Key Points
• In patients with acute ischemic stroke, automated measurements of net water uptake in baseline CTA an NCCT scans can be used as predictor for developing malignant cerebral edema.
• A ...creative approach can lead to broader usability of automated tools.
A pulmonary artery to aorta ratio (PA:A) >1 is a proxy of pulmonary hypertension. It is not known whether this measure carries prognostic information in the general population and in individuals with ...chronic obstructive pulmonary disease (COPD).Between 2003 and 2006, 2197 participants from the population-based Rotterdam Study (mean±sd age 69.7±6.7 years; 51.3% female), underwent cardiac computed tomography (CT) scanning with PA:A quantification, defined as the ratio between the diameters of the pulmonary artery and the aorta. COPD was diagnosed based on spirometry or clinical presentation and obstructive lung function measured by a treating physician. Cox regression was used to investigate the risk of mortality.We observed no association between 1-sd increase of PA:A and mortality in the general population. Larger PA:A was associated with an increased risk of mortality in individuals with COPD, particularly in moderate-to-severe COPD (hazard ratio 1.36, 95% CI 1.03-1.79). We demonstrated that the risk of mortality in COPD was driven by severe COPD, and that this risk increased with decreasing diffusing capacity.Larger PA:A is not associated with mortality in an older general population, but is an independent determinant of mortality in moderate-to-severe COPD. Measuring PA:A in CT scans obtained for other indications may yield important prognostic information in individuals with COPD.
The usefulness of prehospital scales for identifying anterior circulation large vessel occlusion (aLVO) in patients with suspected stroke may vary depending on the severity of their presentation. The ...performance of these scales across the spectrum of deficit severity is unclear. The aim of this study was to evaluate the diagnostic performance of 8 prehospital scales for identifying aLVO across the spectrum of deficit severity.
We used data from the PRESTO study (Prehospital Triage of Patients With Suspected Stroke Symptoms), a prospective observational study comparing prehospital stroke scales in detecting aLVO in suspected stroke patients. We used the National Institutes of Health Stroke Scale (NIHSS) score, assessed in-hospital, as a proxy for the Clinical Global Impression of stroke severity during prehospital assessment by paramedics. We calculated the sensitivity, specificity, positive predictive value, negative predictive value, and the difference in aLVO probabilities with a positive or negative prehospital scale test (ΔP
) for each scale for mild (NIHSS 0-4), intermediate (NIHSS 5-9), moderate (NIHSS 10-14), and severe deficits (NIHSS≥15).
Among 1033 patients with suspected stroke, 119 (11.5%) had an aLVO, of whom 19 (16.0%) had mild, 25 (21.0%) had intermediate, 30 (25.2%) had moderate, and 45 (37.8%) had severe deficits. The scales had low sensitivity and positive predictive value in patients with mild-intermediate deficits, and poor specificity, negative predictive value, and accuracy with moderate-severe deficits. Positive results achieved the highest ΔP
in patients with mild deficits. Negative results achieved the highest ΔP
with severe deficits, but the probability of aLVO with a negative result in the severe range was higher than with a positive test in the mild range.
Commonly-used prehospital stroke scales show variable performance across the range of deficit severity. Probability of aLVO remains high with a negative test in severely affected patients. Studies reporting prehospital stroke scale performance should be appraised in the context of the NIHSS distribution of their samples.
Purpose
In ASPECTS, 10 brain regions are scored visually for presence of acute ischemic stroke damage. We evaluated automated ASPECTS in comparison to expert readers.
Methods
Consecutive, baseline ...non-contrast CT-scans (5-mm slice thickness) from the prospective MR CLEAN trial (
n
= 459, MR CLEAN Netherlands Trial Registry number: NTR1804) were evaluated. A two-observer consensus for ASPECTS regions (normal/abnormal) was used as reference standard for training and testing (0.2/0.8 division). Two other observers provided individual ASPECTS-region scores. The Automated ASPECTS software was applied. A region score specificity of ≥ 90% was used to determine the software threshold for detection of an affected region based on relative density difference between affected and contralateral region. Sensitivity, specificity, and receiver-operating characteristic curves were calculated. Additionally, we assessed intraclass correlation coefficients (ICCs) for automated ASPECTS and observers in comparison to the reference standard in the test set.
Results
In the training set (
n
= 104), with software thresholds for a specificity of ≥ 90%, we found a sensitivity of 33–49% and an area under the curve (AUC) of 0.741–0.785 for detection of an affected ASPECTS region. In the test set (
n
= 355), the results for the found software thresholds were 89–89% (specificity), 41–57% (sensitivity), and 0.750–0.795 (AUC). Comparison of automated ASPECTS with the reference standard resulted in an ICC of 0.526. Comparison of observers with the reference standard resulted in an ICC of 0.383–0.464.
Conclusion
The performance of automated ASPECTS is comparable to expert readers and could support readers in the detection of early ischemic changes.
Increasing evidence suggests involvement of the amount of liver fat in the development of cardiovascular disease. We investigated the relation of liver fat with cardiovascular risk factors and ...subclinical vascular disease in the general population.
