Evaluating the Outcomes and Trainee Performance of a Canadian Medical Imaging Clinician Investigator Program.

Purpose: To measure the research productivity of trainees from the University of Toronto's Medical Imaging Clinician Investigator Program (MI-CIP) and comparing it with the research productivity of trainees from MI-non-CIP and General Surgery (GSx) Clinician Investigator Program. Methods: We identified residents who completed an MI-CIP, MI-non-CIP and GSx-CIP from 2006-2016. In each group of trainees, we assessed 3 research productivity outcomes with non-parametric tests before residency and at 7 years post-CIP completion/post-graduation. Research productivity outcomes include the number of total publications, the number of first-author publications, and the publication's average journal impact factor (IF). Results: We identified 11 MI-CIP trainees (male/female: 9 [82%]/2 [18%]), 74 MI-non-CIP trainees (46 [62%]/28 [38%]) and 41 GSx-CIP trainees (23 [56%]/18 [44%]). MI-CIP trainees had statistically significant higher research productivity than MI-non-CIP in all measured outcomes. The median (interquartile range, IQR) number of total publications of MI-CIP vs MI-non-CIP trainees was 5.0 (8.0) vs 1.0 (2.0) before residency and 6.0 (10.0) vs .0 (2.0) at 7 years post-CIP completion/post-graduation. The median (IQR) first-author publications of MI-CIP vs MI-non-CIP trainees was 2.0 (3.0) vs .0 (1.0) before residency and 2.0 (4.0) vs (.0) (1.0) at 7 years post-CIP completion/post-graduation. The median (IQR) average journal IF of MI-CIP vs MI-non-CIP trainees was 3.2 (2.0) vs .3 (2.4) before residency and 3.9 (3.2) vs .0 (2.6) at 7 years post-CIP completion/post-graduation. Between MI-CIP and GSx-CIP trainees, there were no significant differences in research productivity in all measured outcomes. Conclusion: MI-CIP trainees actively conducted research after graduation. These trainees demonstrated early research engagement before residency. The similar research productivity of MI-CIP vs GSx-CIP trainees shows initial success of MI-CIP trainees.

Carotid Plaque-RADS: A Novel Stroke Risk Classification System.

BACKGROUND: Carotid artery atherosclerosis is highly prevalent in the general population and is a well-established risk factor for acute ischemic stroke. Although the morphological characteristics of vulnerable plaques are well recognized, there is a lack of consensus in reporting and interpreting carotid plaque features. OBJECTIVES: The aim of this paper is to establish a consistent and comprehensive approach for imaging and reporting carotid plaque by introducing the Plaque-RADS (Reporting and Data System) score. METHODS: A panel of experts recognized the necessity to develop a classification system for carotid plaque and its defining characteristics. Using a multimodality analysis approach, the Plaque-RADS categories were established through consensus, drawing on existing published reports. RESULTS: The authors present a universal classification that is applicable to both researchers and clinicians. The Plaque-RADS score offers a morphological assessment in addition to the prevailing quantitative parameter of "stenosis." The Plaque-RADS score spans from grade 1 (indicating complete absence of plaque) to grade 4 (representing complicated plaque). Accompanying visual examples are included to facilitate a clear understanding of the Plaque-RADS categories. CONCLUSIONS: Plaque-RADS is a standardized and reliable system of reporting carotid plaque composition and morphology via different imaging modalities, such as ultrasound, computed tomography, and magnetic resonance imaging. This scoring system has the potential to help in the precise identification of patients who may benefit from exclusive medical intervention and those who require alternative treatments, thereby enhancing patient care. A standardized lexicon and structured reporting promise to enhance communication between radiologists, referring clinicians, and scientists.

Brain Age Estimation on a Dementia Cohort Using FLAIR MRI Biomarkers.

