Atherosclerosis Imaging and the Canadian Atherosclerosis Imaging Network

Synthesis of diffusion-weighted MRI scalar maps from FLAIR volumes using generative adversarial networks

Introduction

Acquisition and pre-processing pipelines for diffusion-weighted imaging (DWI) volumes are resource- and time-consuming. Generating synthetic DWI scalar maps from commonly acquired brain MRI sequences such as fluid-attenuated inversion recovery (FLAIR) could be useful for supplementing datasets. In this work we design and compare GAN-based image translation models for generating DWI scalar maps from FLAIR MRI for the first time.

Methods

We evaluate a pix2pix model, two modified CycleGANs using paired and unpaired data, and a convolutional autoencoder in synthesizing DWI fractional anisotropy (FA) and mean diffusivity (MD) from whole FLAIR volumes. In total, 420 FLAIR and DWI volumes (11,957 images) from multi-center dementia and vascular disease cohorts were used for training/testing. Generated images were evaluated using two groups of metrics: (1) human perception metrics including peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), (2) structural metrics including a newly proposed histogram similarity (Hist-KL) metric and mean squared error (MSE).

Results

Pix2pix demonstrated the best performance both quantitatively and qualitatively with mean PSNR, SSIM, and MSE metrics of 23.41 dB, 0.8, 0.004, respectively for MD generation, and 24.05 dB, 0.78, 0.004, respectively for FA generation. The new histogram similarity metric demonstrated sensitivity to differences in fine details between generated and real images with mean pix2pix MD and FA Hist-KL metrics of 11.73 and 3.74, respectively. Detailed analysis of clinically relevant regions of white matter (WM) and gray matter (GM) in the pix2pix images also showed strong significant (p < 0.001) correlations between real and synthetic FA values in both tissue types (R = 0.714 for GM, R = 0.877 for WM).

Discussion/conclusion

Our results show that pix2pix's FA and MD models had significantly better structural similarity of tissue structures and fine details than other models, including WM tracts and CSF spaces, between real and generated images. Regional analysis of synthetic volumes showed that synthetic DWI images can not only be used to supplement clinical datasets, but demonstrates potential utility in bypassing or correcting registration in data pre-processing.

Automated Midline Estimation for Symmetry Analysis of Cerebral Hemispheres in FLAIR MRI

To perform brain asymmetry studies in large neuroimaging archives, reliable and automatic detection of the interhemispheric fissure (IF) is needed to first extract the cerebral hemispheres. The detection of the IF is often referred to as mid-sagittal plane estimation, as this plane separates the two cerebral hemispheres. However, traditional planar estimation techniques fail when the IF presents a curvature caused by existing pathology or a natural phenomenon known as brain torque. As a result, midline estimates can be inaccurate. In this study, a fully unsupervised midline estimation technique is proposed that is comprised of three main stages: head angle correction, control point estimation and midline generation. The control points are estimated using a combination of intensity, texture, gradient, and symmetry-based features. As shown, the proposed method automatically adapts to IF curvature, is applied on a slice-to-slice basis for more accurate results and also provides accurate delineation of the midline in the septum pellucidum, which is a source of failure for traditional approaches. The method is compared to two state-of-the-art methods for midline estimation and is validated using 75 imaging volumes (~3,000 imaging slices) acquired from 38 centers of subjects with dementia and vascular disease. The proposed method yields the lowest average error across all metrics: Hausdorff distance (HD) was 0.32 ± 0.23, mean absolute difference (MAD) was 1.10 ± 0.38 mm and volume difference was 7.52 ± 5.40 and 5.35 ± 3.97 ml, for left and right hemispheres, respectively. Using the proposed method, the midline was extracted for 5,360 volumes (~275K images) from 83 centers worldwide, acquired by GE, Siemens and Philips scanners. An asymmetry index was proposed that automatically detected outlier segmentations (which were <1% of the total dataset). Using the extracted hemispheres, hemispheric asymmetry texture biomarkers of the normal-appearing brain matter (NABM) were analyzed in a dementia cohort, and significant differences in biomarker means were found across SCI and MCI and SCI and AD.