Between 2003 and 2006, 2351 persons from the population-based Rotterdam Study (mean age 69.6 ± 6.7 years, 47.2% males) underwent non-enhanced computed tomography. We measured the mean liver attenuation value in Hounsfield units and quantified the following markers of subclinical vascular disease: epicardial fat volume and volumes of coronary (CAC), aortic (AAC), extracranial (ECAC), and intracranial carotid calcification (ICAC). Using linear regression, we investigated associations between traditional cardiovascular risk factors and mean liver attenuation. We also investigated relations of mean liver attenuation with markers of subclinical vascular disease, adjusting for cardiovascular risk factors. We found strong associations of waist circumference, diastolic blood pressure, and diabetes with lower mean liver attenuation multivariable-adjusted beta per unit increase in waist circumference: -2.54 (95% CI: -3.10; -1.99); diastolic blood pressure: -0.52 (95% CI: -0.88; -0.17); and the presence of diabetes: -21.91 (95% CI: -31.76; -12.06). Moreover, we found that larger mean liver attenuation values were associated with smaller volumes of epicardial fat and CAC, independent of cardiovascular risk factors beta per 1-SD increase in mean liver attenuation value: -0.05 (95% CI: -0.08; -0.02) and -0.05 (95% CI: -0.10; -0.01), respectively.
Larger amounts of liver fat are related to larger volumes of epicardial fat and CAC, independent of traditional cardiovascular risk factors, providing important novel insights into the role of liver fat as a marker of vascular disease.
Imaging biomarkers, such as the collateral score as determined from Computed Tomography Angiography (CTA) images, play a role in treatment decision making for acute stroke patients. In this ...manuscript, we present an end-to-end learning approach for automatic determination of a collateral score from a CTA image. Our aim was to investigate whether such end-to-end learning approaches can be used for this classification task, and whether the resulting classification can be used in existing outcome prediction models.
The method consists of a preprocessing step, where the CTA image is aligned to an atlas and divided in the two hemispheres: the affected side and the healthy side. Subsequently, a VoxResNet based convolutional neural network is used to extract features at various resolutions from the input images. This is done by using a Siamese model, such that the classification is driven by the comparison between the affected and healthy using a unique set of features for both hemispheres. After masking the resulting features for both sides with the vascular region and global average pooling (per hemisphere) and concatenation of the resulting features, a fully connected layer is used to determine the categorized collateral score.
Several experiments have been performed to optimize the model hyperparameters and training procedure, and to validate the final model performance. The hyperparameter optimization and subsequent model training was done using CTA images from the MR CLEAN Registry, a Dutch multi-center multi-vendor registry of acute stroke patients that underwent endovascular treatment. A separate set of images, from the MR CLEAN Trial, served as an external validation set, where collateral scoring was assessed and compared with both human observers and a recent more traditional model. In addition, the automated collateral scores have been used in an existing functional outcome prediction model that uses both imaging and non-imaging clinical parameters.
The results show that end-to-end learning of collateral scoring in CTA images is feasible, and does perform similar to more traditional methods, and the performance also is within the inter-observer variation. Furthermore, the results demonstrate that the end-to-end classification results also can be used in an existing functional outcome prediction model.
The biological pathways involved in lesion formation after an acute ischemic stroke (AIS) are poorly understood. Despite successful reperfusion treatment, up to two thirds of patients with large ...vessel occlusion remain functionally dependent. Imaging characteristics extracted from DWI and T2-FLAIR follow-up MR sequences could aid in providing a better understanding of the lesion constituents. We built a fully automated pipeline based on a tree ensemble machine learning model to predict poor long-term functional outcome in patients from the MR CLEAN-NO IV trial. Several feature sets were compared, considering only imaging, only clinical, or both types of features. Nested cross-validation with grid search and a feature selection procedure based on SHapley Additive exPlanations (SHAP) was used to train and validate the models. Considering features from both imaging modalities in combination with clinical characteristics led to the best prognostic model (AUC = 0.85, 95%CI 0.81, 0.89). Moreover, SHAP values showed that imaging features from both sequences have a relevant impact on the final classification, with texture heterogeneity being the most predictive imaging biomarker. This study suggests the prognostic value of both DWI and T2-FLAIR follow-up sequences for AIS patients. If combined with clinical characteristics, they could lead to better understanding of lesion pathophysiology and improved long-term functional outcome prediction.
Due to the time-sensitive effect of endovascular treatment, rapid prehospital identification of large-vessel occlusion in individuals with suspected stroke is essential to optimise outcome. ...Interhospital transfers are an important cause of delay of endovascular treatment. Prehospital stroke scales have been proposed to select patients with large-vessel occlusion for direct transport to an endovascular-capable intervention centre. We aimed to prospectively validate eight prehospital stroke scales in the field.
