Dataset Information

Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events.

ABSTRACT: Radiomics, quantitative feature extraction from radiological images, can improve disease diagnosis and prognostication. However, radiomic features are susceptible to image acquisition and segmentation variability. Ideally, only features robust to these variations would be incorporated into predictive models, for good generalisability. We extracted 93 radiomic features from carotid artery computed tomography angiograms of 41 patients with cerebrovascular events. We tested feature robustness to region-of-interest perturbations, image pre-processing settings and quantisation methods using both single- and multi-slice approaches. We assessed the ability of the most robust features to identify culprit and non-culprit arteries using several machine learning algorithms and report the average area under the curve (AUC) from five-fold cross validation. Multi-slice features were superior to single for producing robust radiomic features (67 vs. 61). The optimal image quantisation method used bin widths of 25 or 30. Incorporating our top 10 non-redundant robust radiomics features into ElasticNet achieved an AUC of 0.73 and accuracy of 69% (compared to carotid calcification alone [AUC: 0.44, accuracy: 46%]). Our results provide key information for introducing carotid CT radiomics into clinical practice. If validated prospectively, our robust carotid radiomic set could improve stroke prediction and target therapies to those at highest risk.

SUBMITTER: Le EPV

PROVIDER: S-EPMC7876096 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events.

Le Elizabeth P V EPV Rundo Leonardo L Tarkin Jason M JM Evans Nicholas R NR Chowdhury Mohammed M MM Coughlin Patrick A PA Pavey Holly H Wall Chris C Zaccagna Fulvio F Gallagher Ferdia A FA Huang Yuan Y Sriranjan Rouchelle R Le Anthony A Weir-McCall Jonathan R JR Roberts Michael M Gilbert Fiona J FJ Warburton Elizabeth A EA Schönlieb Carola-Bibiane CB Sala Evis E Rudd James H F JHF

Scientific reports 20210210 1

Radiomics, quantitative feature extraction from radiological images, can improve disease diagnosis and prognostication. However, radiomic features are susceptible to image acquisition and segmentation variability. Ideally, only features robust to these variations would be incorporated into predictive models, for good generalisability. We extracted 93 radiomic features from carotid artery computed tomography angiograms of 41 patients with cerebrovascular events. We tested feature robustness to re ...[more]

PMID: 33568735

Similar Datasets

Project description:One pending issue in cardiovascular epigenetics is whether cerebrovascular events, a major complication of atherosclerosis, are associated with any specific DNA methylation changes in the carotid plaque. To clarify that topic, we profiled the DNA methylomes of human symptomatic carotid plaques (SCPs) obtained from patients who suffered cerebrovascular events (n=19) and asymptomatic counterparts (ACP; n=19) with a high-density microarray (~485,000 CpG sites, Illumina), and crossed DNA methylation data with RNAseq-based expression data from an independent SCP set (n=8). Few (30) CpGs showed a significant (p<0.05; absolute Delta-Beta>0.20) differential methylation between the two groups. Within SCPs, methylation correlated significantly with post-cerebrovascular event time (PCET; range: 3-45 days; r-value range -0.926 to 0.857; p<0.05) for ~45,000 CpGs, the vast majority of which became hypomethylated with increasing PCET. Hypomethylation was not due to erasure of the gene-body and CG-poor region hypermethylation that accompany the progression of stable lesions, but rather targeted promoters and CpG islands. Noticeably, promoter hypomethylation and increased expression were observed for genes involved in the inhibition of the inflammatory response, defence against oxidative stress and active DNA demethylation. Our data show that only weak changes in the DNA methylome distinguish symptomatic from asymptomatic carotid plaques, but a widespread demethylation resulting in permissive transcriptional marks at atheroprotective gene promoters is established in plaques after a cerebrovascular event, thus mirroring previous observations that ruptured plaques tend to revert to a more stable structure. The identified loci are candidate targets to accelerate the pace of carotid plaque stabilization. DNA was quantified by Quant-iT PicoGreen dsDNA Reagent (Invitrogen) and the integrity was analyzed in a 1.3% agarose gel. Bisulfite conversion of 600 ng of each sample was perform according to the manufacturer's recommendation for Illumina Infinium Assay. Effective bisulphite conversion was checked for three controls that were converted simultaneously with the samples. 4 ul of bisulfite converted DNA were used to hybridize on Infinium HumanMethylation 450 BeadChip, following Illumina Infinium HD Methylation protocol. Chip analysis was performed using Illumina HiScan SQ fluorescent scanner. The intensities of the images are extracted using GenomeStudio (2011.2) Methylation module (1.8.5) software. Methylation score of each CpG is represented as beta value.

