Exploration of Multiparameter Hematoma 3D Image Analysis for Predicting Outcome After Intracerebral Hemorrhage

Salazar, Pascal; Napoli, Mario Di; Jafari, Mostafa; Jafarli, Alibay; Ziai, Wendy; Petersen, Alexander; Mayer, Stephan A.; Bershad, Eric M.; Damani, Rahul; Divani, Afshin A.

doi:10.1007/s12028-019-00783-8

Cited by 18 publications

(18 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…F2 seems to re ect the transition from normal lung densities to extended ground glass opaci cations (about-800HU to -600HU) and it was not a signi cant predictor of critically ill status (AUC: 0.53, P:0.60). In this study, the overall results using FPCA are consistent with previous ones for pulmonary disease subtyping 18,19 or patient neurologic outcome prediction 20 con rming the value of the FPCA method: rst as a non-speci c data driven exploration tool, it offers interpretable modes of variations of the CT histograms in the patient whole cohort. Second, it is a generic method giving accurate predictors related to histogram variations without a priori knowledge or delicate radiomic high dimensional parameter selection.…”

Section: Discussionsupporting

confidence: 86%

Predefined and data driven CT densitometric features predict critical illness and hospital length of stay in COVID-19 patients

Shalmon

Salazar

Horie

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The aim of this study was to compare predefined and data-driven parameters of whole lung CT density histograms to predict critical illness outcome and hospital length of stay in a cohort of 80 COVID-19 patients. CT chest images on segmented lungs were retrospectively analyzed. Functional Principal Component Analysis (FPCA) was used to find the main modes of variations on CT density histograms (F1,F2,F3,F4) in the whole patient cohort. The data driven and a priori CT density features, the CT severity score, the COVID-GRAM score and the patient clinical data were assessed for predicting the patient outcome using logistic regression models stratified for contrast enhanced CT and non-enhanced CT, and survival analysis. ROC analysis identified as best predictors of critically ill status: 87.5th percentile CT density (Q875) - AUC: 0.88 95%CI (0.79 0.94), F1-CT - AUC: 0.87 (0.77 0.93) Standard Deviation (SD-CT)- AUC: 0.86 (0.73, 0.93). Multivariate models combining CT-density predictors and Neutrophil-Lymphocyte Ratio showed the highest accuracy with cross-validated AUCs in the 0.91–0.92 range for contrast CT and 0.82–0.88 range for non-contrast CT. SD-CT, Q875 and F1 score were significant predictors of hospital length of stay while controlling for hospital death using competing risks models. Predefined and data-driven parameters of lung CT density histograms can predict critical illness and length of stay to guide management and resources. FPCA method can be used to interpret the CT density histogram variation in a patient cohort and to extract predictive features with minimal a priori knowledge.

show abstract

Section: Discussionsupporting

confidence: 86%

Predefined and data driven CT densitometric features predict critical illness and hospital length of stay in COVID-19 patients

Shalmon

Salazar

Horie

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Selection of features was based on their individual ROC-AUC performance in the full dataset for nodule classification into both 1) H1 or H2 vs H3 and 2) H1 vs H2 or H3. Then, a confounder plot was computed using the R-library “Clumix” 45 , as previously described 46 . Built on a distinct approach that relies on variable similarity (see Supplementary Information Online), the confounder plot shows each feature based on its correlation with the response (H1, H2, H3 nodule class) and on its correlation with a reference feature.…”

Section: Methodsmentioning

confidence: 99%

Assessing invasiveness of subsolid lung adenocarcinomas with combined attenuation and geometric feature models

