White matter lesion extension to automatic brain tissue segmentation on MRI. Neuroimage. 2009;45:1151-1161. [PMID: 19344687 DOI: 10.1016/j.neuroimage.2009.01.011]

Mediation of the Association Between APOE ε4 Genotype, Cognition, and Dementia by Neuropathology Imaging Markers in the Rotterdam Study

BACKGROUND AND OBJECTIVES

Insight into APOE-related pathways is important to unravel pathophysiology and identify therapeutic targets against late-life cognitive decline. We aimed to estimate mediators of APOE ε4 on cognition and dementia through different disease markers on structural in vivo brain imaging.

METHODS

All participants from the population-based Rotterdam Study who underwent brain MRI between 2005 and 2009 were included. Cognition was assessed cross-sectionally during center visits, and participants were followed up for incident dementia until January 1, 2020. Imaging markers included hippocampal volume (HV), volume of white matter hyperintensities (WMHs), Alzheimer disease-specific regional cortical thickness, and presence of ≥2 cerebral microbleeds. We performed causal mediation analyses to decompose the total effect of APOE ε4 carriership on cognition and dementia into natural direct and indirect effects and corresponding percentage mediated. We adjusted models for potential confounders.

RESULTS

Among 5,510 participants (mean age at time of MRI scan: 65.0 [±10.9] years, 55.0% women), 349 developed dementia, of whom 148 were ε4 carriers. Carriers of ε4 had slightly lower Z-scores for global cognition (β = -0.02 [-0.07 to 0.02], age-related cognitive decline = 4.4 months), with 7% (β = -0.00 [0.00-0.00]) of this association mediated by HV and 4% (β = -0.00 [-0.01 to 0.00]) by cortical thickness. In total, an estimated 25% of the effect of ε4 on cognition was mediated by microbleeds (p value = 0.24, [β = -0.00 {-0.01 to 0.00}]) and 12% by WMHs (p value = 0.44, [β = -0.00 {-0.01 to 0.00}]). In multiple mediator analyses, WMHs and microbleeds together accounted for 27% of the mediated effect of APOE ε4 on cognition (p value = 0.48). Carriers of ε4 had higher risk of incident dementia (HR 2.35 [95% CI 2.06-2.65]). For dementia, there was little to no evidence of mediation by either HV (3%, p value = 0.09, OR = 1.01 [1.00-1.03]) or regional cortical thickness (0%, p value = 0.79, OR = 1.00 [0.99-1.02]). In total, 1% of the effect of ε4 on dementia was mediated by WMHs (p value 0.29, OR = 1.00 [1.00-1.02]) and 5% by microbleeds (p value = 0.06), OR = 1.03 (1.00-1.07). In multiple mediator analyses, all 4 imaging markers together explained 6% of the mediated effect on incident dementia (p value = 0.04).

DISCUSSION

In this population-based cohort study, we found that an estimated one-fourth of the effect of APOE ε4 on cognition is mediated by structural brain imaging markers, driven mainly by cerebral microbleeds. For dementia, mediation by these markers was limited.

Dynamic PET imaging in patients with unilateral carotid occlusion shows lateralized cerebral hypoperfusion, but no amyloid binding

BackgroundCarotid occlusive disease is a risk factor for cognitive decline. A possible underlying etiology is that hemodynamic impairment results in decreased cerebral perfusion, exacerbated amyloid-β accumulation (Aβ) and poorer cognitive performance.ObjectiveWe aimed to determine whether patients with unilateral internal carotid artery (ICA) occlusion have less cerebral perfusion and more Aβ in the ipsilateral than in the contralateral hemisphere, and whether perfusion and Aβ are associated with cognitive functioning.MethodsWe included 20 patients (age 67.2 ± 7.0 years, 8 females, MMSE 29 [27-29]) with unilateral ICA occlusion, which underwent neuropsychological assessment and dynamic 18F-Florbetaben positron emission tomography (PET). Global and regional relative perfusion (R1) and binding potential (BPND) were obtained from the PET-images using a simplified reference tissue model. We performed Wilcoxon signed-rank tests to examine differences between hemispheres within subjects and linear regression to investigate associations with cognitive functioning.ResultsMedian global R1 was 0.911 (0.883-0.950) and global BPND was 0.172 (0.129-0.187). R1 was lower in the hemisphere ipsilateral to the ICA occlusion than in the contralateral hemisphere (0.899 [0.876-0.921] versus 0.935 [0.889-0.970]). BPND did not differ significantly between hemispheres (ipsilateral 0.172 [0.124-0.181] versus contralateral 0.168 [0.137-0.191]). Neither cerebral perfusion nor Aβ burden were associated with cognitive functioning.ConclusionsPatients with unilateral ICA occlusion did not have more Aβ in the ipsilateral hemisphere than in the contralateral hemisphere despite ipsilateral hypoperfusion. Perfusion and Aβ were unrelated to cognitive functioning. This indicates that cognitive impairment in patients with ICA occlusion is not due to exacerbated Aβ accumulation.