BACKGROUND AND PURPOSE: The prodromal stage of Alzheimer's disease presents an imperative intervention window. This work focuses on using brain age prediction models and biomarkers from FLAIR MR imaging to identify subjects who progress to Alzheimer's disease (converting mild cognitive impairment) or those who remain stable (stable mild cognitive impairment). MATERIALS AND METHODS: A machine learning model was trained to predict the age of normal control subjects on the basis of volume, intensity, and texture features from 3239 FLAIR MRI volumes. The brain age gap estimation (BrainAGE) was computed as the difference between the predicted and true age, and it was used as a biomarker for both cross-sectional and longitudinal analyses. Differences in biomarker means, slopes, and intercepts were investigated using ANOVA and Tukey post hoc test. Correlation analysis was performed between brain age gap estimation and established Alzheimer's disease indicators. RESULTS: The brain age prediction model showed accurate results (mean absolute error = 2.46 years) when testing on held out normal control data. The computed BrainAGE metric showed significant differences between the stable mild cognitive impairment and converting mild cognitive impairment groups in cross-sectional and longitudinal analyses, most notably showing significant differences up to 4 years before conversion to Alzheimer's disease. A significant correlation was found between BrainAGE and previously established Alzheimer's disease conversion biomarkers. CONCLUSIONS: The BrainAGE metric can allow clinicians to consider a single explainable value that summarizes all the biomarkers because it considers many dimensions of disease and can determine whether the subject has normal aging patterns or if he or she is trending into a high-risk category using a single value.

Detecting conversion from mild cognitive impairment to Alzheimer's disease using FLAIR MRI biomarkers.

Mild cognitive impairment (MCI) is the prodromal phase of Alzheimer's disease (AD) and while it presents as an imperative intervention window, it is difficult to detect which subjects convert to AD (cMCI) and which ones remain stable (sMCI). The objective of this work was to investigate fluid-attenuated inversion recovery (FLAIR) MRI biomarkers and their ability to differentiate between sMCI and cMCI subjects in cross-sectional and longitudinal data. Three types of biomarkers were investigated: volume, intensity and texture. Volume biomarkers included total brain volume, cerebrospinal fluid volume (CSF), lateral ventricular volume, white matter lesion volume, subarachnoid CSF, and grey matter (GM) and white matter (WM), all normalized to intracranial volume. The mean intensity, kurtosis, and skewness of the GM and WM made up the intensity features. Texture features quantified homogeneity and microstructural tissue changes of GM and WM regions. Composite indices were also considered, which are biomarkers that represent an aggregate sum (z-score normalization and summation) of all biomarkers. The FLAIR MRI biomarkers successfully identified high-risk subjects as significant differences (p < 0.05) were found between the means of the sMCI and cMCI groups and the rate of change over time for several individual biomarkers as well as the composite indices for both cross-sectional and longitudinal analyses. Classification accuracy and feature importance analysis showed volume biomarkers to be most predictive, however, best performance was obtained when complimenting the volume biomarkers with the intensity and texture features. Using all the biomarkers, accuracy of 86.2 % and 69.2 % was achieved for normal control-AD and sMCI-cMCI classification respectively. Survival analysis demonstrated that the majority of the biomarkers showed a noticeable impact on the AD conversion probability 4 years prior to conversion. Composite indices were the top performers for all analyses including feature importance, classification, and survival analysis. This demonstrated their ability to summarize various dimensions of disease into single-valued metrics. Significant correlation (p < 0.05) with phosphorylated-tau and amyloid-beta CSF biomarkers was found with all the FLAIR biomarkers. The proposed biomarker system is easily attained as FLAIR is routinely acquired, models are not computationally intensive and the results are explainable, thus making this pipeline easily integrated into clinical workflow.

Segmentation of Brain Metastases Using Background Layer Statistics (BLAST).

BACKGROUND AND PURPOSE: Accurate segmentation of brain metastases is important for treatment planning and evaluating response. The aim of this study was to assess the performance of a semiautomated algorithm for brain metastases segmentation using Background Layer Statistics (BLAST). MATERIALS AND METHODS: Nineteen patients with 48 parenchymal and dural brain metastases were included. Segmentation was performed by 4 neuroradiologists and 1 radiation oncologist. K-means clustering was used to identify normal gray and white matter (background layer) in a 2D parameter space of signal intensities from postcontrast T2 FLAIR and T1 MPRAGE sequences. The background layer was subtracted and operator-defined thresholds were applied in parameter space to segment brain metastases. The remaining voxels were back-projected to visualize segmentations in image space and evaluated by the operators. Segmentation performance was measured by calculating the Dice-Sørensen coefficient and Hausdorff distance using ground truth segmentations made by the investigators. Contours derived from the segmentations were evaluated for clinical acceptance using a 5-point Likert scale. RESULTS: The median Dice-Sørensen coefficient was 0.82 for all brain metastases and 0.9 for brain metastases of ≥10 mm. The median Hausdorff distance was 1.4 mm. Excellent interreader agreement for brain metastases volumes was found with an intraclass correlation coefficient = 0.9978. The median segmentation time was 2.8 minutes/metastasis. Forty-five contours (94%) had a Likert score of 4 or 5, indicating that the contours were acceptable for treatment, requiring no changes or minor edits. CONCLUSIONS: We show accurate and reproducible segmentation of brain metastases using BLAST and demonstrate its potential as a tool for radiation planning and evaluating treatment response.