Project description:ObjectivesTo evaluate inter- and intra- scan mode and scanner repeatability and reproducibility of radiomics features within and between single-energy CT (SECT) and dual-energy CT (DECT).MethodsA standardized phantom with sixteen rods of clinical-relevant densities was scanned on seven DECT-capable scanners and three SECT-only scanners. The acquisition parameters were selected to present typical abdomen-pelvic examinations with the same voxel size. Images of SECT at 120 kVp and corresponding 120 kVp-like virtual monochromatic images (VMIs) in DECT which were generated according to scanners were analyzed. Regions of interest were drawn with rigid registrations to avoid variations due to segmentation. Radiomics features were extracted via Pyradiomics platform. Test-retest repeatability was evaluated by Bland-Altman analysis for repeated scans. Intra-scanner reproducibility for different scan modes was tested by intraclass correlation coefficient (ICC) and concordance correlation coefficient (CCC). Inter-scanner reproducibility among different scanners for same scan mode was assessed by coefficient of variation (CV) and quartile coefficient of dispersion (QCD).ResultsThe test-retest analysis presented that 92.91% and 87.02% of the 94 assessed features were repeatable for SECT 120kVp and DECT 120 kVp-like VMIs, respectively. The intra-scanner analysis for SECT 120kVp vs DECT 120 kVp-like VMIs demonstrated that 10.76% and 10.28% of features were with ICC > 0.90 and CCC > 0.90, respectively. The inter-scanner analysis showed that 17.09% and 27.73% of features for SECT 120kVp were with CV < 10% and QCD < 10%, and 15.16% and 32.78% for DECT 120 kVp-like VMIs, respectively.ConclusionsThe majority of radiomics features were non-reproducible within and between SECT and DECT.Key points• Although the test-retest analysis showed high repeatability for radiomics features, the overall reproducibility of radiomics features within and between SECT and DECT was low. • Only about one-tenth of radiomics features extracted from SECT images and corresponding DECT images did match each other, even their average photon energy levels were considered alike, indicating that the scan mode potentially altered the radiomics features. • Less than one-fifth of radiomics features were reproducible among multiple SECT and DECT scanners, regardless of their fixed acquisition and reconstruction parameters, suggesting the necessity of scanning protocol adjustment and post-scan harmonization process.

Project description:BackgroundBrain lesions on diffusion-weighted imaging (DWI) are frequently found after carotid artery stenting (CAS), but their clinical relevance remains unclear.ObjectivesThis study sought to investigate whether periprocedural ischemic DWI lesions after CAS or carotid endarterectomy (CEA) are associated with an increased risk of recurrent cerebrovascular events.MethodsIn the magnetic resonance imaging (MRI) substudy of ICSS (International Carotid Stenting Study), 231 patients with symptomatic carotid stenosis were randomized to undergo CAS (n=124) or CEA (n=107). MRIs were performed 1 to 7 days before and 1 to 3 days after treatment. The primary outcome event was stroke or transient ischemic attack in any territory occurring between the post-treatment MRI and the end of follow-up. Time to occurrence of the primary outcome event was compared between patients with (DWI+) and without (DWI-) new DWI lesions on the post-treatment scan in the CAS and CEA groups separately.ResultsMedian time of follow-up was 4.1 years (interquartile range: 3.0 to 5.2). In the CAS group, recurrent stroke or transient ischemic attack occurred more often among DWI+ patients (12 of 62) than among DWI- patients (6 of 62), with a cumulative 5-year incidence of 22.8% (standard error [SE]: 7.1%) and 8.8% (SE: 3.8%), respectively (unadjusted hazard ratio: 2.85; 95% confidence interval: 1.05 to 7.72; p=0.04). In DWI+ and DWI- patients, 8 and 2 events, respectively, occurred within 6 months after treatment. In the CEA group, there was no difference in recurrent cerebrovascular events between DWI+ and DWI- patients.ConclusionsIschemic brain lesions discovered on DWI after CAS seem to be a marker of increased risk for recurrent cerebrovascular events. Patients with periprocedural DWI lesions might benefit from more aggressive and prolonged antiplatelet therapy after CAS. (A Randomised Comparison of the Risks, Benefits and Cost Effectiveness of Primary Carotid Stenting With Carotid Endarterectomy: International Carotid Stenting Study; ISRCTN25337470).