Generalizability of trial criteria on amyloid-lowering therapy against Alzheimer's disease to individuals with mild cognitive impairment or early Alzheimer's disease in the general population

Treatment with anti-amyloid-β monoclonal antibodies slowed cognitive decline in recent RCTs in patients with mild cognitive impairment (MCI) and early dementia due to Alzheimer's disease (AD). However, stringent trial eligibility criteria may affect generalisability to clinical practice. We extracted eligibility criteria for trials of aducanumab, lecanemab and donanemab, and applied these to participants with MCI and early clinical AD dementia from the population-based Rotterdam Study. Participants underwent questionnaires, genotyping, brain-MRI, cognitive testing, and cardiovascular assessment. We determined amyloid status using a validated prediction model based on age and APOE-genotype. Of 968 participants (mean age: 75 years, 56% women), 779 had MCI and 189 dementia. Across trials, around 40% of participants would be ineligible because of predicted amyloid negativity. At least one clinical exclusion criterion was present in 76.3% of participants for aducanumab, 75.8% for lecanemab, and 59.8% for donanemab. Common criteria were cardiovascular disease (35.2%), anticoagulant (31.2%), psychotropic or immunological medication use (20.4%), anxiety or depression (15.9%), or lack of social support (15.6%). One-third were ineligible based on brain-MRI findings alone, similar across trials and predominantly due to cerebral small-vessel disease. Combining amyloid, clinical, and imaging criteria, eligibility ranged from 9% (95% CI:7.0-11.1) for aducanumab, 8% (6.2-9.9) lecanemab to 15% (12.4-17.5) for donanemab. Findings from recent RCTs reporting protective effects of monoclonal antibodies against amyloid-β are applicable to less than 15% of community-dwelling individuals with MCI or early AD. These findings underline that evidence for drug efficacy and safety is lacking for the vast majority of patients with MCI/AD in routine clinical practice.

Neurofind: using deep learning to make individualised inferences in brain-based disorders

Within precision psychiatry, there is a growing interest in normative models given their ability to parse heterogeneity. While they are intuitive and informative, the technical expertise and resources required to develop normative models may not be accessible to most researchers. Here we present Neurofind, a new freely available tool that bridges this gap by wrapping sound and previously tested methods on data harmonisation and advanced normative models into a web-based platform that requires minimal input from the user. We explain how Neurofind was developed, how to use the Neurofind website in four simple steps ( www.neurofind.ai ), and provide exemplar applications. Neurofind takes as input structural MRI images and outputs two main metrics derived from independent normative models: (1) Outlier Index Score, a deviation score from the normative brain morphology, and (2) Brain Age, the predicted age based on an individual's brain morphometry. The tool was trained on 3362 images of healthy controls aged 20-80 from publicly available datasets. The volume of 101 cortical and subcortical regions was extracted and modelled with an adversarial autoencoder for the Outlier index model and a support vector regression for the Brain age model. To illustrate potential applications, we applied Neurofind to 364 images from three independent datasets of patients diagnosed with Alzheimer's disease and schizophrenia. In Alzheimer's disease, 55.2% of patients had very extreme Outlier Index Scores, mostly driven by larger deviations in temporal-limbic structures and ventricles. Patients were also homogeneous in how they deviated from the norm. Conversely, only 30.1% of schizophrenia patients were extreme outliers, due to deviations in the hippocampus and pallidum, and patients tended to be more heterogeneous than controls. Both groups showed signs of accelerated brain ageing.

Validation of a novel neuroimaging signature for dementia and clinical Alzheimer's disease in the population-based Rotterdam study

BACKGROUND

A novel neuroimaging signature of regional cortical thickness on brain MRI recently showed high potential for Alzheimer's disease and related dementias (ADRD) risk stratification in the community. How these findings translate to other populations, remains undetermined.

OBJECTIVE

We aimed to replicate this novel ADRD neuroimaging marker in the population-based Rotterdam Study.

METHODS

We included all participants from the population-based Rotterdam Study with brain-MRI between 2005-2016, and derived the signature using FreeSurfer. We computed hazard ratios and C-statistics for 10-year dementia risk, and betas for cross-sectional associations with cognition, comparing the novel signature to hippocampal volume, mean cortical thickness, and another cortical thickness signature (Dickerson's).

RESULTS

Of 3249 participants (mean age 71.3 ± 8.0 years), 294 developed dementia (74.8% clinical AD) during a mean follow-up of 8.1 years. The novel ADRD signature had similar magnitude of associations as Dickerson's signature and cortical thickness for AD dementia (HR per 1-SD increase 0.87;0.78-0.96), but performed worse than all markers for all-cause dementia. Of the four neuroimaging markers, hippocampal volume showed the strongest associations with both risk of all-cause dementia and clinical AD dementia. The ADRD had the weakest association with general cognitive function (β per 1-SD increase 0.04;0.02-0.06), and executive function (β per 1-SD increase 0.02;0.00-0.04), followed by cortical thickness and Dickerson's, and hippocampal volume showed the strongest associations.

CONCLUSIONS

In this community-based study, the novel cortical thickness signature did not outperform hippocampal volume for dementia risk stratification. The importance of replication studies underlines the value of the current study. Replicating research findings is essential to establish robust biomarkers for dementia risk prediction.