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.

Alzheimer's and vascular disease classification using regional texture biomarkers in FLAIR MRI.

Interactions between subcortical vascular disease and dementia due to Alzheimer's disease (AD) are unclear, and clinical overlap between the diseases makes diagnosis challenging. Existing studies have shown regional microstructural changes specific to each disease, and that textures in fluid-attenuated inversion recovery (FLAIR) MRI images may characterize abnormalities in tissue microstructure. This work aims to investigate regional FLAIR biomarkers that can differentiate dementia cohorts with and without subcortical vascular disease. FLAIR and diffusion MRI (dMRI) volumes were obtained in 65 mild cognitive impairment (MCI), 21 AD, 44 subcortical vascular MCI (scVMCI), 22 Mixed etiology, and 48 healthy elderly patients. FLAIR texture and intensity biomarkers were extracted from the normal appearing brain matter (NABM), WML penumbra, blood supply territory (BST), and white matter tract regions of each patient. All FLAIR biomarkers were correlated to dMRI metrics in each region and global WML load, and biomarker means between groups were compared using ANOVA. Binary classifications were performed using Random Forest classifiers to investigate the predictive nature of the regional biomarkers, and SHAP feature analysis was performed to further investigate optimal regions of interest for differentiating disease groups. The regional FLAIR biomarkers were strongly correlated to MD, while all biomarker regions but white matter tracts were strongly correlated to WML burden. Classification between Mixed disease and healthy, AD, and scVMCI patients yielded accuracies of 97%, 81%, and 72% respectively using WM tract biomarkers. Classification between scVMCI and healthy, MCI, and AD patients yielded accuracies of 89%, 84%, and 79% respectively using penumbra biomarkers. Only the classification between AD and healthy patients had optimal results using NABM biomarkers. This work presents novel regional FLAIR biomarkers that may quantify white matter degeneration related to subcortical vascular disease, and which indicate that investigating degeneration in specific regions may be more important than assessing global WML burden in vascular disease groups.

Glycemic Index Versus Wheat Fiber on Arterial Wall Damage in Diabetes: A Randomized Controlled Trial.

OBJECTIVE: High cereal fiber and low-glycemic index (GI) diets are associated with reduced cardiovascular disease (CVD) risk in cohort studies. Clinical trial evidence on event incidence is lacking. Therefore, to make trial outcomes more directly relevant to CVD, we compared the effect on carotid plaque development in diabetes of a low-GI diet versus a whole-grain wheat-fiber diet. RESEARCH DESIGN AND METHODS: The study randomized 169 men and women with well-controlled type 2 diabetes to counseling on a low GI-diet or whole-grain wheat-fiber diet for 3 years. Change in carotid vessel wall volume (VWV) (prespecified primary end point) was assessed by MRI as an indication of arterial damage. RESULTS: Of 169 randomized participants, 134 completed the study. No treatment differences were seen in VWV. However, on the whole-grain wheat-fiber diet, VWV increased significantly from baseline, 23 mm3 (95% CI 4, 41; P = 0.016), but not on the low-GI diet, 8 mm3 (95% CI -10, 26; P = 0.381). The low-GI diet resulted in preservation of renal function, as estimated glomerular filtration rate, compared with the reduction following the wheat-fiber diet. HbA1c was modestly reduced over the first 9 months in the intention-to-treat analysis and extended with greater compliance to 15 months in the per-protocol analysis. CONCLUSIONS: Since the low-GI diet was similar to the whole-grain wheat-fiber diet recommended for cardiovascular risk reduction, the low-GI diet may also be effective for CVD risk reduction.

Delusional Severity Is Associated with Abnormal Texture in FLAIR MRI.