Advances in Multimodality Carotid Plaque Imaging: AJR Expert Panel Narrative Review

Contemporary imaging methods provide detailed visualization of carotid athero-sclerotic plaque, enabling a major evolution of in vivo carotid plaque imaging evaluation. The degree of luminal stenosis in the carotid artery bifurcation, as assessed by ultrasound, has historically served as the primary imaging feature for determining ischemic stroke risk and the potential need for surgery. However, stroke risk may be more strongly driven by the presence of specific characteristics of vulnerable plaque, as visualized on CT and MRI, than by traditional ultrasound-based assessment of luminal narrowing. This review highlights six promising imaging-based plaque characteristics that harbor unique information regarding plaque vulnerability: maximum plaque thickness and volume, calcification, ulceration, intraplaque hemorrhage, lipid-rich necrotic core, and thin or ruptured fibrous cap. Increasing evidence supports the association of these plaque characteristics with risk of ischemic stroke, although these characteristics have varying suitability for clinical implementation. Key aspects of CT and MRI protocols for carotid plaque imaging are also considered. Practical next steps and hurdles are explored for implementing routine imaging assessment of these plaque characteristics in addition to, or even as replacement for, traditional assessment of the degree of vascular stenosis on ultrasound, in the identification of individuals at high risk of ischemic stroke.

A Voxel-Based Fully Convolution Network and Continuous Max-Flow for Carotid Vessel-Wall-Volume Segmentation From 3D Ultrasound Images

Vessel-wall-volume (VWV) is an important three-dimensional ultrasound (3DUS) metric used in the assessment of carotid plaque burden and monitoring changes in carotid atherosclerosis in response to medical treatment. To generate the VWV measurement, we proposed an approach that combined a voxel-based fully convolution network (Voxel-FCN) and a continuous max-flow module to automatically segment the carotid media-adventitia (MAB) and lumen-intima boundaries (LIB) from 3DUS images. Voxel-FCN includes an encoder consisting of a general 3D CNN and a 3D pyramid pooling module to extract spatial and contextual information, and a decoder using a concatenating module with an attention mechanism to fuse multi-level features extracted by the encoder. A continuous max-flow algorithm is used to improve the coarse segmentation provided by the Voxel-FCN. Using 1007 3DUS images, our approach yielded a Dice-similarity-coefficient (DSC) of 93.2±3.0% for the MAB in the common carotid artery (CCA), and 91.9±5.0% in the bifurcation by comparing algorithm and expert manual segmentations. We achieved a DSC of 89.5±6.7% and 89.3±6.8% for the LIB in the CCA and the bifurcation respectively. The mean errors between the algorithm-and manually-generated VWVs were 0.2±51.2 mm³ for the CCA and -4.0±98.2 mm³ for the bifurcation. The algorithm segmentation accuracy was comparable to intra-observer manual segmentation but our approach required less than 1s, which will not alter the clinical work-flow as 10s is required to image one side of the neck. Therefore, we believe that the proposed method could be used clinically for generating VWV to monitor progression and regression of carotid plaques.

Whole volume brain extraction for multi-centre, multi-disease FLAIR MRI datasets

Automatic segmentation of the brain from magnetic resonance images (MRI) is a fundamental step in many neuroimaging processing frameworks. There are mature technologies for this task for T1- and T2-weighted MRI; however, a widely-accepted brain extraction method for Fluid-Attenuated Inversion Recovery (FLAIR) MRI has yet to be established. FLAIR MRI are becoming increasingly important for the analysis of neurodegenerative diseases and tools developed for this sequence would have clinical value. To maximize translation opportunities and for large scale research studies, algorithms for brain extraction in FLAIR MRI should generalize to multi-centre (MC) data. To this end, this work proposes a fully automated, whole volume brain extraction methodology for MC FLAIR MRI datasets. The framework is built using a novel standardization framework which reduces acquisition artifacts, standardizes the intensities of tissues and normalizes the spatial coordinates of brain tissue across MC datasets. Using the standardized datasets, an intuitive set of features based on intensity, spatial location and gradients are extracted and classified using a random forest (RF) classifier to segment the brain tissue class. A series of experiments were conducted to optimize classifier parameters, and to determine segmentation accuracy for standardized and unstandardized (original) data, as a function of scanner vendor, feature type and disease type. The models are trained, tested and validated on 156 image volumes (∼8000 image slices) from two multi-centre, multi-disease datasets, acquired with varying imaging parameters from 30 centres and three scanner vendors. The image datasets, denoted as CAIN and ADNI for vascular and dementia disease, respectively, represent a diverse collection of MC data to test the generalization capabilities of the proposed design. Results demonstrate the importance of standardization for segmentation of MC data, as models trained on standardized data yielded a drastic improvement in brain extraction accuracy compared to the original, unstandardized data (CAIN: DSC = 91% and ADNI: DSC = 86% vs. CAIN: 78% and ADNI: 65%). It was also found that models created from one scanner vendor based on unstandardized data yielded poor segmentation results in data acquired from other scanner vendors, which was improved through standardization. These results demonstrate that to create consistency in segmentations from multi-institutional datasets it is paramount that MC variability be mitigated to improve stability and to ensure generalization of machine learning algorithms for MRI.

