Integrating regional perfusion CT information to improve prediction of infarction after stroke.
Ontology highlight
ABSTRACT:
SUBMITTER: Klug J
PROVIDER: S-EPMC7922756 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:IntroductionWe investigated whether baseline CT angiography (CTA) and CT perfusion (CTP) in acute ischemic stroke could improve prediction of infarct presence and infarct volume on follow-up imaging.MethodsWe analyzed 906 patients with suspected anterior circulation stroke from the prospective multicenter Dutch acute stroke study (DUST). All patients underwent baseline non-contrast CT, CTA, and CTP and follow-up non-contrast CT/MRI after 3 days. Multivariable regression models were developed including patient characteristics and non-contrast CT, and subsequently, CTA and CTP measures were added. The increase in area under the curve (AUC) and R (2) was assessed to determine the additional value of CTA and CTP.ResultsAt follow-up, 612 patients (67.5%) had a detectable infarct on CT/MRI; median infarct volume was 14.8 mL (interquartile range (IQR) 2.8-69.6). Regarding infarct presence, the AUC of 0.82 (95% confidence interval (CI) 0.79-0.85) for patient characteristics and non-contrast CT was improved with addition of CTA measures (AUC 0.85 (95% CI 0.82-0.87); p < 0.001) and was even higher after addition of CTP measures (AUC 0.89 (95% CI 0.87-0.91); p < 0.001) and combined CTA/CTP measures (AUC 0.89 (95% CI 0.87-0.91); p < 0.001). For infarct volume, adding combined CTA/CTP measures (R (2) = 0.58) was superior to patient characteristics and non-contrast CT alone (R (2) = 0.44) and to addition of CTA alone (R (2) = 0.55) or CTP alone (R (2) = 0.54; all p < 0.001).ConclusionIn the acute stage, CTA and CTP have additional value over patient characteristics and non-contrast CT for predicting infarct presence and infarct volume on follow-up imaging. These findings could be applied for patient selection in future trials on ischemic stroke treatment.
| S-EPMC4819789 | biostudies-literature
Project description:PurposeThe aim of this study was to evaluate whether the addition of brain CT imaging data to a model incorporating clinical risk factors improves prediction of ischemic stroke recurrence over 5 years of follow-up.MethodsA total of 638 patients with ischemic stroke from three centers were selected from the Dutch acute stroke study (DUST). CT-derived candidate predictors included findings on non-contrast CT, CT perfusion, and CT angiography. Five-year follow-up data were extracted from medical records. We developed a multivariable Cox regression model containing clinical predictors and an extended model including CT-derived predictors by applying backward elimination. We calculated net reclassification improvement and integrated discrimination improvement indices. Discrimination was evaluated with the optimism-corrected c-statistic and calibration with a calibration plot.ResultsDuring 5 years of follow-up, 56 patients (9%) had a recurrence. The c-statistic of the clinical model, which contained male sex, history of hyperlipidemia, and history of stroke or transient ischemic attack, was 0.61. Compared with the clinical model, the extended model, which contained previous cerebral infarcts on non-contrast CT and Alberta Stroke Program Early CT score greater than 7 on mean transit time maps derived from CT perfusion, had higher discriminative performance (c-statistic 0.65, P = 0.01). Inclusion of these CT variables led to a significant improvement in reclassification measures, by using the net reclassification improvement and integrated discrimination improvement indices.ConclusionData from CT imaging significantly improved the discriminatory performance and reclassification in predicting ischemic stroke recurrence beyond a model incorporating clinical risk factors only.
| S-EPMC7966192 | biostudies-literature
Project description:BackgroundIn ischaemic stroke patients undergoing reperfusion therapy, the amount of salvageable tissue, that is, extent of the ischaemic penumbra, predicts the clinical outcomes. CT perfusion (CTP) enables quantification of penumbral tissues to guide decision making, and current programmes have automated its analysis. More advanced machine learning techniques utilising the CTP maps may improve prediction beyond the ischaemic volume measures.MethodWe determined whether applying convolutional neural networks (CNN), a key machine learning technique in modelling image-label relationships, to post-processed CTP maps improved prediction of outcome, assessed by 3 months modified Rankin scale (mRS). Patients who underwent thrombolysis but not thrombectomy were included. CTP maps of a retrospective cohort of 230 patients with middle cerebral artery stroke were used to develop the model, which was validated in an independent cohort of 129 patients.ResultsWe constructed a CNN model that predicted a favourable post-thrombolysis outcome (mRS 0-2 at 3 months) with an area under receiver-operator characteristics curve (AUC) of 0.792 (95% CI, 0.707-0.877). This model outperformed a currently clinically used MISTAR software using previously validated thresholds (AUC = 0.583, 95% CI, 0.480-0.686) and a model modified using thresholds from the derivation cohort (AUC = 0.670, 95% CI, 0.571-0.769). By combining CNN-derived features and baseline demographic features, the prediction AUC was improved to 0.865 (95% CI, 0.794-0.936).ConclusionCNN improved prediction of post-thrombolysis outcome, and may be useful in selecting which patients benefit from thrombolysis.