Background: This study examines the relationship between delusional severity in cognitively impaired adults with automatically computed volume and texture biomarkers from the Normal Appearing Brain Matter (NABM) in FLAIR MRI. Methods: Patients with mild cognitive impairment (MCI, n = 24) and Alzheimer's Disease (AD, n = 18) with delusions of varying severities based on Neuropsychiatric Inventory-Questionnaire (NPI-Q) (1­mild, 2­moderate, 3­severe) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were analyzed for this task. The NABM region, which is gray matter (GM) and white matter (WM) combined, was automatically segmented in FLAIR MRI volumes with intensity standardization and thresholding. Three imaging biomarkers were computed from this region, including NABM volume and two texture markers called "Integrity" and "Damage". Together, these imaging biomarkers quantify structural changes in brain volume, microstructural integrity and tissue damage. Multivariable regression was used to investigate relationships between imaging biomarkers and delusional severities (1, 2 and 3). Sex, age, education, APOE4 and baseline cerebrospinal fluid (CSF) tau were included as co-variates. Results: Biomarkers were extracted from a total of 42 participants with longitudinal time points representing 164 imaging volumes. Significant associations were found for all three NABM biomarkers between delusion level 3 and level 1. Integrity was also sensitive enough to show differences between delusion level 1 and delusion level 2. A significant specified interaction was noted with severe delusions (level 3) and CSF tau for all imaging biomarkers (p < 0.01). APOE4 homozygotes were also significantly related to the biomarkers. Conclusion: Cognitively impaired older adults with more severe delusions have greater global brain disease burden in the WM and GM combined (NABM) as measured using FLAIR MRI. Relative to patients with mild delusions, tissue degeneration in the NABM was more pronounced in subjects with higher delusional symptoms, with a significant association with CSF tau. Future studies are required to establish potential tau-associated mechanisms of increased delusional severity.

Fabrication of Customizable Intraplaque Hemorrhage Phantoms for Magnetic Resonance Imaging.

PURPOSE: Magnetic resonance (MR) imaging detection of methemoglobin, a molecular marker of intraplaque hemorrhage (IPH), in atherosclerotic plaque is a promising method of assessing stroke risk. However, the multicenter imaging studies required to further validate this technique necessitate the development of IPH phantoms to standardize images acquired across different scanners. This study developed a set of phantoms that modeled methemoglobin-laden IPH for use in MR image standardization. PROCEDURES: A time-stable material mimicking the MR properties of methemoglobin in IPH was created by doping agarose hydrogel with gadolinium and sodium alginate. This material was used to create a phantom that consisted of 9 cylindrical IPH sites (with sizes from 1 to 8 mm). Anatomical replicas of IPH-positive atherosclerosis were also created using 3D printed molds. These plaque replicas also modeled other common plaque components including a lipid core and atheroma cap. T1 mapping and a magnetization-prepared rapid acquisition gradient echo (MPRAGE) carotid imaging protocol were used to assess phantom realism and long-term stability. RESULTS: Cylindrical phantom IPH sites possessed a T1 time of 335 ± 51 ms and exhibited little change in size or MPRAGE signal intensity over 31 days; the mean (SD) magnitude of changes in size and signal were 6.4 % (2.7 %) and 7.3 % (6.7 %), respectively. IPH sites incorporated into complex anatomical plaque phantoms exhibited contrast comparable to clinical images. CONCLUSIONS: The cylindrical IPH phantom accurately modeled the short T1 time characteristic of methemoglobin-laden IPH, with the IPH sites exhibiting little variation in imaging properties over 31 days. Furthermore, MPRAGE images of the anatomical atherosclerosis replicas closely matched those of clinical plaques. In combination, these phantoms will allow for IPH imaging protocol standardization and thus facilitate future multicenter IPH imaging.

The impact of reporting magnetic resonance imaging incidental findings in the Canadian alliance for healthy hearts and minds cohort.

BACKGROUND: In the Canadian Alliance for Healthy Hearts and Minds (CAHHM) cohort, participants underwent magnetic resonance imaging (MRI) of the brain, heart, and abdomen, that generated incidental findings (IFs). The approach to managing these unexpected results remain a complex issue. Our objectives were to describe the CAHHM policy for the management of IFs, to understand the impact of disclosing IFs to healthy research participants, and to reflect on the ethical obligations of researchers in future MRI studies. METHODS: Between 2013 and 2019, 8252 participants (mean age 58 ± 9 years, 54% women) were recruited with a follow-up questionnaire administered to 909 participants (40% response rate) at 1-year. The CAHHM policy followed a restricted approach, whereby routine feedback on IFs was not provided. Only IFs of severe structural abnormalities were reported. RESULTS: Severe structural abnormalities occurred in 8.3% (95% confidence interval 7.7-8.9%) of participants, with the highest proportions found in the brain (4.2%) and abdomen (3.1%). The majority of participants (97%) informed of an IF reported no change in quality of life, with 3% of participants reporting that the knowledge of an IF negatively impacted their quality of life. Furthermore, 50% reported increased stress in learning about an IF, and in 95%, the discovery of an IF did not adversely impact his/her life insurance policy. Most participants (90%) would enrol in the study again and perceived the MRI scan to be beneficial, regardless of whether they were informed of IFs. While the implications of a restricted approach to IF management was perceived to be mostly positive, a degree of diagnostic misconception was present amongst participants, indicating the importance of a more thorough consent process to support participant autonomy. CONCLUSION: The management of IFs from research MRI scans remain a challenging issue, as participants may experience stress and a reduced quality of life when IFs are disclosed. The restricted approach to IF management in CAHHM demonstrated a fair fulfillment of the overarching ethical principles of respect for autonomy, concern for wellbeing, and justice. The approach outlined in the CAHHM policy may serve as a framework for future research studies. Clinical trial registration https://clinicaltrials.gov/ct2/show/NCT02220582 .

Effect of Cognitive Reserve on the Association of Vascular Brain Injury With Cognition: Analysis of the PURE and CAHHM Studies.

BACKGROUND AND OBJECTIVES: To determine whether cognitive reserve attenuates the association of vascular brain injury with cognition. METHODS: Cross-sectional data were analyzed from 2 harmonized studies: the Canadian Alliance for Healthy Hearts and Healthy Minds (CAHHM) and the Prospective Urban and Rural Epidemiology (PURE) study. Markers of cognitive reserve were education, involvement in social activities, marital status, height, and leisure physical activity, which were combined into a composite score. Vascular brain injury was defined as nonlacunar brain infarcts or high white matter hyperintensity (WMH) burden on MRI. Cognition was assessed using the Montreal Cognitive Assessment Tool (MoCA) and the Digit Symbol Substitution Test (DSST). RESULTS: There were 10,916 participants age 35-81. Mean age was 58.8 years (range 35-81) and 55.8% were female. Education, moderate leisure physical activity, being in a marital partnership, being taller, and participating in social groups were each independently associated with higher cognition, as was the composite cognitive reserve score. Vascular brain injury was associated with lower cognition (ß -0.35 [95% confidence interval [CI] -0.53 to -0.17] for MoCA and ß -2.19 [95% CI -3.22 to -1.15] for DSST) but the association was not modified by the composite cognitive reserve variable (interaction p = 0.59 for MoCA and p = 0.72 for DSST). CONCLUSIONS: Both vascular brain injury and markers of cognitive reserve are associated with cognition. However, the effects were independent such that the adverse effects of covert vascular brain injury were not attenuated by higher cognitive reserve. To improve cognitive brain health, interventions to both prevent cerebrovascular disease and promote positive lifestyles are needed.

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.

Multinational Survey of Current Practice from Imaging to Treatment of Atherosclerotic Carotid Stenosis.