Imaging biomarkers of vulnerable carotid plaques for stroke risk prediction and their potential clinical implications

Stroke represents a massive public health problem. Carotid atherosclerosis plays a fundamental part in the occurence of ischaemic stroke. European and US guidelines for prevention of stroke in patients with carotid plaques are based on quantification of the percentage reduction in luminal diameter due to the atherosclerotic process to select the best therapeutic approach. However, better strategies for prevention of stroke are needed because some subtypes of carotid plaques (eg, vulnerable plaques) can predict the occurrence of stroke independent of the degree of stenosis. Advances in imaging techniques have enabled routine characterisation and detection of the features of carotid plaque vulnerability. Intraplaque haemorrhage is accepted by neurologists and radiologists as one of the features of vulnerable plaques, but other characteristics-eg, plaque volume, neovascularisation, and inflammation-are promising as biomarkers of carotid plaque vulnerability. These biomarkers could change current management strategies based merely on the degree of stenosis.

[18F]-Fluorodeoxyglucose PET/CT imaging as a marker of carotid plaque inflammation: Comparison to immunohistology and relationship to acuity of events

BACKGROUND

[18F]-fluorodeoxyglucose (¹⁸FDG) uptake imaged with positron emission tomography (PET) and computed tomography (CT) may serve as a biomarker of plaque inflammation. This study evaluated the relationship between carotid plaque ¹⁸FDG uptake and a) intraplaque expression of macrophage and macrophage-like cellular CD68 immunohistology; b) intraplaque inflammatory burden using leukocyte-sensitive CD45 immunohistology; c) symptomatic patient presentation; d) time from last cerebrovascular event.

METHODS

54 patients scheduled for carotid endarterectomy underwent ¹⁸FDG PET/CT imaging. Maximum 18FDG uptake (SUV_max) and tissue-to-blood ratio (TBR_max) was measured for carotid plaques. Quantitative immunohistological analysis of macrophage-like cell expression (CD68) and leukocyte content (CD45) was performed.

RESULTS

¹⁸FDG uptake was related to CD68 macrophage expression (TBR_max: r = 0.51, p < 0.001), and total-plaque leukocyte CD45 expression (TBR_max: r = 0.632, p = 0.009, p < 0.001). ¹⁸FDG TBR_max uptake in carotid plaque associated with patient symptoms was greater than asymptomatic plaque (3.58 ± 1.01 vs. 3.13 ± 1.10, p = 0.008). ¹⁸FDG uptake differed between an acuity threshold of <90 days and >90 days (SUV_max:3.15 ± 0.87 vs. 2.52 ± 0.45, p = 0.015).

CONCLUSIONS

In this CAIN cohort, ¹⁸FDG uptake imaged with PET/CT serves a surrogate marker of intraplaque inflammatory macrophage, macrophage-like cell and leukocyte burden. 18FDG uptake is greater in plaque associated with patient symptoms and those with recent cerebrovascular events. Future studies are needed to relate ¹⁸FDG uptake and disease progression.

Carotid Artery Wall Imaging: Perspective and Guidelines from the ASNR Vessel Wall Imaging Study Group and Expert Consensus Recommendations of the American Society of Neuroradiology

Identification of carotid artery atherosclerosis is conventionally based on measurements of luminal stenosis and surface irregularities using in vivo imaging techniques including sonography, CT and MR angiography, and digital subtraction angiography. However, histopathologic studies demonstrate considerable differences between plaques with identical degrees of stenosis and indicate that certain plaque features are associated with increased risk for ischemic events. The ability to look beyond the lumen using highly developed vessel wall imaging methods to identify plaque vulnerable to disruption has prompted an active debate as to whether a paradigm shift is needed to move away from relying on measurements of luminal stenosis for gauging the risk of ischemic injury. Further evaluation in randomized clinical trials will help to better define the exact role of plaque imaging in clinical decision-making. However, current carotid vessel wall imaging techniques can be informative. The goal of this article is to present the perspective of the ASNR Vessel Wall Imaging Study Group as it relates to the current status of arterial wall imaging in carotid artery disease.

Age-Specific Sex Differences in Magnetic Resonance Imaging-Depicted Carotid Intraplaque Hemorrhage

BACKGROUND AND PURPOSE

Stroke rates are higher in men compared with women in the fourth through seventh decades of life, and higher rates may result from differences in carotid intraplaque hemorrhage (IPH), an unstable atherosclerotic plaque component. We report age-specific sex differences in the presence of magnetic resonance imaging-depicted carotid IPH.

METHODS

Patients (n=1115) underwent magnetic resonance imaging for carotid IPH between 2005 and 2014. Low-grade carotid stenosis patients (n=906) without prior endarterectomy were eligible for this cross-sectional study.

RESULTS

Of the 906 patients included (mean age±SD in years, 66.98±15.15), 63 (6.95%) had carotid IPH. In men and women, carotid IPH was present in 11.43% (48 of 420) and 3.09% (15 of 486), respectively (P<0.0001). Multivariable logistic regression analysis confirmed greater odds of carotid IPH in men for all ages: 45 to 54 (odds ratio=45.45; 95% confidence interval, 3.43-500), 55 to 64 years (odds ratio=21.74; 95% confidence interval, 3.21-142.86), 65 to 74 years (odds ratio=10.42; 95% confidence interval, 2.91-37.04), and ≥75 years (odds ratio=5.00; 95% confidence interval, 2.31-10.75). Male sex modified the effect of age on the presence of carotid IPH (β=0.074; SE=0.036; P=0.0411).

CONCLUSIONS

Men have greater age-specific odds of magnetic resonance imaging-depicted carotid IPH compared with women. With increasing age post-menopause, the odds of carotid IPH in women becomes closer to that of men. Delayed onset of carotid IPH in women, an unstable plaque component, may partly explain differential stroke rates between sexes, and further studies are warranted.

Low-glycaemic index diet to improve glycaemic control and cardiovascular disease in type 2 diabetes: design and methods for a randomised, controlled, clinical trial

INTRODUCTION

Type 2 diabetes (T2DM) produces macrovascular and microvascular damage, significantly increasing the risk of cardiovascular disease (CVD), renal failure and blindness. As rates of T2DM rise, the need for effective dietary and other lifestyle changes to improve diabetes management become more urgent. Low-glycaemic index (GI) diets may improve glycaemic control in diabetes in the short term; however, there is a lack of evidence on the long-term adherence to low-GI diets, as well as on the association with surrogate markers of CVD beyond traditional risk factors. Recently, advances have been made in measures of subclinical arterial disease through the use of MRI, which, along with standard measures from carotid ultrasound (CUS) scanning, have been associated with CVD events. We therefore designed a randomised, controlled, clinical trial to assess whether low-GI dietary advice can significantly improve surrogate markers of CVD and long-term glycaemic control in T2DM.

METHODS AND ANALYSIS

169 otherwise healthy individuals with T2DM were recruited to receive intensive counselling on a low-GI or high-cereal fibre diet for 3 years. To assess macrovascular disease, MRI and CUS are used, and to assess microvascular disease, retinal photography and 24-hour urinary collections are taken at baseline and years 1 and 3. Risk factors for CVD are assessed every 3 months.

ETHICS AND DISSEMINATION

The study protocol and consent form have been approved by the research ethics board of St. Michael's Hospital. If the study shows a benefit, these data will support the use of low-GI and/or high-fibre foods in the management of T2DM and its complications.

TRIAL REGISTRATION NUMBER

NCT01063374; Pre-results.