Changes in Cerebral Hemodynamics and Progression of Subclinical Vascular Brain Disease: A Population-Based Cohort Study

BACKGROUND

Cerebral hypoperfusion is associated with vascular brain injury and neurodegeneration, but their longitudinal relationship is largely unknown, especially in healthy older adults.

METHODS

We investigated the longitudinal relationship between cerebral hemodynamics and subclinical vascular brain disease in community-dwelling older adults without stroke or dementia at baseline. Participants underwent brain magnetic resonance imaging scans every 3 to 4 years between 2005 and 2016. Cerebral blood flow (CBF) was measured through 2-dimensional phase-contrast magnetic resonance imaging; the cerebrovascular resistance index (CVRi) was defined as the ratio of mean arterial blood pressure to total CBF. Simultaneous progression in subclinical brain disease was evaluated through repeated magnetic resonance imaging assessment of white matter hyperintensities (WMH), cerebral microbleeds, lacune, and brain atrophy. The longitudinal relationship was estimated using generalized estimating equations, with adjustment for age, sex, smoking habits, body mass index, systolic blood pressure (for CBF measures), lipid level, history of diabetes and cardiovascular disease, and the baseline burden of magnetic resonance imaging markers.

RESULTS

Among 3623 older adults (mean age, 61.4±9.3 years; 54.6% women), large decreases and increases in CBF and increases in CVRi over time were associated with white matter hyperintensity progression. The risk ratios for white matter hyperintensity progression were 1.36 (95% CI, 1.19-1.55) for large decreases in total CBF (lowest quartile), 1.02 (95% CI, 0.91-1.14) for moderate decreases (second quartile), and 1.28 (95% CI, 1.14-1.45) for large increases (highest quartile), compared with stable CBF (third quartile). The corresponding risk ratios for changes in CVRi were 1.13 (95% CI, 1.00-1.30), 1.25 (95% CI, 1.09-1.43), and 1.33 (95% CI, 1.16-1.52) for the second to fourth (versus lowest) quartiles, respectively, showing a dose-response relationship. The changes in CBF also demonstrate a similar U-shaped association with the progression of brain atrophy and incident microbleeds, whereas increases in CVRi were associated with lower microbleed risk.

CONCLUSIONS

Longitudinal changes in CBF and CVRi may capture distinct pathophysiologies linking cerebral hemodynamics to subclinical brain disease, extending beyond single-time point measurements.

Hearing loss and its relation to longitudinal changes in white matter microstructure in older adults: The Rotterdam Study

Hearing loss is considered a potentially modifiable risk factor for dementia. The sensory deprivation theory postulates that hearing loss adversely affects cognition in older adults through structural brain changes, but longitudinal studies are scarce. To find evidence for a possible detrimental effect of hearing loss on white matter microstructure, we carried out a longitudinal study in the population-based Rotterdam Study. A total of 1877 participants with a median age at baseline of 56.4 years (IQR: [52.2-60.0]) underwent audiometry and had longitudinal diffusion imaging data available with a mean follow-up of 4.0 years. A lower level of hearing acuity was associated with worse white matter microstructure in the left uncinate fasciculus and superior longitudinal fasciculus at baseline. Poorer hearing acuity was also associated with faster microstructural deterioration over time in the left superior longitudinal fasciculus. The strongest effects were observed for low-frequency hearing thresholds, while the high-frequency thresholds showed the weakest associations. These results suggest that hearing loss may contribute to the age-related decline in brain structure, consistent with the sensory deprivation theory.

Sex differences in cognitive functioning in patients with heart failure

BACKGROUND

Cognitive impairment is common in patients with heart failure (HF) and impacts patients' life. Sex differences in HF-characteristics are well-established. We hypothesized that women and men with HF also differ in cognitive functioning and that this may be related to sex differences in HF-characteristics and vascular brain injury.

METHODS

In the Heart-Brain Connection Study, 162 clinically stable HF patients (mean age 69.7 ± 10.0, 33 % women) underwent neuropsychological assessments and brain-MRI. Test results were standardized into z-scores for memory, language, attention/speed, executive functioning, and global cognition. Using linear models adjusted for age and education, we calculated sex differences (women-to-men: W-M∆) in cognitive functioning and examined effects of HF- and vascular brain injury-characteristics on these differences.

RESULTS

Men more often had an ischemic cause of HF and lower NYHA-classes, whereas women more often had preserved left ventricular ejection fractions (LVEF). Women had a higher volume of white matter hyperintensities (WMHs) whereas non-lacunar infarcts and microbleeds were more prevalent in men. Women performed better on global cognition than men (W-M∆ in z-score 0.20, 95 %CI 0.03-0.37), predominantly on memory (0.40, 0.02-0.78). These differences were associated with ischemic HF-etiology, as adjustment attenuated these sex differences. After adjustment for non-lacunar infarcts, global cognition difference persisted, but the difference in memory functioning attenuated. Adjustments for NYHA-class, LVEF, WMHs, and microbleeds did not change the results.

CONCLUSION

Women and men with HF differ in cognitive functioning, predominantly in memory functioning, these differences were related to some sex differences in HF-characteristics and vascular brain injury, but not to all.

The blood metabolome of cognitive function and brain health in middle-aged adults - influences of genes, gut microbiome, and exposome

Increasing evidence suggests the involvement of metabolic alterations in neurological disorders, including Alzheimer's disease (AD), and highlights the significance of the peripheral metabolome, influenced by genetic factors and modifiable environmental exposures, for brain health. In this study, we examined 1,387 metabolites in plasma samples from 1,082 dementia-free middle-aged participants of the population-based Rotterdam Study. We assessed the relation of metabolites with general cognition (G-factor) and magnetic resonance imaging (MRI) markers using linear regression and estimated the variance of these metabolites explained by genes, gut microbiome, lifestyle factors, common clinical comorbidities, and medication using gradient boosting decision tree analysis. Twenty-one metabolites and one metabolite were significantly associated with total brain volume and total white matter lesions, respectively. Fourteen metabolites showed significant associations with G-factor, with ergothioneine exhibiting the largest effect (adjusted mean difference = 0.122, P = 4.65×10-7). Associations for nine of the 14 metabolites were replicated in an independent, older cohort. The metabolite signature of incident AD in the replication cohort resembled that of cognition in the discovery cohort, emphasizing the potential relevance of the identified metabolites to disease pathogenesis. Lifestyle, clinical variables, and medication were most important in determining these metabolites' blood levels, with lifestyle, explaining up to 28.6% of the variance. Smoking was associated with ten metabolites linked to G-factor, while diabetes and antidiabetic medication were associated with 13 metabolites linked to MRI markers, including N-lactoyltyrosine. Antacid medication strongly affected ergothioneine levels. Mediation analysis revealed that lower ergothioneine levels may partially mediate negative effects of antacids on cognition (31.5%). Gut microbial factors were more important for the blood levels of metabolites that were more strongly associated with cognition and incident AD in the older replication cohort (beta-cryptoxanthin, imidazole propionate), suggesting they may be involved later in the disease process. The detailed results on how multiple modifiable factors affect blood levels of cognition- and brain imaging-related metabolites in dementia-free participants may help identify new AD prevention strategies.

Cluster-Based White Matter Signatures and the Risk of Dementia, Stroke, and Mortality in Community-Dwelling Adults

BACKGROUND AND OBJECTIVES

Markers of white matter (WM) injury on brain MRI are important indicators of brain health. Different patterns of WM atrophy, WM hyperintensities (WMHs), and microstructural integrity could reflect distinct pathologies and disease risks, but large-scale imaging studies investigating WM signatures are lacking. This study aims to identify distinct WM signatures using brain MRI in community-dwelling adults, determine underlying risk factor profiles, and assess risks of dementia, stroke, and mortality associated with each signature.

METHODS

Between 2005 and 2016, we measured WMH volume, WM volume, fractional anisotropy (FA), and mean diffusivity (MD) using automated pipelines on structural and diffusion MRI in community-dwelling adults aged older than 45 years of the Rotterdam study. Continuous surveillance was conducted for dementia, stroke, and mortality. We applied hierarchical clustering to identify separate WM injury clusters and Cox proportional hazard models to determine their risk of dementia, stroke, and mortality.

RESULTS

We included 5,279 participants (mean age 65.0 years, 56.0% women) and identified 4 distinct data-driven WM signatures: (1) above-average microstructural integrity and little WM atrophy and WMH; (2) above-average microstructural integrity and little WMH, but substantial WM atrophy; (3) poor microstructural integrity and substantial WMH, but little WM atrophy; and (4) poor microstructural integrity with substantial WMH and WM atrophy. Prevalence of cardiovascular risk factors, lacunes, and cerebral microbleeds was higher in clusters 3 and 4 than in clusters 1 and 2. During a median 10.7 years of follow-up, 291 participants developed dementia, 220 had a stroke, and 910 died. Compared with cluster 1, dementia risk was increased for all clusters, notably cluster 3 (hazard ratio [HR] 3.06, 95% CI 2.12-4.42), followed by cluster 4 (HR 2.31, 95% CI 1.58-3.37) and cluster 2 (HR 1.67, 95% CI 1.17-2.38). Compared with cluster 1, risk of stroke was higher only for clusters 3 (HR 1.55, 95% CI 1.02-2.37) and 4 (HR 1.94, 95% CI 1.30-2.89), whereas mortality risk was increased in all clusters (cluster 2: HR 1.27, 95% CI 1.06-1.53, cluster 3: HR 1.65, 95% CI 1.35-2.03, cluster 4: HR 1.76, 95% CI 1.44-2.15), compared with cluster 1. Models including clusters instead of an individual imaging marker showed a superior goodness of fit for dementia and mortality, but not for stroke.

DISCUSSION

Clustering can derive WM signatures that are differentially associated with dementia, stroke, and mortality risk. Future research should incorporate spatial information of imaging markers.

Association of ambient air pollution with cognitive functioning and markers of structural brain damage: The Maastricht study

INTRODUCTION

Given the absence of curative interventions and the rising global incidence of dementia, research is increasingly focusing on lifestyle factors for prevention. However, identifying shared environmental risk for dementia, next to individual factors, is crucial for optimal risk reduction strategies. Therefore, in the present study we investigated the association between air pollution, cognitive functioning, and markers of structural brain damage.

METHODS

We used cross-sectional data from 4,002 participants of The Maastricht Study on volumetric markers of brain integrity (white and grey matter volume, cerebrospinal fluid volume, white matter hyperintensities volume, presence of cerebral small vessel disease) and cognitive functioning (memory, executive functioning and attention, processing speed, overall cognition). Individuals were matched by postal code of residence to nationwide data on air pollution exposure (particulate matter < 2.5 μm (PM2.5), particulate matter <10 μm (PM10), nitrogen dioxide (NO2), soot). Potentia linear and non-linear associations were investigated with linear, logistic, and restricted cubic splines regression. All analyses were adjusted for demographic characteristics and a compound score of modifiable dementia risk and protective factors.

RESULTS

Exposure to air pollutants was not related to cognitive functioning and most brain markers. We found curvilinear relationships between high PM2.5 exposures and grey matter and cerebrospinal fluid volume. Participants in the low and high range of exposure had lower grey matter volume. Higher cerebrospinal fluid volumes were only associated with high range of exposure, independent of demographic and individual modifiable dementia risk factors. After additional post hoc analyses, controlling for urbanicity, the associations for grey matter volume became non-significant. In men only, higher exposure to all air pollutants was associated with lower white matter volumes. No significant associations with white matter hyperintensities volume or cerebral small vessel disease were observed.

DISCUSSION

Our findings suggest that higher PM2.5 exposure is associated with more brain atrophy.

The neurovascular unit and its correlation with cognitive performance in patients with cerebral small vessel disease: a canonical correlation analysis approach

Growing evidence indicates an important role of neurovascular unit (NVU) dysfunction in the pathophysiology of cerebral small vessel disease (cSVD). Individually measurable functions of the NVU have been correlated with cognitive function, but a combined analysis is lacking. We aimed to perform a unified analysis of NVU function and its relation with cognitive performance. The relationship between NVU function in the white matter and cognitive performance (both latent variables composed of multiple measurable variables) was investigated in 73 patients with cSVD (mean age 70 ± 10 years, 41% women) using canonical correlation analysis. MRI-based NVU function measures included (1) the intravoxel incoherent motion derived perfusion volume fraction (f) and microvascular diffusivity (D*), reflecting cerebral microvascular flow; (2) the IVIM derived intermediate volume fraction (fint), indicative of the perivascular clearance system; and (3) the dynamic contrast-enhanced MRI derived blood-brain barrier (BBB) leakage rate (Ki) and leakage volume fraction (VL), reflecting BBB integrity. Cognitive performance was composed of 13 cognitive test scores. Canonical correlation analysis revealed a strong correlation between the latent variables NVU function and cognitive performance (r 0.73; p = 0.02). For the NVU, the dominating variables were D*, fint, and Ki. Cognitive performance was driven by multiple cognitive tests comprising different cognitive domains. The functionality of the NVU is correlated with cognitive performance in cSVD. Instead of focusing on individual pathophysiological mechanisms, future studies should target NVU dysfunction as a whole to acquire a coherent understanding of the complex disease mechanisms that occur in the NVU in cSVD.Trial registration: NTR3786 (Dutch Trial Register).

The associations of socioeconomic position with structural brain damage and connectivity and cognitive functioning: The Maastricht Study

BACKGROUND

Socioeconomic inequalities in cognitive impairment may partly act through structural brain damage and reduced connectivity. This study investigated the extent to which the association of early-life socioeconomic position (SEP) with later-life cognitive functioning is mediated by later-life SEP, and whether the associations of SEP with later-life cognitive functioning can be explained by structural brain damage and connectivity.

METHODS

We used cross-sectional data from the Dutch population-based Maastricht Study (n = 4,839; mean age 59.2 ± 8.7 years, 49.8% women). Early-life SEP was assessed by self-reported poverty during childhood and parental education. Later-life SEP included education, occupation, and current household income. Participants underwent cognitive testing and 3-T magnetic resonance imaging to measure volumes of white matter hyperintensities, grey matter, white matter, cerebrospinal fluid, and structural connectivity. Multiple linear regression analyses tested the associations between SEP, markers of structural brain damage and connectivity, and cognitive functioning. Mediation was tested using structural equation modeling.

RESULTS

Although there were direct associations between both indicators of SEP and later-life cognitive functioning, a large part of the association between early-life SEP and later-life cognitive functioning was explained by later-life SEP (72.2%). The extent to which structural brain damage or connectivity acted as mediators between SEP and cognitive functioning was small (up to 5.9%).

CONCLUSIONS

We observed substantial SEP differences in later-life cognitive functioning. Associations of structural brain damage and connectivity with cognitive functioning were relatively small, and only marginally explained the SEP gradients in cognitive functioning.

Orthostatic hypotension, cognition and structural brain imaging in hemodynamically impaired patients

BACKGROUND

Orthostatic hypotension (OH) is associated with an increased risk of dementia, potentially attributable to cerebral hypoperfusion. We investigated which patterns and characteristics of OH are related to cognition or to potentially underlying structural brain injury in hemodynamically impaired patients and healthy reference participants.

METHODS

Participants with carotid occlusive disease or heart failure, and reference participants from the Heart-Brain Connection Study underwent OH measurements, neuropsychological assessment and brain MRI. We analyzed the association between OH, global cognitive functioning, white matter hyperintensity (WMH) volume and brain parenchymal fraction with linear regression. We stratified by participant group, severity and duration of OH, chronotropic incompetence and presence of orthostatic symptoms.

RESULTS

Of 337 participants (mean age 67.3 ± 8.8 years, 118 (35.0%) women), 113 (33.5%) had OH. Overall, presence of OH was not associated with cognitive functioning (β: -0.12 [-0.24-0.00]), but we did observe worse cognitive functioning in those with severe OH (≥ 30/15 mmHg; β: -0.18 [-0.34 to -0.02]) and clinically manifest OH (β: -0.30 [-0.52 to -0.08]). These associations did not differ significantly by OH duration or chronotropic incompetence, and were similar between patient groups and reference participants. Similarly, both severe OH and clinically manifest OH were associated with a lower brain parenchymal fraction, and severe OH also with a somewhat higher WMH volume.

CONCLUSIONS

Severe OH and clinically manifest OH are associated with worse cognitive functioning. This supports the notion that specific patterns and characteristics of OH determine its impact on brain health.

Plasma trimethylamine N-oxide (TMAO): associations with cognition, neuroimaging, and dementia

BACKGROUND

The gut-derived metabolite Trimethylamine N-oxide (TMAO) and its precursors - betaine, carnitine, choline, and deoxycarnitine - have been associated with an increased risk of cardiovascular disease, but their relation to cognition, neuroimaging markers, and dementia remains uncertain.

METHODS

In the population-based Rotterdam Study, we used multivariable regression models to study the associations between plasma TMAO, its precursors, and cognition in 3,143 participants. Subsequently, we examined their link to structural brain MRI markers in 2,047 participants, with a partial validation in the Leiden Longevity Study (n = 318). Among 2,517 participants, we assessed the risk of incident dementia using multivariable Cox proportional hazard models. Following this, we stratified the longitudinal associations by medication use and sex, after which we conducted a sensitivity analysis for individuals with impaired renal function.

RESULTS

Overall, plasma TMAO was not associated with cognition, neuroimaging markers or incident dementia. Instead, higher plasma choline was significantly associated with poor cognition (adjusted mean difference: -0.170 [95% confidence interval (CI) -0.297;-0.043]), brain atrophy and more markers of cerebral small vessel disease, such as white matter hyperintensity volume (0.237 [95% CI: 0.076;0.397]). By contrast, higher carnitine concurred with lower white matter hyperintensity volume (-0.177 [95% CI: -0.343;-0.010]). Only among individuals with impaired renal function, TMAO appeared to increase risk of dementia (hazard ratio (HR): 1.73 [95% CI: 1.16;2.60]). No notable differences were observed in stratified analyses.

CONCLUSIONS

Plasma choline, as opposed to TMAO, was found to be associated with cognitive decline, brain atrophy, and markers of cerebral small vessel disease. These findings illustrate the complexity of relationships between TMAO and its precursors, and emphasize the need for concurrent study to elucidate gut-brain mechanisms.

Brain structure and connectivity mediate the association between lifestyle and cognition: The Maastricht Study

Life-course exposure to risk and protective factors impacts brain macro- and micro-structure, which in turn affects cognition. The concept of brain-age gap assesses brain health by comparing an individual's neuroimaging-based predicted age with their calendar age. A higher BAG implies accelerated brain ageing and is expected to be associated with worse cognition. In this study, we comprehensively modelled mutual associations between brain health and lifestyle factors, brain age and cognition in a large, middle-aged population. For this study, cognitive test scores, lifestyle and 3T MRI data for n = 4881 participants [mean age (± SD) = 59.2 (±8.6), 50.1% male] were available from The Maastricht Study, a population-based cohort study with extensive phenotyping. Whole-brain volumes (grey matter, cerebrospinal fluid and white matter hyperintensity), cerebral microbleeds and structural white matter connectivity were calculated. Lifestyle factors were combined into an adapted LIfestyle for BRAin health weighted sum score, with higher score indicating greater dementia risk. Cognition was calculated by averaging z-scores across three cognitive domains (memory, information processing speed and executive function and attention). Brain-age gap was calculated by comparing calendar age to predictions from a neuroimaging-based multivariable regression model. Paths between LIfestyle for BRAin health tertiles, brain-age gap and cognitive function were tested using linear regression and structural equation modelling, adjusting for sociodemographic and clinical confounders. The results show that cerebrospinal fluid, grey matter, white matter hyperintensity and cerebral microbleeds best predicted brain-age gap (R 2 = 0.455, root mean squared error = 6.44). In regression analysis, higher LIfestyle for BRAin health scores (greater dementia risk) were associated with higher brain-age gap (standardized regression coefficient β = 0.126, P < 0.001) and worse cognition (β = -0.046, P = 0.013), while higher brain-age gap was associated with worse cognition (β=-0.163, P < 0.001). In mediation analysis, 24.7% of the total difference in cognition between the highest and lowest LIfestyle for BRAin health tertile was mediated by brain-age gap (β indirect = -0.049, P < 0.001; β total = -0.198, P < 0.001) and an additional 3.8% was mediated via connectivity (β indirect = -0.006, P < 0.001; β total = -0.150, P < 0.001). Findings suggest that associations between health- and lifestyle-based risk/protective factors (LIfestyle for BRAin health) and cognition can be partially explained by structural brain health markers (brain-age gap) and white matter connectivity markers. Lifestyle interventions targeted at high-risk individuals in mid-to-late life may be effective in promoting and preserving cognitive function in the general public.