BACKGROUND: In the last 20-30 years, there have been many advances in imaging and therapeutic strategies for symptomatic and asymptomatic individuals with carotid artery stenosis. Our aim was to examine contemporary multinational practice standards. METHODS: Departmental Review Board approval for this study was obtained, and 3 authors prepared the 44 multiple choice survey questions. Endorsement was obtained by the European Society of Neuroradiology, American Society of Functional Neuroradiology, and African Academy of Neurology. A link to the online questionnaire was sent to their respective members and members of the Faculty Advocating Collaborative and Thoughtful Carotid Artery Treatments (FACTCATS). The questionnaire was open from May 16 to July 16, 2019. RESULTS: The responses from 223 respondents from 46 countries were included in the analyses including 65.9% from academic university hospitals. Neuroradiologists/radiologists comprised 68.2% of respondents, followed by neurologists (15%) and vascular surgeons (12.9%). In symptomatic patients, half (50.4%) the respondents answered that the first exam they used to evaluate carotid bifurcation was ultrasound, followed by computed tomography angiography (CTA, 41.6%) and then magnetic resonance imaging (MRI 8%). In asymptomatic patients, the first exam used to evaluate carotid bifurcation was ultrasound in 88.8% of respondents, CTA in 7%, and MRA in 4.2%. The percent stenosis upon which carotid endarterectomy or stenting was recommended was reduced in the presence of imaging evidence of "vulnerable plaque features" by 66.7% respondents for symptomatic patients and 34.2% for asymptomatic patients with a smaller subset of respondents even offering procedural intervention to patients with <50% symptomatic or asymptomatic stenosis. CONCLUSIONS: We found heterogeneity in current practices of carotid stenosis imaging and management in this worldwide survey with many respondents including vulnerable plaque imaging into their decision analysis despite the lack of proven benefit from clinical trials. This study highlights the need for new clinical trials using vulnerable plaque imaging to select high-risk patients despite maximal medical therapy who may benefit from procedural intervention.

Reduced Cognitive Assessment Scores Among Individuals With Magnetic Resonance Imaging-Detected Vascular Brain Injury.

Background and Purpose- Little is known about the association between covert vascular brain injury and cognitive impairment in middle-aged populations. We investigated if scores on a cognitive screen were lower in individuals with higher cardiovascular risk, and those with covert vascular brain injury. Methods- Seven thousand five hundred forty-seven adults, aged 35 to 69 years, free of cardiovascular disease underwent a cognitive assessment using the Digital Symbol Substitution test and Montreal Cognitive Assessment, and magnetic resonance imaging (MRI) to detect covert vascular brain injury (high white matter hyperintensities, lacunar, and nonlacunar brain infarctions). Cardiovascular risk factors were quantified using the INTERHEART (A Global Study of Risk Factors for Acute Myocardial Infarction) risk score. Multivariable mixed models tested for independent determinants of reduced cognitive scores. The population attributable risk of risk factors and MRI vascular brain injury on low cognitive scores was calculated. Results- The mean age of participants was 58 (SD, 9) years; 55% were women. Montreal Cognitive Assessment and Digital Symbol Substitution test scores decreased significantly with increasing age (P<0.0001), INTERHEART risk score (P<0.0001), and among individuals with high white matter hyperintensities, nonlacunar brain infarction, and individuals with 3+ silent brain infarctions. Adjusted for age, sex, education, ethnicity covariates, Digital Symbol Substitution test was significantly lowered by 1.0 (95% CI, -1.3 to -0.7) point per 5-point cardiovascular risk score increase, 1.9 (95% CI, -3.2 to -0.6) per high white matter hyperintensities, 3.5 (95% CI, -6.4 to -0.7) per nonlacunar stroke, and 6.8 (95% CI, -11.5 to -2.2) when 3+ silent brain infarctions were present. No postsecondary education accounted for 15% (95% CI, 12-17), moderate and high levels of cardiovascular risk factors accounted for 19% (95% CI, 8-30), and MRI vascular brain injury accounted for 10% (95% CI, -3 to 22) of low test scores. Conclusions- Among a middle-aged community-dwelling population, scores on a cognitive screen were lower in individuals with higher cardiovascular risk factors or MRI vascular brain injury. Much of the population attributable risk of low cognitive scores can be attributed to lower educational attainment, higher cardiovascular risk factors, and MRI vascular brain injury.

Cardiovascular risk scoring and magnetic resonance imaging detected subclinical cerebrovascular disease.