Project description:Vulnerable carotid atherosclerotic plaque (CAP) significantly contributes to ischemic stroke. Neovascularization within plaques is an emerging biomarker linked to plaque vulnerability that can be detected using contrast-enhanced ultrasound (CEUS). Computed tomography angiography (CTA) is a common method used in clinical cerebrovascular assessments that can be employed to evaluate the vulnerability of CAPs. Radiomics is a technique that automatically extracts radiomic features from images. This study aimed to identify radiomic features associated with the neovascularization of CAP and construct a prediction model for CAP vulnerability based on radiomic features. CTA data and clinical data of patients with CAPs who underwent CTA and CEUS between January 2018 and December 2021 in Beijing Hospital were retrospectively collected. The data were divided into a training cohort and a testing cohort using a 7:3 split. According to the examination of CEUS, CAPs were dichotomized into vulnerable and stable groups. 3D Slicer software was used to delineate the region of interest in CTA images, and the Pyradiomics package was used to extract radiomic features in Python. Machine learning algorithms containing logistic regression (LR), support vector machine (SVM), random forest (RF), light gradient boosting machine (LGBM), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), and multi-layer perception (MLP) were used to construct the models. The confusion matrix, receiver operating characteristic (ROC) curve, accuracy, precision, recall, and f-1 score were used to evaluate the performance of the models. A total of 74 patients with 110 CAPs were included. In all, 1,316 radiomic features were extracted, and 10 radiomic features were selected for machine-learning model construction. After evaluating several models on the testing cohorts, it was discovered that model_RF outperformed the others, achieving an AUC value of 0.93 (95% CI: 0.88-0.99). The accuracy, precision, recall, and f-1 score of model_RF in the testing cohort were 0.85, 0.87, 0.85, and 0.85, respectively. Radiomic features associated with the neovascularization of CAP were obtained. Our study highlights the potential of radiomics-based models for improving the accuracy and efficiency of diagnosing vulnerable CAP. In particular, the model_RF, utilizing radiomic features extracted from CTA, provides a noninvasive and efficient method for accurately predicting the vulnerability status of CAP. This model shows great potential for offering clinical guidance for early detection and improving patient outcomes.

Project description:Background Studies have shown that pericoronary artery inflammation can be accurately detected via increased attenuation on computed tomography. Our purpose was to evaluate the association between pericarotid inflammation, measured by density of carotid perivascular fat on computed tomography angiography, with stroke and transient ischemic attack. Methods and Results We screened computed tomography angiography examinations for patients with unilateral internal carotid artery ( ICA ) stenosis ≥50% to 99%. A blinded neuroradiologist placed regions-of-interest in the pericarotid fat on the slice showing maximal stenosis. Two-sample t tests were performed to assess between-subject differences in mean Hounsfield Units in carotid perivascular fat between symptomatic and asymptomatic patients. Paired t tests were used to assess within-subject differences in mean Hounsfield Units between stenotic versus nonstenotic ICA s in a given patient. We included 94 patients, including 42 symptomatic and 52 asymptomatic patients. In the between-subject analysis of stenotic ICA s, we found symptomatic patients had higher mean pericarotid fat density compared with asymptomatic patients (-66.2±19.2 versus -77.1±20.4, P=0.009). When comparing nonstenotic ICA s, there was no significant difference between pericarotid fat density in symptomatic compared with asymptomatic patients (-81.0±13.3 versus -85.3±18.0: P=0.198). Within-subject comparison showed statistically significant increased density in stenotic ICA versus nonstenotic ICA with mean Hounsfield Units difference of 11.1 ( P<0.0001). Conclusions We found increased density, a surrogate marker for perivascular inflammation, in the fat surrounding ICA s ipsilateral to stroke or transient ischemic attack compared with asymptomatic ICA s. Our findings suggest that inflammation associated with culprit carotid plaques extends beyond the vessel lumen and can be identified using simple methods on computed tomography angiography imaging.

Dataset Information

Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events.

Publications

Assessing robustness of carotid artery CT angiography radiomics in the identification of culprit lesions in cerebrovascular events.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets