Analytical Approaches for the Segmentation of the Zebrafish Brain Vasculature.

With advancements in imaging techniques, data visualization allows new insights into fundamental biological processes of development and disease. However, although biomedical science is heavily reliant on imaging data, interpretation of datasets is still often based on subjective visual assessment rather than rigorous quantitation. This overview presents steps to validate image processing and segmentation using the zebrafish brain vasculature data acquired with light sheet fluorescence microscopy as a use case. Blood vessels are of particular interest to both medical and biomedical science. Specific image enhancement filters have been developed that enhance blood vessels in imaging data prior to segmentation. Using the Sato enhancement filter as an example, we discuss how filter application can be evaluated and optimized. Approaches from the medical field such as simulated, experimental, and augmented datasets can be used to gain the most out of the data at hand. Using such datasets, we provide an overview of how biologists and data analysts can assess the accuracy, sensitivity, and robustness of their segmentation approaches that allow extraction of objects from images. Importantly, even after optimization and testing of a segmentation workflow (e.g., from a particular reporter line to another or between immunostaining processes), its generalizability is often limited, and this can be tested using double-transgenic reporter lines. Lastly, due to the increasing importance of deep learning networks, a comparative approach can be adopted to study their applicability to biological datasets. In summary, we present a broad methodological overview ranging from image enhancement to segmentation with a mixed approach of experimental, simulated, and augmented datasets to assess and validate vascular segmentation using the zebrafish brain vasculature as an example. © 2022 The Authors. Current Protocols published by Wiley Periodicals LLC. HIGHLIGHTS: Simulated, experimental, and augmented datasets provide an alternative to overcome the lack of segmentation gold standards and phantom models for zebrafish cerebrovascular segmentation. Direct generalization of a segmentation approach to the data for which it was not optimized (e.g., different transgenics or antibody stainings) should be treated with caution. Comparison of different deep learning segmentation methods can be used to assess their applicability to data. Here, we show that the zebrafish cerebral vasculature can be segmented with U-Net-based architectures, which outperform SegNet architectures.

Zebrafish vascular quantification: a tool for quantification of three-dimensional zebrafish cerebrovascular architecture by automated image analysis.

Zebrafish transgenic lines and light sheet fluorescence microscopy allow in-depth insights into three-dimensional vascular development in vivo. However, quantification of the zebrafish cerebral vasculature in 3D remains highly challenging. Here, we describe and test an image analysis workflow for 3D quantification of the total or regional zebrafish brain vasculature, called zebrafish vasculature quantification (ZVQ). It provides the first landmark- or object-based vascular inter-sample registration of the zebrafish cerebral vasculature, producing population average maps allowing rapid assessment of intra- and inter-group vascular anatomy. ZVQ also extracts a range of quantitative vascular parameters from a user-specified region of interest, including volume, surface area, density, branching points, length, radius and complexity. Application of ZVQ to 13 experimental conditions, including embryonic development, pharmacological manipulations and morpholino-induced gene knockdown, shows that ZVQ is robust, allows extraction of biologically relevant information and quantification of vascular alteration, and can provide novel insights into vascular biology. To allow dissemination, the code for quantification, a graphical user interface and workflow documentation are provided. Together, ZVQ provides the first open-source quantitative approach to assess the 3D cerebrovascular architecture in zebrafish.

Tracer kinetic assessment of blood-brain barrier leakage and blood volume in cerebral small vessel disease: Associations with disease burden and vascular risk factors.

Subtle blood-brain barrier (BBB) permeability increases have been shown in small vessel disease (SVD) using various analysis methods. Following recent consensus recommendations, we used Patlak tracer kinetic analysis, considered optimal in low permeability states, to quantify permeability-surface area product (PS), a BBB leakage estimate, and blood plasma volume (vP) in 201 patients with SVD who underwent dynamic contrast-enhanced MRI scans. We ran multivariable regression models with a quantitative or qualitative metric of white matter hyperintensity (WMH) severity, demographic and vascular risk factors. PS increased with WMH severity in grey (B = 0.15, Confidence Interval (CI): [0.001,0.299], p = 0.049) and normal-appearing white matter (B = 0.015, CI: [-0.008,0.308], p = 0.062). Patients with more severe WMH had lower vP in WMH (B = -0.088, CI: [-0.138,-0.039], p < 0.001), but higher vP in normal-appearing white matter (B = 0.031, CI: [-0.004,0.065], p = 0.082). PS and vP were lower at older ages in WMH, grey and white matter. We conclude higher PS in normal-appearing tissue with more severe WMH suggests impaired BBB integrity beyond visible lesions indicating that the microvasculature is compromised in normal-appearing white matter and WMH. BBB dysfunction is an important mechanism in SVD, but associations with clinical variables are complex and underlying damage affecting vascular surface area may alter interpretation of tracer kinetic results.

Single-Input Multi-Output U-Net for Automated 2D Foetal Brain Segmentation of MR Images.

In this work, we develop the Single-Input Multi-Output U-Net (SIMOU-Net), a hybrid network for foetal brain segmentation inspired by the original U-Net fused with the holistically nested edge detection (HED) network. The SIMOU-Net is similar to the original U-Net but it has a deeper architecture and takes account of the features extracted from each side output. It acts similar to an ensemble neural network, however, instead of averaging the outputs from several independently trained models, which is computationally expensive, our approach combines outputs from a single network to reduce the variance of predications and generalization errors. Experimental results using 200 normal foetal brains consisting of over 11,500 2D images produced Dice and Jaccard coefficients of 94.2 ± 5.9% and 88.7 ± 6.9%, respectively. We further tested the proposed network on 54 abnormal cases (over 3500 images) and achieved Dice and Jaccard coefficients of 91.2 ± 6.8% and 85.7 ± 6.6%, respectively.

Post-stroke Cognition at 1 and 3 Years Is Influenced by the Location of White Matter Hyperintensities in Patients With Lacunar Stroke.

Lacunar strokes are a common type of ischemic stroke. They are known to have long-term cognitive deficits, but the influencing factors are still largely unknown. We investigated if the location of the index lacunar stroke or regional WMH and their change at 1 year could predict the cognitive performance at 1 and 3 years post-stroke in lacunar stroke patients. We used lacunar lesion location and WMH-segmented data from 118 patients, mean age 64.9 who had a brain MRI scan soon after presenting with symptoms, of which 88 had a repeated scan 12 months later. Premorbid intelligence (National Adult Reading Test) and current intelligence [Addenbrooke's Cognitive Exam-Revised (ACE-R)] were measured at 1, 12, and 36 months after the stroke. ANCOVA analyses adjusting for baseline cognition/premorbid intelligence, vascular risk factors, age, sex and total baseline WMH volume found that the recent small subcortical infarcts (RSSI) in the internal/external capsule/lentiform nucleus and centrum semiovale did not predict cognitive scores at 12 and 36 months. However, RSSI location moderated voxel-based associations of WMH change from baseline to 1 year with cognitive scores at 1 and 3 years. WMH increase in the external capsule, intersection between the anterior limb of the internal and external capsules, and optical radiation, was associated with worsening of ACE-R scores 1 and 3 years post-stroke after accounting for the location of the index infarct, age and baseline cognition.

Lacunar Stroke Lesion Extent and Location and White Matter Hyperintensities Evolution 1 Year Post-lacunar Stroke.

Lacunar strokes are a common type of ischemic stroke. They are associated with long-term disability, but the factors affecting the dynamic of the infarcted lesion and the brain imaging features associated with them, reflective of small vessel disease (SVD) severity, are still largely unknown. We investigated whether the distribution, volume and 1-year evolution of white matter hyperintensities (WMH), one of these SVD features, relate to the extent and location of these infarcts, accounting for vascular risk factors. We used imaging and clinical data from all patients [n = 118, mean age 64.9 (SD 11.75) years old] who presented to a regional hospital with a lacunar stroke syndrome within the years 2010 and 2013 and consented to participate in a study of stroke mechanisms. All patients had a brain MRI scan at presentation, and 88 had another scan 12 months after. Acute lesions (i.e., recent small subcortical infarcts, RSSI) were identified in 79 patients and lacunes in 77. Number of lacunes was associated with baseline WMH volume (B = 0.370, SE = 0.0939, P = 0.000174). RSSI volume was not associated with baseline WMH volume (B = 3.250, SE = 2.117, P = 0.129), but predicted WMH volume change (B = 2.944, SE = 0.913, P = 0.00184). RSSI location was associated with the spatial distribution of WMH and the pattern of 1-year WMH evolution. Patients with the RSSI in the centrum semiovale (n = 33) had significantly higher baseline volumes of WMH, recent and old infarcts, than patients with the RSSI located elsewhere [median 33.69, IQR (14.37 50.87) ml, 0.001 ≤ P ≤ 0.044]. But patients with the RSSI in the internal/external capsule/lentiform nucleus experienced higher increase of WMH volume after a year [n = 21, median (IQR) from 18 (11.70 31.54) ml to 27.41 (15.84 40.45) ml]. Voxel-wise analyses of WMH distribution in patients grouped per RSSI location revealed group differences increased in the presence of vascular risk factors, especially hypertension and recent or current smoking habit. In our sample of patients presenting to the clinic with lacunar strokes, lacunar strokes extent influenced WMH volume fate; and RSSI location and WMH spatial distribution and dynamics were intertwined, with differential patterns emerging in the presence of vascular risk factors. These results, if confirmed in wider samples, open potential avenues in stroke rehabilitation to be explored further.

A four-dimensional computational model of dynamic contrast-enhanced magnetic resonance imaging measurement of subtle blood-brain barrier leakage.

Dynamic contrast-enhanced MRI (DCE-MRI) is increasingly used to quantify and map the spatial distribution of blood-brain barrier (BBB) leakage in neurodegenerative disease, including cerebral small vessel disease and dementia. However, the subtle nature of leakage and resulting small signal changes make quantification challenging. While simplified one-dimensional simulations have probed the impact of noise, scanner drift, and model assumptions, the impact of spatio-temporal effects such as gross motion, k-space sampling and motion artefacts on parametric leakage maps has been overlooked. Moreover, evidence on which to base the design of imaging protocols is lacking due to practical difficulties and the lack of a reference method. To address these problems, we present an open-source computational model of the DCE-MRI acquisition process for generating four dimensional Digital Reference Objects (DROs), using a high-resolution brain atlas and incorporating realistic patient motion, extra-cerebral signals, noise and k-space sampling. Simulations using the DROs demonstrated a dominant influence of spatio-temporal effects on both the visual appearance of parameter maps and on measured tissue leakage rates. The computational model permits greater understanding of the sensitivity and limitations of subtle BBB leakage measurement and provides a non-invasive means of testing and optimising imaging protocols for future studies.

Examining the Relationship between Semiquantitative Methods Analysing Concentration-Time and Enhancement-Time Curves from Dynamic-Contrast Enhanced Magnetic Resonance Imaging and Cerebrovascular Dysfunction in Small Vessel Disease.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can be used to examine the distribution of an intravenous contrast agent within the brain. Computational methods have been devised to analyse the contrast uptake/washout over time as reflections of cerebrovascular dysfunction. However, there have been few direct comparisons of their relative strengths and weaknesses. In this paper, we compare five semiquantitative methods comprising the slope and area under the enhancement-time curve, the slope and area under the concentration-time curve ( S l o p e C o n and A U C C o n ), and changes in the power spectrum over time. We studied them in cerebrospinal fluid, normal tissues, stroke lesions, and white matter hyperintensities (WMH) using DCE-MRI scans from a cohort of patients with small vessel disease (SVD) who presented mild stroke. The total SVD score was associated with A U C C o n in WMH ( p < 0.05 ), but not with the other four methods. In WMH, we found higher A U C C o n was associated with younger age ( p < 0.001 ) and fewer WMH ( p < 0.001 ), whereas S l o p e C o n increased with younger age ( p > 0.05 ) and WMH burden ( p > 0.05 ). Our results show the potential of different measures extracted from concentration-time curves extracted from the same DCE examination to demonstrate cerebrovascular dysfunction better than those extracted from enhancement-time curves.

