudies-literature
Project description:PurposeTo train and validate a predictive model of mortality for hospitalized COVID-19 patients based on lung densitometry.MethodsTwo-hundred-fifty-one patients with respiratory symptoms underwent CT few days after hospitalization. "Aerated" (AV), "consolidated" (CV) and "intermediate" (IV) lung sub-volumes were quantified by an operator-independent method based on individual HU maximum gradient recognition. AV, CV, IV, CV/AV, IV/AV, and HU of the first peak position were extracted. Relevant clinical parameters were prospectively collected. The population was composed by training (n = 166) and validation (n = 85) consecutive cohorts, and backward multi-variate logistic regression was applied on the training group to build a CT_model. Similarly, models including only clinical parameters (CLIN_model) and both CT/clinical parameters (COMB_model) were developed. Model's performances were assessed by goodness-of-fit (H&L-test), calibration and discrimination. Model's performances were tested in the validation group.ResultsForty-three patients died (25/18 in training/validation). CT_model included AVmax (i.e. maximum AV between lungs), CV and CV/AE, while CLIN_model included random glycemia, C-reactive protein and biological drugs (protective). Goodness-of-fit and discrimination were similar (H&L:0.70 vs 0.80; AUC:0.80 vs 0.80). COMB_model including AVmax, CV, CV/AE, random glycemia, biological drugs and active cancer, outperformed both models (H&L:0.91; AUC:0.89, 95%CI:0.82-0.93). All models showed good calibration (R2:0.77-0.97). Despite several patient's characteristics were different between training and validation cohorts, performances in the validation cohort confirmed good calibration (R2:0-70-0.81) and discrimination for CT_model/COMB_model (AUC:0.72/0.76), while CLIN_model performed worse (AUC:0.64).ConclusionsFew automatically extracted densitometry parameters with clear functional meaning predicted mortality of COVID-19 patients. Combined with clinical features, the resulting predictive model showed higher discrimination/calibration.
| S-EPMC8084622 | biostudies-literature
Project description:ObjectivesTo test radiomics-based features extracted from noncontrast CT of patients with spontaneous intracerebral haemorrhage for prediction of haematoma expansion and poor functional outcome and compare them with radiological signs and clinical factors.Materials and methodsSeven hundred fifty-four radiomics-based features were extracted from 1732 scans derived from the TICH-2 multicentre clinical trial. Features were harmonised and a correlation-based feature selection was applied. Different elastic-net parameterisations were tested to assess the predictive performance of the selected radiomics-based features using grid optimisation. For comparison, the same procedure was run using radiological signs and clinical factors separately. Models trained with radiomics-based features combined with radiological signs or clinical factors were tested. Predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC) score.ResultsThe optimal radiomics-based model showed an AUC of 0.693 for haematoma expansion and an AUC of 0.783 for poor functional outcome. Models with radiological signs alone yielded substantial reductions in sensitivity. Combining radiomics-based features and radiological signs did not provide any improvement over radiomics-based features alone. Models with clinical factors had similar performance compared to using radiomics-based features, albeit with low sensitivity for haematoma expansion. Performance of radiomics-based features was boosted by incorporating clinical factors, with time from onset to scan and age being the most important contributors for haematoma expansion and poor functional outcome prediction, respectively.ConclusionRadiomics-based features perform better than radiological signs and similarly to clinical factors on the prediction of haematoma expansion and poor functional outcome. Moreover, combining radiomics-based features with clinical factors improves their performance.Key points• Linear models based on CT radiomics-based features perform better than radiological signs on the prediction of haematoma expansion and poor functional outcome in the context of intracerebral haemorrhage. • Linear models based on CT radiomics-based features perform similarly to clinical factors known to be good predictors. However, combining these clinical factors with radiomics-based features increases their predictive performance.
| S-EPMC8452575 | biostudies-literature
Project description:Prediction in natural environments is a challenging task, and there is a lack of clarity around how a myopic organism can make short-term predictions given limited data availability and cognitive resources. In this context, we may ask what kind of resources are available to the organism to help it address the challenge of short-term prediction within its own cognitive limits. We point to one potentially important resource: ordinal patterns, which are extensively used in physics but not in the study of cognitive processes. We explain the potential importance of ordinal patterns for short-term prediction, and how natural constraints imposed through (i) ordinal pattern types, (ii) their transition probabilities and (iii) their irreversibility signature may support short-term prediction. Having tested these ideas on a massive dataset of Bitcoin prices representing a highly fluctuating environment, we provide preliminary empirical support showing how organisms characterized by bounded rationality may generate short-term predictions by relying on ordinal patterns.
| S-EPMC7890473 | biostudies-literature
Project description:Purpose4D perfusion magnetic resonance imaging (MRI) with intravenous injection of contrast agent allows for a radiation-free assessment of regional lung function. It is therefore a valuable method to monitor response to treatment in patients with chronic obstructive pulmonary disease (COPD). This study was designed to evaluate its potential for monitoring short-term response to hyperoxia in COPD patients.Materials and methods19 prospectively enrolled COPD patients (median age 66y) underwent paired dynamic contrast-enhanced 4D perfusion MRI within 35min, first breathing 100% oxygen (injection 1, O2) and then room air (injection 2, RA), which was repeated on two consecutive days (day 1 and 2). Post-processing software was employed to calculate mean transit time (MTT), pulmonary blood volume (PBV) and pulmonary blood flow (PBF), based on the indicator dilution theory, for the automatically segmented whole lung and 12 regions of equal volume.ResultsComparing O2 with RA conditions, PBF and PBV were found to be significantly lower at O2, consistently on both days (p<10-8). Comparing day 2 to day 1, MTT was shorter by 0.59±0.63 s (p<10-8), PBF was higher by 22±80 ml/min/100ml (p<3·10-4), and PBV tended to be lower by 0.2±7.2 ml/100ml (p = 0.159) at both, RA and O2, conditions.ConclusionThe second injection (RA) yielded higher PBF and PBV, which apparently contradicts the established hypothesis that hyperoxia increases lung perfusion. Quantification of 4D perfusion MRI by current software approaches may thus be limited by residual circulating contrast agent in the short-term and even the next day.
| S-EPMC6287948 | biostudies-literature
Project description:Since intelligent systems were developed to collect traffic data, this data can be collected at high volume, velocity, and variety, resulting in big traffic data. In previous studies, dealing with the large volume of big traffic data has always been discussed. In this study, big traffic data were used to predict traffic state on a section of suburban road from Karaj to Chalous located in the north of Iran. Due to the many and various extracted features, data dimensions management is necessary. This management was accomplished using principal component analysis to reduce the number of features, genetic algorithms to select features influencing traffic states, and cyclic features to change the nature of features. The data set obtained from each method is used as input to the models. The models used include long short-term memory, support vector machine, and random forest. The results show that using cyclic features can increase traffic state prediction's accuracy than the model, including all the initial features (base model). Long short-term memory model with 71 cyclic features offers the highest accuracy, equivalent to 88.09%. Additionally, this model's reduced number of features led to a shorter modelling execution time, from 456 s (base model) to 201 s.
| S-EPMC10794253 | biostudies-literature