| S-EPMC10472959 | biostudies-literature
Project description:We aimed to investigate optimal perfusion thresholds defining ischemic core and penumbra for hemispheric-cortical gray matter (GM) and subcortical white matter (WM). A total of 65 sub-6 h ischemic stroke patients were assessed, who underwent acute computed tomography perfusion (CTP) and acute magnetic resonance imaging. CTP maps were generated by both standard singular value deconvolution (sSVD) and SVD with delay and dispersion correction (ddSVD). Analyses were undertaken to calculate sensitivity, specificity, and area under the curve (AUC) for each CTP threshold for core and penumbra in GM and WM. With sSVD, the core was best defined in GM by cerebral blood flow (CBF) < 30% (AUC: 0.73) and in WM by CBF < 20% (AUC: 0.67). With ddSVD, GM core was best defined by CBF < 35% (AUC: 0.75) and in WM by CBF < 25% (AUC: 0.68). A combined GM/WM threshold overestimated core compared to diffusion-weighted imaging, CBF < 25% from sSVD (1.88 ml, P = 0.007) and CBF < 30% from ddSVD (1.27 ml, P = 0.011). The perfusion lesion was best defined by Tmax > 5 s (AUC: 0.80) in GM and Tmax > 7 s (AUC: 0.75) in WM. With sSVD, a delay time (DT) > 3 s from ddSVD was the optimal for both GM (AUC: 0.78) and WM (AUC: 0.75). Using tissue-specific thresholds for GM/WM provides more accurate estimation of acute ischemic core.
| S-EPMC6421247 | biostudies-literature
Project description:In the context of recent guidelines for stroke management, application of computed tomographic perfusion (CTP) is anticipated to increase, especially within 6-24 hours after stroke onset. There are commercially available software packages to help to detect the infarct core and the penumbra, however, the results of these software are not always straightforward that might cause incorrect CTP interpretation. We present here a case of acute ischemic stroke in which the CTP software misinterpreted the results by labeling the normal hemisphere as infarction and ischemia due to luxury perfusion of the infarcted hemisphere. Awareness of laterality of the patient's symptoms and the understanding of functioning of perfusion analysis software is necessary to avoid this pitfall.
| S-EPMC6168926 | biostudies-literature
Project description:IntroductionMalignant cerebral edema develops in a small subset of patients with hemispheric strokes, precipitating deterioration and death if decompressive hemicraniectomy (DHC) is not performed in a timely manner. Predicting which stroke patients will develop malignant edema is imprecise based on clinical data alone. Head computed tomography (CT) imaging is often performed at baseline and 24-h. We determined the incremental value of incorporating imaging-derived features from serial CTs to enhance prediction of malignant edema.MethodsWe identified hemispheric stroke patients at three sites with NIHSS ≥ 7 who had baseline as well as 24-h clinical and CT imaging data. We extracted quantitative imaging features from baseline and follow-up CTs, including CSF volume, intracranial reserve (CSF/cranial volume), as well as midline shift (MLS) and infarct-related hypodensity volume. Potentially lethal malignant edema was defined as requiring DHC or dying with MLS over 5-mm. We built machine-learning models using logistic regression first with baseline data and then adding 24-h data including reduction in CSF volume (ΔCSF). Model performance was evaluated with cross-validation using metrics of recall (sensitivity), precision (predictive value), as well as area under receiver-operating-characteristic and precision-recall curves (AUROC, AUPRC).ResultsTwenty of 361 patients (6%) died or underwent DHC. Baseline clinical variables alone had recall of 60% with low precision (7%), AUROC 0.59, AUPRC 0.15. Adding baseline intracranial reserve improved recall to 80% and AUROC to 0.82 but precision remained only 16% (AUPRC 0.28). Incorporating ΔCSF improved AUPRC to 0.53 (AUROC 0.91) while all imaging features further improved prediction (recall 90%, precision 38%, AUROC 0.96, AUPRC 0.66).ConclusionIncorporating quantitative CT-based imaging features from baseline and 24-h CT enhances identification of patients with malignant edema needing DHC. Further refinements and external validation of such imaging-based machine-learning models are required.
| S-EPMC7738418 | biostudies-literature
Project description:Imaging protocols for acute ischemic stroke varies significantly from center to center leading to challenges in research translation. We aimed to assess the inter-rater reliability of collateral grading systems derived from dynamic computed tomography angiography (CTA) and an optimized multiphase CTA and, to analyze the association of the two CTA modalities with CT perfusion (CTP) compartments by comparing the accuracy of dynamic CTA (dCTA) and optimized multiphase CTA (omCTA) in identifying CT perfusion (CTP) target mismatch patients. Acute ischemic stroke patients with a proximal large vessel occlusion who underwent whole brain CTP were included in the study. Collateral status were assessed using ASPECTS collaterals (Alberta Stroke Program Early CT Score on Collaterals) and ASITN/SIR collateral system (the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology) on dCTA and omCTA. Eighty-one patients were assessed, with a median ischemic core volume of 29 mL. The collateral assessment with ASPECTS collaterals using dCTA have a similar inter-rater agreement (K-alpha: 0.71) compared to omCTA (K-alpha: 0.69). However, the agreement between dCTA and CTP in classifying patients with target mismatch was higher compared to omCTA (Kappa, dCTA: 0.81; omCTA: 0.64). We found dCTA was more accurate than omCTA in identifying target mismatch patients with proximal large vessel occlusion.