Margerie-Mellon

Gill

Salazar³

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

The aim of this study was to develop and test multiclass predictive models for assessing the invasiveness of individual lung adenocarcinomas presenting as subsolid nodules on computed tomography (CT). 227 lung adenocarcinomas were included: 31 atypical adenomatous hyperplasia and adenocarcinomas in situ (class H1), 64 minimally invasive adenocarcinomas (class H2) and 132 invasive adenocarcinomas (class H3). Nodules were segmented, and geometric and CT attenuation features including functional principal component analysis features (FPC1 and FPC2) were extracted. After a feature selection step, two predictive models were built with ordinal regression: Model 1 based on volume (log) (logarithm of the nodule volume) and FPC1, and Model 2 based on volume (log) and Q.875 (CT attenuation value at the 87.5% percentile). Using the 200-repeats Monte-Carlo cross-validation method, these models provided a multiclass classification of invasiveness with discriminative power AUCs of 0.83 to 0.87 and predicted the class probabilities with less than a 10% average error. The predictive modelling approach adopted in this paper provides a detailed insight on how the value of the main predictors contribute to the probability of nodule invasiveness and underlines the role of nodule CT attenuation features in the nodule invasiveness classification.

show abstract

“…Additionally, we used the ABC/2 method to measure hematoma volume. While some studies support its use, there is some evidence suggesting this method is not as accurate as volumetric or planimetric measurement methods [25,26]. The FDA revoked emergency use of Hydroxychloroquine, after our study period.…”

Section: Strengths and Limitationsmentioning

confidence: 85%

Anticoagulation use and Hemorrhagic Stroke in SARS-CoV-2 Patients Treated at a New York Healthcare System

et al. 2020

View full text Add to dashboard Cite

Background and Purpose: While the thrombotic complications of COVID-19 have been well described, there are limited data on clinically significant bleeding complications including hemorrhagic stroke. The clinical characteristics, underlying stroke mechanism, and outcomes in this particular subset of patients are especially salient as therapeutic anticoagulation becomes increasingly common in the treatment and prevention of thrombotic complications of COVID-19. Methods: We conducted a retrospective cohort study of patients with hemorrhagic stroke (both non-traumatic intracerebral hemorrhage and spontaneous non-aneurysmal subarachnoid hemorrhage) who were hospitalized between March 1, 2020, and May 15, 2020, within a major healthcare system in New York, during the coronavirus pandemic. Patients with hemorrhagic stroke on admission and who developed hemorrhage during hospitalization were both included. We compared the clinical characteristics of patients with hemorrhagic stroke and COVID-19 to those without COVID-19 admitted to our hospital system between March 1, 2020, and May 15, 2020 (contemporary controls), and March 1, 2019, and May 15, 2019 (historical controls). Demographic variables and clinical characteristics between the individual groups were compared using Fischer's exact test for categorical variables and nonparametric test for continuous variables. We adjusted for multiple comparisons using the Bonferroni method. Results: During the study period in 2020, out of 4071 patients who were hospitalized with COVID-19, we identified 19 (0.5%) with hemorrhagic stroke. Of all COVID-19 with hemorrhagic stroke, only three had isolated non-aneurysmal SAH with no associated intraparenchymal hemorrhage. Among hemorrhagic stroke in patients with COVID-19, coagulopathy was the most common etiology (73.7%); empiric anticoagulation was started in 89.5% of these patients versus 4.2% in contemporary controls (p ≤ .001) and 10.0% in historical controls (p ≤ .001). Compared to contemporary and historical controls, patients with COVID-19 had higher initial NIHSS scores, INR, PTT, and fibrinogen levels. Patients with COVID-19 also had higher rates of in-hospital mortality (84.6% vs. 4.6%, p ≤ 0.001). Sensitivity analyses excluding patients with strictly subarachnoid hemorrhage yielded similar results.

show abstract

Exploration of Multiparameter Hematoma 3D Image Analysis for Predicting Outcome After Intracerebral Hemorrhage

Cited by 18 publications

References 41 publications

Predefined and data driven CT densitometric features predict critical illness and hospital length of stay in COVID-19 patients

Predefined and data driven CT densitometric features predict critical illness and hospital length of stay in COVID-19 patients

Assessing invasiveness of subsolid lung adenocarcinomas with combined attenuation and geometric feature models

Anticoagulation use and Hemorrhagic Stroke in SARS-CoV-2 Patients Treated at a New York Healthcare System

Contact Info

Product

Resources

About