Assessment of Carotid Artery Stenosis and the Use of Statins

General thinking has previously centered on managing carotid artery stenosis (CAS) by carotid endarterectomy and subsequently, stenting for higher risk patients. However for CAS and other forms of vascular disease, especially when asymptomatic, there is new emphasis on defining underlying mechanisms. Knowledge of these mechanisms can lead to medical treatments that result in possible atherosclerotic plaque stabilization, and even plaque regression, including in the patient with CAS. For now, the key medication class for a medical approach are the statins. Their use is supported by good cardiovascular clinical trial evidence including some directed carotid artery studies, especially with a demonstrated decrease in carotid intima-media thickness. Procedural controversy still exists but the current era in medicine offers significant support for medical management of asymptomatic CAS while techniques to recognize the vulnerable plaque evolve. If CAS converts to a symptomatic status, early referral for endarterectomy or stenting is indicated.

Cardiac PET: metabolic and functional imaging of the myocardium

Cardiac PET has evolved over the past 30 years to gain wider acceptance as a valuable modality for a variety of cardiac conditions. Wider availability of scanners as well as changes in reimbursement policies in more recent years has further increased its use. Moreover, with the emergence of novel radionuclides as well as further advances in scanner technology, the use of cardiac PET can be expected to increase further in both clinical practice and the research arena. PET has demonstrated superior diagnostic accuracy for the diagnosis of coronary artery disease in comparison with single-photon emission tomography while it provides robust prognostic value. The addition of absolute flow quantification increases sensitivity for 3-vessel disease as well as providing incremental functional and prognostic information. Metabolic imaging using (18)F-fluorodeoxyglucose can be used to guide revascularization in the setting of heart failure and also to detect active inflammation in conditions such as cardiac sarcoidosis and within atherosclerotic plaque, improving our understanding of the processes that underlie these conditions. However, although the pace of new developments is rapid, there remains a gap in evidence for many of these advances and further studies are required.

Atherosclerotic plaque component segmentation in combined carotid MRI and CTA data incorporating class label uncertainty

Atherosclerotic plaque composition can indicate plaque vulnerability. We segment atherosclerotic plaque components from the carotid artery on a combination of in vivo MRI and CT-angiography (CTA) data using supervised voxelwise classification. In contrast to previous studies the ground truth for training is directly obtained from 3D registration with histology for fibrous and lipid-rich necrotic tissue, and with μCT for calcification. This registration does, however, not provide accurate voxelwise correspondence. We therefore evaluate three approaches that incorporate uncertainty in the ground truth used for training: I) soft labels are created by Gaussian blurring of the original binary histology segmentations to reduce weights at the boundaries between components, and are weighted by the estimated registration accuracy of the histology and in vivo imaging data (measured by overlap), II) samples are weighted by the local contour distance of the lumen and outer wall between histology and in vivo data, and III) 10% of each class is rejected by Gaussian outlier rejection. Classification was evaluated on the relative volumes (% of tissue type in the vessel wall) for calcified, fibrous and lipid-rich necrotic tissue, using linear discriminant (LDC) and support vector machine (SVM) classification. In addition, the combination of MRI and CTA data was compared to using only one imaging modality. Best results were obtained by LDC and outlier rejection: the volume error per vessel was 0.9±1.0% for calcification, 12.7±7.6% for fibrous and 12.1±8.1% for necrotic tissue, with Spearman rank correlation coefficients of 0.91 (calcification), 0.80 (fibrous) and 0.81 (necrotic). While segmentation using only MRI features yielded low accuracy for calcification, and segmentation using only CTA features yielded low accuracy for necrotic tissue, the combination of features from MRI and CTA gave good results for all studied components.

Imaging atherosclerosis with hybrid [18F]fluorodeoxyglucose positron emission tomography/computed tomography imaging: what Leonardo da Vinci could not see

Prodigious efforts and landmark discoveries have led toward significant advances in our understanding of atherosclerosis. Despite significant efforts, atherosclerosis continues globally to be a leading cause of mortality and reduced quality of life. With surges in the prevalence of obesity and diabetes, atherosclerosis is expected to have an even more pronounced impact upon the global burden of disease. It is imperative to develop strategies for the early detection of disease. Positron emission tomography (PET) imaging utilizing [(18)F]fluorodeoxyglucose (FDG) may provide a non-invasive means of characterizing inflammatory activity within atherosclerotic plaque, thus serving as a surrogate biomarker for detecting vulnerable plaque. The aim of this review is to explore the rationale for performing FDG imaging, provide an overview into the mechanism of action, and summarize findings from the early application of FDG PET imaging in the clinical setting to evaluate vascular disease. Alternative imaging biomarkers and approaches are briefly discussed.