The effect of hyperglycemia on neurovascular coupling and cerebrovascular patterning in zebrafish.

Neurovascular coupling (through which local cerebral blood flow changes in response to neural activation are mediated) is impaired in many diseases including diabetes. Current preclinical rodent models of neurovascular coupling rely on invasive surgery and instrumentation, but transgenic zebrafish coupled with advances in imaging techniques allow non-invasive quantification of cerebrovascular anatomy, neural activation, and cerebral vessel haemodynamics. We therefore established a novel non-invasive, non-anaesthetised zebrafish larval model of neurovascular coupling, in which visual stimulus evokes neuronal activation in the optic tectum that is associated with a specific increase in red blood cell speed in tectal blood vessels. We applied this model to the examination of the effect of glucose exposure on cerebrovascular patterning and neurovascular coupling. We found that chronic exposure of zebrafish to glucose impaired tectal blood vessel patterning and neurovascular coupling. The nitric oxide donor sodium nitroprusside rescued all these adverse effects of glucose exposure on cerebrovascular patterning and function. Our results establish the first non-mammalian model of neurovascular coupling, offering the potential to perform more rapid genetic modifications and high-throughput screening than is currently possible using rodents. Furthermore, using this zebrafish model, we reveal a potential strategy to ameliorate the effects of hyperglycemia on cerebrovascular function.

Analysis of dynamic texture and spatial spectral descriptors of dynamic contrast-enhanced brain magnetic resonance images for studying small vessel disease.

Cerebral small vessel disease (SVD) comprises various pathological processes affecting small brain vessels and damaging white and grey matter. In this paper, we propose a framework comprising region of interest sampling, dynamic spectral and texture description, functional principal component analysis, and statistical analysis to study exogenous contrast agent distribution over time in various brain regions in patients with recent mild stroke and SVD features.We compared our results against current semi-quantitative surrogates of dysfunction such as signal enhancement area and slope. Biological sex, stroke lesion type and overall burden of white matter hyperintensities (WMH) were significant predictors of intensity, spectral, and texture features extracted from the ventricular region (p-value < 0.05), explaining between a fifth and a fourth of the data variance (0.20 ≤Adj.R2 ≤ 0.25). We observed that spectral feature reflected more the dysfunction compared to other descriptors since the overall WMH burden explained consistently the power spectra variability in blood vessels, cerebrospinal fluid, deep grey matter and white matter. Our preliminary results show the potential of the framework for the analysis of dynamic contrast-enhanced brain magnetic resonance imaging acquisitions in SVD since significant variation in our metrics was related to the burden of SVD features. Therefore, our proposal may increase sensitivity to detect subtle features of small vessel dysfunction. A public version of the code will be released on our research website.

Predictors of Lesion Cavitation After Recent Small Subcortical Stroke.

Morphologic evolution of recent small subcortical infarcts (RSSI) ranges from lesion disappearance to lacune formation and the reasons for this variability are still poorly understood. We hypothesized that diffusion tensor imaging (DTI) and blood-brain-barrier (BBB) abnormalities early on can predict tissue damage 1 year after an RSSI. We studied prospectively recruited patients with a symptomatic MRI-defined RSSI who underwent baseline and two pre-specified MRI examinations at 1-3-month and 1-year post-stroke. We defined the extent of long-term tissue destruction, termed cavitation index, as the ratio of the 1-year T1-weighted cavity volume to the baseline RSSI volume on FLAIR. We calculated fractional anisotropy and mean diffusivity (MD) of the RSSI and normal-appearing white matter, and BBB leakage in different tissues on dynamic contrast-enhanced MRI. Amongst 60 patients, at 1-year post-stroke, 44 patients showed some degree of RSSI cavitation on FLAIR, increasing to 50 on T2- and 56 on T1-weighted high-resolution scans, with a median cavitation index of 7% (range, 1-36%). Demographic, clinical, and cerebral small vessel disease features were not associated with the cavitation index. While lower baseline MD of the RSSI (rs = - 0.371; p = 0.004) and more contrast leakage into CSF (rs = 0.347; p = 0.007) were associated with the cavitation index in univariable analysis, only BBB leakage in CSF remained independently associated with cavitation (beta = 0.315, p = 0.046). Increased BBB leakage into CSF may indicate worse endothelial dysfunction and increased risk of tissue destruction post RSSI. Although cavitation was common, it only affected a small proportion of the original RSSI.

Identification of the presence of ischaemic stroke lesions by means of texture analysis on brain magnetic resonance images.

BACKGROUND: The differential quantification of brain atrophy, white matter hyperintensities (WMH) and stroke lesions is important in studies of stroke and dementia. However, the presence of stroke lesions is usually overlooked by automatic neuroimage processing methods and the-state-of-the-art deep learning schemes, which lack sufficient annotated data. We explore the use of radiomics in identifying whether a brain magnetic resonance imaging (MRI) scan belongs to an individual that had a stroke or not. MATERIALS AND METHODS: We used 1800 3D sets of MRI data from three prospective studies: one of stroke mechanisms and two of cognitive ageing, evaluated 114 textural features in WMH, cerebrospinal fluid, deep grey and normal-appearing white matter, and attempted to classify the scans using a random forest and support vector machine classifiers with and without feature selection. We evaluated the discriminatory power of each feature independently in each population and corrected the result against Type 1 errors. We also evaluated the influence of clinical parameters in the classification results. RESULTS: Subtypes of ischaemic strokes (i.e. lacunar vs. cortical) cannot be discerned using radiomics, but the presence of a stroke-type lesion can be ascertained with accuracies ranging from 0.7 < AUC < 0.83. Feature selection, tissue type, stroke subtype and MRI sequence did not seem to determine the classification results. From all clinical variables evaluated, age correlated with the proportion of images classified correctly using either different or the same descriptors (Pearson r = 0.31 and 0.39 respectively, p < 0.001). CONCLUSIONS: Texture features in conventionally automatically segmented tissues may help in the identification of the presence of previous stroke lesions on an MRI scan, and should be taken into account in transfer learning strategies of the-state-of-the-art deep learning schemes.

Neurologic Deficits in Patients With Newly Diagnosed Celiac Disease Are Frequent and Linked With Autoimmunity to Transglutaminase 6.

BACKGROUND & AIMS: Celiac disease is an autoimmune disorder induced by ingestion of gluten that affects 1% of the population and is characterized by gastrointestinal symptoms, weight loss, and anemia. We evaluated the presence of neurologic deficits and investigated whether the presence of antibodies to Transglutaminase 6 (TG6) increases the risk of neurologic defects in patients with a new diagnosis of celiac disease. METHODS: We performed a prospective cohort study at a secondary-care gastroenterology center of 100 consecutive patients who received a new diagnosis of celiac disease based on gastroscopy and duodenal biopsy. We collected data on neurologic history, and patients were evaluated in a clinical examination along with magnetic resonance imaging of the brain, magnetic resonance (MR) spectroscopy of the cerebellum, and measurements of antibodies against TG6 in serum samples. The first 52 patients recruited underwent repeat MR spectroscopy at 1 year after a gluten-free diet (GFD). The primary aim was to establish if detection of antibodies against TG6 can be used to identify patients with celiac disease and neurologic dysfunction. RESULTS: Gait instability was reported in 24% of the patients, persisting sensory symptoms in 12%, and frequent headaches in 42%. Gait ataxia was found in 29% of patients, nystagmus in 11%, and distal sensory loss in 10%. Sixty percent of patients had abnormal results from magnetic resonance imaging, 47% had abnormal results from MR spectroscopy of the cerebellum, and 25% had brain white matter lesions beyond that expected for their age group. Antibodies against TG6 were detected in serum samples from 40% of patients-these patients had significant atrophy of subcortical brain regions compared with patients without TG6 autoantibodies. In patients with abnormal results from MR spectroscopy of the cerebellum, those on the GFD had improvements detected in the repeat MR spectroscopy 1 year later. CONCLUSIONS: In a prospective cohort study of patients with a new diagnosis of celiac disease at a gastroenterology clinic, neurologic deficits were common and 40% had circulating antibodies against TG6. We observed a significant reduction in volume of specific brain regions in patients with TG6 autoantibodies, providing evidence for a link between autoimmunity to TG6 and brain atrophy in patients with celiac disease. There is a need for early diagnosis, increased awareness of the neurologic manifestations among clinicians, and reinforcement of adherence to a strict GFD by patients to avoid permanent neurologic disability.

Convolutional Neural Networks for Direct Inference of Pharmacokinetic Parameters: Application to Stroke Dynamic Contrast-Enhanced MRI.

Background and Purpose: The T1-weighted dynamic contrast enhanced (DCE)-MRI is an imaging technique that provides a quantitative measure of pharmacokinetic (PK) parameters characterizing microvasculature of tissues. For the present study, we propose a new machine learning (ML) based approach to directly estimate the PK parameters from the acquired DCE-MRI image-time series that is both more robust and faster than conventional model fitting. Materials and Methods: We specifically utilize deep convolutional neural networks (CNNs) to learn the mapping between the image-time series and corresponding PK parameters. DCE-MRI datasets acquired from 15 patients with clinically evident mild ischaemic stroke were used in the experiments. Training and testing were carried out based on leave-one-patient-out cross- validation. The parameter estimates obtained by the proposed CNN model were compared against the two tracer kinetic models: (1) Patlak model, (2) Extended Tofts model, where the estimation of model parameters is done via voxelwise linear and nonlinear least squares fitting respectively. Results: The trained CNN model is able to yield PK parameters which can better discriminate different brain tissues, including stroke regions. The results also demonstrate that the model generalizes well to new cases even if a subject specific arterial input function (AIF) is not available for the new data. Conclusion: A ML-based model can be used for direct inference of the PK parameters from DCE image series. This method may allow fast and robust parameter inference in population DCE studies. Parameter inference on a 3D volume-time series takes only a few seconds on a GPU machine, which is significantly faster compared to conventional non-linear least squares fitting.

White matter hyperintensity reduction and outcomes after minor stroke.

OBJECTIVE: To assess factors associated with white matter hyperintensity (WMH) change in a large cohort after observing obvious WMH shrinkage 1 year after minor stroke in several participants in a longitudinal study. METHODS: We recruited participants with minor ischemic stroke and performed clinical assessments and brain MRI. At 1 year, we assessed recurrent cerebrovascular events and dependency and repeated the MRI. We assessed change in WMH volume from baseline to 1 year (normalized to percent intracranial volume [ICV]) and associations with baseline variables, clinical outcomes, and imaging parameters using multivariable analysis of covariance, model of changes, and multinomial logistic regression. RESULTS: Among 190 participants (mean age 65.3 years, range 34.3-96.9 years, 112 [59%] male), WMH decreased in 71 participants by 1 year. At baseline, participants whose WMH decreased had similar WMH volumes but higher blood pressure (p = 0.0064) compared with participants whose WMH increased. At 1 year, participants with WMH decrease (expressed as percent ICV) had larger reductions in blood pressure (ß = 0.0053, 95% confidence interval [CI] 0.00099-0.0097 fewer WMH per 1-mm Hg decrease, p = 0.017) and in mean diffusivity in normal-appearing white matter (ß = 0.075, 95% CI 0.0025-0.15 fewer WMH per 1-unit mean diffusivity decrease, p = 0.043) than participants with WMH increase; those with WMH increase experienced more recurrent cerebrovascular events (32%, vs 16% with WMH decrease, ß = 0.27, 95% CI 0.047-0.50 more WMH per event, p = 0.018). CONCLUSIONS: Some WMH may regress after minor stroke, with potentially better clinical and brain tissue outcomes. The role of risk factor control requires verification. Interstitial fluid alterations may account for some WMH reversibility, offering potential intervention targets.

Metric to quantify white matter damage on brain magnetic resonance images.

PURPOSE: Quantitative assessment of white matter hyperintensities (WMH) on structural Magnetic Resonance Imaging (MRI) is challenging. It is important to harmonise results from different software tools considering not only the volume but also the signal intensity. Here we propose and evaluate a metric of white matter (WM) damage that addresses this need. METHODS: We obtained WMH and normal-appearing white matter (NAWM) volumes from brain structural MRI from community dwelling older individuals and stroke patients enrolled in three different studies, using two automatic methods followed by manual editing by two to four observers blind to each other. We calculated the average intensity values on brain structural fluid-attenuation inversion recovery (FLAIR) MRI for the NAWM and WMH. The white matter damage metric is calculated as the proportion of WMH in brain tissue weighted by the relative image contrast of the WMH-to-NAWM. The new metric was evaluated using tissue microstructure parameters and visual ratings of small vessel disease burden and WMH: Fazekas score for WMH burden and Prins scale for WMH change. RESULTS: The correlation between the WM damage metric and the visual rating scores (Spearman ρ > =0.74, p < 0.0001) was slightly stronger than between the latter and WMH volumes (Spearman ρ > =0.72, p < 0.0001). The repeatability of the WM damage metric was better than WM volume (average median difference between measurements 3.26% (IQR 2.76%) and 5.88% (IQR 5.32%) respectively). The follow-up WM damage was highly related to total Prins score even when adjusted for baseline WM damage (ANCOVA, p < 0.0001), which was not always the case for WMH volume, as total Prins was highly associated with the change in the intense WMH volume (p = 0.0079, increase of 4.42 ml per unit change in total Prins, 95%CI [1.17 7.67]), but not with the change in less-intense, subtle WMH, which determined the volumetric change. CONCLUSION: The new metric is practical and simple to calculate. It is robust to variations in image processing methods and scanning protocols, and sensitive to subtle and severe white matter damage.

Application of Texture Analysis to Study Small Vessel Disease and Blood-Brain Barrier Integrity.

OBJECTIVES: We evaluate the alternative use of texture analysis for evaluating the role of blood-brain barrier (BBB) in small vessel disease (SVD). METHODS: We used brain magnetic resonance imaging from 204 stroke patients, acquired before and 20 min after intravenous gadolinium administration. We segmented tissues, white matter hyperintensities (WMH) and applied validated visual scores. We measured textural features in all tissues pre- and post-contrast and used ANCOVA to evaluate the effect of SVD indicators on the pre-/post-contrast change, Kruskal-Wallis for significance between patient groups and linear mixed models for pre-/post-contrast variations in cerebrospinal fluid (CSF) with Fazekas scores. RESULTS: Textural "homogeneity" increase in normal tissues with higher presence of SVD indicators was consistently more overt than in abnormal tissues. Textural "homogeneity" increased with age, basal ganglia perivascular spaces scores (p < 0.01) and SVD scores (p < 0.05) and was significantly higher in hypertensive patients (p < 0.002) and lacunar stroke (p = 0.04). Hypertension (74% patients), WMH load (median = 1.5 ± 1.6% of intracranial volume), and age (mean = 65.6 years, SD = 11.3) predicted the pre/post-contrast change in normal white matter, WMH, and index stroke lesion. CSF signal increased with increasing SVD post-contrast. CONCLUSION: A consistent general pattern of increasing textural "homogeneity" with increasing SVD and post-contrast change in CSF with increasing WMH suggest that texture analysis may be useful for the study of BBB integrity.

Reliability of an automatic classifier for brain enlarged perivascular spaces burden and comparison with human performance.

In the brain, enlarged perivascular spaces (PVS) relate to cerebral small vessel disease (SVD), poor cognition, inflammation and hypertension. We propose a fully automatic scheme that uses a support vector machine (SVM) to classify the burden of PVS in the basal ganglia (BG) region as low or high. We assess the performance of three different types of descriptors extracted from the BG region in T2-weighted MRI images: (i) statistics obtained from Wavelet transform's coefficients, (ii) local binary patterns and (iii) bag of visual words (BoW) based descriptors characterizing local keypoints obtained from a dense grid with the scale-invariant feature transform (SIFT) characteristics. When the latter were used, the SVM classifier achieved the best accuracy (81.16%). The output from the classifier using the BoW descriptors was compared with visual ratings done by an experienced neuroradiologist (Observer 1) and by a trained image analyst (Observer 2). The agreement and cross-correlation between the classifier and Observer 2 (κ = 0.67 (0.58-0.76)) were slightly higher than between the classifier and Observer 1 (κ = 0.62 (0.53-0.72)) and comparable between both the observers (κ = 0.68 (0.61-0.75)). Finally, three logistic regression models using clinical variables as independent variable and each of the PVS ratings as dependent variable were built to assess how clinically meaningful were the predictions of the classifier. The goodness-of-fit of the model for the classifier was good (area under the curve (AUC) values: 0.93 (model 1), 0.90 (model 2) and 0.92 (model 3)) and slightly better (i.e. AUC values: 0.02 units higher) than that of the model for Observer 2. These results suggest that, although it can be improved, an automatic classifier to assess PVS burden from brain MRI can provide clinically meaningful results close to those from a trained observer.

Sample size considerations for trials using cerebral white matter hyperintensity progression as an intermediate outcome at 1 year after mild stroke: results of a prospective cohort study.

BACKGROUND: White matter hyperintensities (WMHs) are commonly seen on in brain imaging and are associated with stroke and cognitive decline. Therefore, they may provide a relevant intermediate outcome in clinical trials. WMH can be measured as a volume or visually on the Fazekas scale. We investigated predictors of WMH progression and design of efficient studies using WMH volume and Fazekas score as an intermediate outcome. METHODS: We prospectively recruited 264 patients with mild ischaemic stroke and measured WMH volume, Fazekas score, age and cardiovascular risk factors at baseline and 1 year. We modelled predictors of WMH burden at 1 year and used the results in sample size calculations for hypothetical randomised controlled trials with different analysis plans and lengths of follow-up. RESULTS: Follow-up WMH volume was predicted by baseline WMH: a 0.73-ml (95% CI 0.65-0.80, p < 0.0001) increase per 1-ml baseline volume increment, and a 2.93-ml increase (95% CI 1.76-4.10, p < 0.0001) per point on the Fazekas scale. Using a mean difference of 1 ml in WMH volume between treatment groups, 80% power and 5% alpha, adjusting for all predictors and 2-year follow-up produced the smallest sample size (n = 642). Other study designs produced samples sizes from 2054 to 21,270. Sample size calculations using Fazekas score as an outcome with the same power and alpha, as well as an OR corresponding to a 1-ml difference, were sensitive to assumptions and ranged from 2504 to 18,886. CONCLUSIONS: Baseline WMH volume and Fazekas score predicted follow-up WMH volume. Study size was smallest using volumes and longer-term follow-up, but this must be balanced against resources required to measure volumes versus Fazekas scores, bias due to dropout and scanner drift. Samples sizes based on Fazekas scores may be best estimated with simulation studies.