We did a multicentre, prospective, observational cohort study of adults with suspected stroke (aged ≥18 years) who were transported by ambulance to one of eight hospitals in southwest Netherlands. Suspected stroke was defined by a positive Face-Arm-Speech-Time (FAST) test. We included individuals with blood glucose of at least 2·5 mmol/L. People who presented more than 6 h after symptom onset were excluded from the analysis. After structured training, paramedics used a mobile app to assess items from eight prehospital stroke scales: Rapid Arterial oCclusion Evaluation (RACE), Los Angeles Motor Scale (LAMS), Cincinnati Stroke Triage Assessment Tool (C-STAT), Gaze-Face-Arm-Speech-Time (G-FAST), Prehospital Acute Stroke Severity (PASS), Cincinnati Prehospital Stroke Scale (CPSS), Conveniently-Grasped Field Assessment Stroke Triage (CG-FAST), and the FAST-PLUS (Face-Arm-Speech-Time plus severe arm or leg motor deficit) test. The primary outcome was the clinical diagnosis of ischaemic stroke with a proximal intracranial large-vessel occlusion in the anterior circulation (aLVO) on CT angiography. Baseline neuroimaging was centrally assessed by neuroradiologists to validate the true occlusion status. Prehospital stroke scale performance was expressed as the area under the receiver operating characteristic curve (AUC) and was compared with National Institutes of Health Stroke Scale (NIHSS) scores assessed by clinicians at the emergency department. This study was registered at the Netherlands Trial Register, NL7387.
Between Aug 13, 2018, and Sept 2, 2019, 1039 people (median age 72 years IQR 61–81) with suspected stroke were identified by paramedics, of whom 120 (12%) were diagnosed with aLVO. Of all prehospital stroke scales, the AUC for RACE was highest (0·83, 95% CI 0·79–0·86), followed by the AUC for G-FAST (0·80, 0·76–0·84), CG-FAST (0·80, 0·76–0·84), LAMS (0·79, 0·75–0·83), CPSS (0·79, 0·75–0·83), PASS (0·76, 0·72–0·80), C-STAT (0·75, 0·71–0·80), and FAST-PLUS (0·72, 0·67–0·76). The NIHSS as assessed by a clinician in the emergency department did somewhat better than the prehospital stroke scales with an AUC of 0·86 (95% CI 0·83–0·89).
Prehospital stroke scales detect aLVO with acceptable-to-good accuracy. RACE, G-FAST, and CG-FAST are the best performing prehospital stroke scales out of the eight scales tested and approach the performance of the clinician-assessed NIHSS. Further studies are needed to investigate whether use of these scales in regional transportation strategies can optimise outcomes of patients with ischaemic stroke.
BeterKeten Collaboration and Theia Foundation (Zilveren Kruis).
Purpose
Outcome of endovascular treatment in acute ischemic stroke patients is depending on the collateral circulation maintaining blood flow to the ischemic territory. We evaluated the inter-rater ...reliability and accuracy of raters and an automated algorithm for assessing the collateral score (CS, range: 0–3) in acute ischemic stroke patients.
Methods
Baseline CTA scans with an intracranial anterior occlusion from the MR CLEAN study (
n
=500) were used. For each core lab CS, ten CTA scans with sufficient quality were randomly selected. After a training session in collateral scoring, all selected CTA scans were individually evaluated for a visual CS by three groups: 7 radiologists, 13 junior and 9 senior radiology residents. Two additional radiologists scored CS to be used as reference, with a third providing a CS to produce a 2 out of 3 consensus CS in case of disagreement. An automated algorithm was also used to compute CS. Inter-rater agreement was reported with intraclass correlation coefficient (ICC). Accuracy of visual and automated CS were calculated.
Results
39 CTA scans were assessed (1 corrupt CTA-scan excluded). All groups showed a moderate ICC (0.689-0.780) in comparison to the reference standard. Overall human accuracy was 65± 7% and increased to 88± 5% for dichotomized CS (0–1, 2–3). Automated CS accuracy was 62%, and 90% for dichotomized CS. No significant difference in accuracy was found between groups with different levels of expertise.
Conclusion
After training, inter-rater reliability in collateral scoring was not influenced by experience. Automated CS performs similar to residents and radiologists in determining a collateral score.
Extracting the cerebral anterior vessel tree of patients with an intracranial large vessel occlusion (LVO) is relevant to investigate potential biomarkers that can contribute to treatment decision ...making. The purpose of our work is to develop a method that can achieve this from routinely acquired computed tomography angiography (CTA) and computed tomography perfusion (CTP) images.
To this end, we regard the anterior vessel tree as a set of bifurcations and connected centerlines. The method consists of a proximal policy optimization (PPO) based deep reinforcement learning (DRL) approach for tracking centerlines, a convolutional neural network based bifurcation detector, and a breadth-first vessel tree construction approach taking the tracking and bifurcation detection results as input. We experimentally determine the added values of various components of the tracker. Both DRL vessel tracking and CNN bifurcation detection were assessed in a cross validation experiment using 115 subjects. The anterior vessel tree formation was evaluated on an independent test set of 25 subjects, and compared to interobserver variation on a small subset of images.
The DRL tracking result achieves a median overlapping rate until the first error (1.8 mm off the reference standard) of 100, 46, 100 % on 8032 vessels over 115 subjects. The bifurcation detector reaches an average recall and precision of 76% and 87% respectively during the vessel tree formation process. The final vessel tree formation achieves a median recall of 68% and precision of 70%, which is in line with the interobserver agreement.
Display omitted
PDF