Intracranial arteriosclerosis and the risk of dementia: A population-based cohort study

BACKGROUND

The impact of intracranial arteriosclerosis on dementia remains largely unclear.

METHODS

In 2339 stroke-free and dementia-free participants (52.2% women, mean age 69.5 years) from the general population, we assessed intracranial carotid artery calcification (ICAC) and vertebrobasilar artery calcification (VBAC) as proxy for arteriosclerosis. Associations with dementia were assessed using Cox models. In addition, indirect effects through cerebral small vessel disease (cSVD) and subcortical brain structure volumes were assessed using causal mediation analyses.

RESULTS

During a median of 13.4 years (25th-75th percentiles 9.9-14.5) of follow-up, 282 participants developed dementia. Both ICAC presence (hazard ratio [HR]: 1.53, 95% confidence interval [CI]: 1.00-2.32]) and volume (HR per standard deviation: 1.19, 95% CI: 1.01-1.40) increased dementia risk. For VBAC, severe calcifications increased dementia risk (HR for third vs first volume tertile: 1.89, 95% CI: 1.00-3.59). These effects were mediated partly through increased cSVD (percentage mediated for ICAC: 13% and VBAC: 24%).

DISCUSSION

Intracranial arteriosclerosis increases the risk of dementia.

Plasma neurofilament light chain in relation to 10-year change in cognition and neuroimaging markers: a population-based study

Neurofilament light chain (NfL) is a promising biomarker for risk stratification and disease monitoring of dementia, but its utility in the preclinical disease stage remains uncertain. We determined the association of plasma NfL with (change in) neuroimaging markers and cognition in the population-based Rotterdam Study, using linear and logistic regression and mixed-effects models. Plasma NfL levels were measured using the Simoa NF-light™ assay in 4705 dementia-free participants (mean age 71.9 years, 57% women), who underwent cognitive assessment and brain MRI with repeated assessments over a 10-year follow-up period. Higher plasma NfL was associated with worse cognitive performance at baseline (g-factor: β =  - 0.12 (- 0.15; - 0.09), p < 0.001), and accelerated cognitive decline during follow-up on the Stroop color naming task (β = 0.04 (0.02; 0.06), p < 0.001), with a smaller trend for decline in global cognition (g-factor β =  - 0.02 (- 0.04; 0.00), p = 0.044). In the subset of 975 participants with brain MRI, higher NfL was associated with poorer baseline white matter integrity (e.g., global mean diffusivity: β = 0.12 (0.06; 0.19), p < 0.001), with similar trends for volume of white matter hyperintensities (β = 0.09 (0.02; 0.16), p = 0.011) and presence of lacunes (OR = 1.55 (1.13; 2.14), p = 0.007). Plasma NfL was not associated with volumes or thickness of the total gray matter, hippocampus, or Alzheimer signature regions. In conclusion, higher plasma NfL levels are associated with cognitive decline and larger burden of primarily white matter pathology in the general population.

Effect of Magnetic Resonance Image Quality on Structural and Functional Brain Connectivity: The Maastricht Study

In population-based cohort studies, magnetic resonance imaging (MRI) is vital for examining brain structure and function. Advanced MRI techniques, such as diffusion-weighted MRI (dMRI) and resting-state functional MRI (rs-fMRI), provide insights into brain connectivity. However, biases in MRI data acquisition and processing can impact brain connectivity measures and their associations with demographic and clinical variables. This study, conducted with 5110 participants from The Maastricht Study, explored the relationship between brain connectivity and various image quality metrics (e.g., signal-to-noise ratio, head motion, and atlas-template mismatches) that were obtained from dMRI and rs-fMRI scans. Results revealed that in particular increased head motion (R2 up to 0.169, p < 0.001) and reduced signal-to-noise ratio (R2 up to 0.013, p < 0.001) negatively impacted structural and functional brain connectivity, respectively. These image quality metrics significantly affected associations of overall brain connectivity with age (up to -59%), sex (up to -25%), and body mass index (BMI) (up to +14%). Associations with diabetes status, educational level, history of cardiovascular disease, and white matter hyperintensities were generally less affected. This emphasizes the potential confounding effects of image quality in large population-based neuroimaging studies on brain connectivity and underscores the importance of accounting for it.