AIMS: Cardiovascular risk factors are used for risk stratification in primary prevention. We sought to determine if simple cardiac risk scores are associated with magnetic resonance imaging (MRI)-detected subclinical cerebrovascular disease including carotid wall volume (CWV), carotid intraplaque haemorrhage (IPH), and silent brain infarction (SBI). METHODS AND RESULTS: A total of 7594 adults with no history of cardiovascular disease (CVD) underwent risk factor assessment and a non-contrast enhanced MRI of the carotid arteries and brain using a standardized protocol in a population-based cohort recruited between 2014 and 2018. The non-lab-based INTERHEART risk score (IHRS) was calculated in all participants; the Framingham Risk Score was calculated in a subset who provided blood samples (n = 3889). The association between these risk scores and MRI measures of CWV, carotid IPH, and SBI was determined. The mean age of the cohort was 58 (8.9) years, 55% were women. Each 5-point increase (∼1 SD) in the IHRS was associated with a 9 mm3 increase in CWV, adjusted for sex (P < 0.0001), a 23% increase in IPH [95% confidence interval (CI) 9-38%], and a 32% (95% CI 20-45%) increase in SBI. These associations were consistent for lacunar and non-lacunar brain infarction. The Framingham Risk Score was also significantly associated with CWV, IPH, and SBI. CWV was additive and independent to the risk scores in its association with IPH and SBI. CONCLUSION: Simple cardiovascular risk scores are significantly associated with the presence of MRI-detected subclinical cerebrovascular disease, including CWV, IPH, and SBI in an adult population without known clinical CVD.

Prediction of Stroke Risk by Detection of Hemorrhage in Carotid Plaques: Meta-Analysis of Individual Patient Data.

OBJECTIVES: The goal of this study was to compare the risk of stroke between patients with carotid artery disease with and without the presence of intraplaque hemorrhage (IPH) on magnetic resonance imaging. BACKGROUND: IPH in carotid stenosis increases the risk of cerebrovascular events. Uncertainty remains whether risk of stroke alone is increased and whether stroke is predicted independently of known risk factors. METHODS: Data were pooled from 7 cohort studies including 560 patients with symptomatic carotid stenosis and 136 patients with asymptomatic carotid stenosis. Hazards of ipsilateral ischemic stroke (primary outcome) were compared between patients with and without IPH, adjusted for clinical risk factors. RESULTS: IPH was present in 51.6% of patients with symptomatic carotid stenosis and 29.4% of patients with asymptomatic carotid stenosis. During 1,121 observed person-years, 66 ipsilateral strokes occurred. Presence of IPH at baseline increased the risk of ipsilateral stroke both in symptomatic (hazard ratio [HR]: 10.2; 95% confidence interval [CI]: 4.6 to 22.5) and asymptomatic (HR: 7.9; 95% CI: 1.3 to 47.6) patients. Among patients with symptomatic carotid stenosis, annualized event rates of ipsilateral stroke in those with IPH versus those without IPH were 9.0% versus 0.7% (<50% stenosis), 18.1% versus 2.1% (50% to 69% stenosis), and 29.3% versus 1.5% (70% to 99% stenosis). Annualized event rates among patients with asymptomatic carotid stenosis were 5.4% in those with IPH versus 0.8% in those without IPH. Multivariate analysis identified IPH (HR: 11.0; 95% CI: 4.8 to 25.1) and severe degree of stenosis (HR: 3.3; 95% CI: 1.4 to 7.8) as independent predictors of ipsilateral stroke. CONCLUSIONS: IPH is common in patients with symptomatic and asymptomatic carotid stenosis and is a stronger predictor of stroke than any known clinical risk factors. Magnetic resonance imaging might help identify patients with carotid disease who would benefit from revascularization.

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.

Sample-Size Determination Methodologies for Machine Learning in Medical Imaging Research: A Systematic Review.

PURPOSE: The required training sample size for a particular machine learning (ML) model applied to medical imaging data is often unknown. The purpose of this study was to provide a descriptive review of current sample-size determination methodologies in ML applied to medical imaging and to propose recommendations for future work in the field. METHODS: We conducted a systematic literature search of articles using Medline and Embase with keywords including "machine learning," "image," and "sample size." The search included articles published between 1946 and 2018. Data regarding the ML task, sample size, and train-test pipeline were collected. RESULTS: A total of 167 articles were identified, of which 22 were included for qualitative analysis. There were only 4 studies that discussed sample-size determination methodologies, and 18 that tested the effect of sample size on model performance as part of an exploratory analysis. The observed methods could be categorized as pre hoc model-based approaches, which relied on features of the algorithm, or post hoc curve-fitting approaches requiring empirical testing to model and extrapolate algorithm performance as a function of sample size. Between studies, we observed great variability in performance testing procedures used for curve-fitting, model assessment methods, and reporting of confidence in sample sizes. CONCLUSIONS: Our study highlights the scarcity of research in training set size determination methodologies applied to ML in medical imaging, emphasizes the need to standardize current reporting practices, and guides future work in development and streamlining of pre hoc and post hoc sample size approaches.