| S-EPMC6819495 | biostudies-literature
Project description:The prediction of infarction volume after stroke onset depends on the shape of the growth dynamics of the infarction. To understand growth patterns that predict lesion volume changes, we studied currently available models described in literature and compared the models with Adaptive Neuro-Fuzzy Inference System [ANFIS], a method previously unused in the prediction of infarction growth and infarction volume (IV). We included 67 patients with malignant middle cerebral artery [MMCA] stroke who underwent decompressive hemicraniectomy. All patients had at least three cranial CT scans prior to the surgery. The rate of growth and volume of infarction measured on the third CT was predicted with ANFIS without statistically significant difference compared to the ground truth [P = 0.489]. This was not possible with linear, logarithmic or exponential methods. ANFIS was able to predict infarction volume [IV3] over a wide range of volume [163.7-600 cm3] and time [22-110 hours]. The cross correlation [CRR] indicated similarity between the ANFIS-predicted IV3 and original data of 82% for ANFIS, followed by logarithmic 70%, exponential 63% and linear 48% respectively. Our study shows that ANFIS is superior to previously defined methods in the prediction of infarction growth rate (IGR) with reasonable accuracy, over wide time and volume range.
| S-EPMC5548812 | biostudies-other
Project description:Antibiotic resistance in bacteria limits the effect of corresponding antibiotics, and the classification of antibiotic resistance genes (ARGs) is important for the treatment of bacterial infections and for understanding the dynamics of microbial communities. Although several methods have been developed to classify ARGs, none of them work well when the ARGs diverge from those in the reference ARG databases. We develop a novel method, ARG-SHINE, for ARG classification. ARG-SHINE utilizes state-of-the-art learning to rank machine learning approach to ensemble three component methods with different features, including sequence homology, protein domain/family/motif and raw amino acid sequences for the deep convolutional neural network. Compared with other methods, ARG-SHINE achieves better performance on two benchmark datasets in terms of accuracy, macro-average f1-score and weighted-average f1-score. ARG-SHINE is used to classify newly discovered ARGs through functional screening and achieves high prediction accuracy. ARG-SHINE is freely available at https://github.com/ziyewang/ARG_SHINE.
| S-EPMC8341004 | biostudies-literature
Project description:Exposure assessment traditionally relies on biomarkers that measure chemical concentrations in individual biological media (i.e., blood, urine, etc.). However, chemicals distribute unevenly among different biological media; thus, each medium provides incomplete information about body burden. We propose that machine learning and statistical approaches can create integrated exposure estimates from multiple biomarker matrices that better represent the overall body burden, which we term multi-media biomarkers (MMBs). We measured lead (Pb) in blood, urine, hair and nails from 251 Italian adolescents aged 11-14 years from the Public Health Impact of Metals Exposure (PHIME) cohort. We derived aggregated MMBs from the four biomarkers and then tested their association with Wechsler Intelligence Scale for Children (WISC) IQ scores. We used three approaches to derive the Pb MMB: one supervised learning technique, weighted quantile sum regression (WQS), and two unsupervised learning techniques, independent component analysis (ICA) and non-negative matrix factorization (NMF). Overall, the Pb MMB derived using WQS was most consistently associated with IQ scores and was the only method to be statistically significant for Verbal IQ, Performance IQ and Total IQ. A one standard deviation increase in the WQS MMB was associated with lower Verbal IQ (β [95% CI] = -2.2 points [-3.7, -0.6]), Performance IQ (-1.9 points [-3.5, -0.4]) and Total IQ (-2.1 points [-3.8, -0.5]). Blood Pb was negatively associated with only Verbal IQ, with a one standard deviation increase in blood Pb being associated with a -1.7 point (95% CI: [-3.3, -0.1]) decrease in Verbal IQ. Increases of one standard deviation in the ICA MMB were associated with lower Verbal IQ (-1.7 points [-3.3, -0.1]) and lower Total IQ (-1.7 points [-3.3, -0.1]). Similarly, an increase of one standard deviation in the NMF MMB was associated with lower Verbal IQ (-1.8 points [-3.4, -0.2]) and lower Total IQ (-1.8 points [-3.4, -0.2]). Weights highlighting the contributions of each medium to the MMB revealed that blood Pb was the largest contributor to most MMBs, although the weights varied from more than 80% for the ICA and NMF MMBs to between 30% and 54% for the WQS-derived MMBs. Our results suggest that MMBs better reflect the total body burden of a chemical that may be acting on target organs than individual biomarkers. Estimating MMBs improved our ability to estimate the full impact of Pb on IQ. Compared with individual Pb biomarkers, including blood, a Pb MMB derived using WQS was more strongly associated with IQ scores. MMBs may increase statistical power when the choice of exposure medium is unclear or when the sample size is small. Future work will need to validate these methods in other cohorts and for other chemicals.
| S-EPMC7167344 | biostudies-literature