Where is VALDO? VAscular Lesions Detection and segmentatiOn challenge at MICCAI 2021

Imaging markers of cerebral small vessel disease provide valuable information on brain health, but their manual assessment is time-consuming and hampered by substantial intra- and interrater variability. Automated rating may benefit biomedical research, as well as clinical assessment, but diagnostic reliability of existing algorithms is unknown. Here, we present the results of the VAscular Lesions DetectiOn and Segmentation (Where is VALDO?) challenge that was run as a satellite event at the international conference on Medical Image Computing and Computer Aided Intervention (MICCAI) 2021. This challenge aimed to promote the development of methods for automated detection and segmentation of small and sparse imaging markers of cerebral small vessel disease, namely enlarged perivascular spaces (EPVS) (Task 1), cerebral microbleeds (Task 2) and lacunes of presumed vascular origin (Task 3) while leveraging weak and noisy labels. Overall, 12 teams participated in the challenge proposing solutions for one or more tasks (4 for Task 1-EPVS, 9 for Task 2-Microbleeds and 6 for Task 3-Lacunes). Multi-cohort data was used in both training and evaluation. Results showed a large variability in performance both across teams and across tasks, with promising results notably for Task 1-EPVS and Task 2-Microbleeds and not practically useful results yet for Task 3-Lacunes. It also highlighted the performance inconsistency across cases that may deter use at an individual level, while still proving useful at a population level.

Small vessel disease burden and functional brain connectivity in mild cognitive impairment

Background

The role of small vessel disease in the development of dementia is not yet completely understood. Functional brain connectivity has been shown to differ between individuals with and without cerebral small vessel disease. However, a comprehensive measure of small vessel disease quantifying the overall damage on the brain is not consistently used and studies using such measure in mild cognitive impairment individuals are missing.

Method

Functional brain connectivity differences were analyzed between mild cognitive impairment individuals with absent or low (n = 34) and high (n = 34) small vessel disease burden using data from the Parelsnoer Institute, a Dutch multicenter study. Small vessel disease was characterized using an ordinal scale considering: lacunes, microbleeds, perivascular spaces in the basal ganglia, and white matter hyperintensities. Resting state functional MRI data using 3 Tesla scanners was analyzed with group-independent component analysis using the CONN toolbox.

Results

Functional connectivity between areas of the cerebellum and between the cerebellum and the thalamus and caudate nucleus was higher in the absent or low small vessel disease group compared to the high small vessel disease group.

Conclusion

These findings might suggest that functional connectivity of mild cognitive impairment individuals with low or absent small vessel disease burden is more intact than in mild cognitive impairment individuals with high small vessel disease. These brain areas are mainly responsible for motor, attentional and executive functions, domains which in previous studies were found to be mostly associated with small vessel disease markers. Our results support findings on the involvement of the cerebellum in cognitive functioning.

Serum immunoglobulins and biomarkers of dementia: a population-based study

BACKGROUND

Inflammation plays a key role in the development of dementia, but its link to early biomarkers, particularly those in plasma or neuroimaging, remains elusive. This study aimed to investigate the association between serum immunoglobulins and biomarkers of dementia.

METHODS

Between 1997 and 2009, serum immunoglobulins (IgA, IgG and IgM) were measured in dementia-free participants of the population-based Rotterdam Study. A random subset of participants had assessment of biomarkers in plasma (total tau (t-tau), neurofilament light chain (NfL), amyloid-β40 (Aβ-40), amyloid-β42 (Aβ-42), while another subset of participants underwent neuroimaging to quantify brain volume, white matter structural integrity and markers of cerebral small vessel disease. Linear regression models were constructed to determine cross-sectional associations between IgA, IgG, IgM and biomarkers of dementia, with adjustment for potential confounders. Multiple testing correction was applied using the false discovery rate. As a sensitivity analysis, we re-ran the models for participants within the reference range of immunoglobulins, excluding those using immunomodulating drugs, and conducted a stratified analysis by APOE-ε4 carriership and sex.

RESULTS

Of 8,768 participants with serum immunoglobulins, 3,455 participants (65.8 years [interquartile range (IQR): 61.5-72.0], 57.2% female) had plasma biomarkers available and 3,139 participants (57.4 years [IQR: 52.7-60.7], 54.4% female) had neuroimaging data. Overall, no associations between serum immunoglobulins and biomarkers of dementia remained significant after correction for multiple testing. However, several suggestive associations were noted: higher serum IgA levels concurred with lower plasma levels of Aβ-42 (standardized adjusted mean difference: -0.015 [95% confidence interval (CI): -0.029--0.002], p = 2.8 × 10-2), and a lower total brain volume, mainly driven by less gray matter (-0.027 [-0.046--0.008], p = 6.0 × 10-3) and more white matter hyperintensities (0.047 [0.016 - 0.077], p = 3.0 × 10-3). In sensitivity analyses, higher IgM was linked to lower t-tau, Aβ-40, and Aβ-42, but also a loss of white matter microstructural integrity. Stratified analyses indicate that these associations potentially differ between carriers and non-carriers of the APOE-ε4 allele and men and women.

CONCLUSIONS

While associations between serum immunoglobulins and early markers of dementia could not be established in this population-based sample, it may be valuable to consider factors such as APOE-ε4 allele carriership and sex in future investigations.

Social Health Is Associated With Tract-Specific Brain White Matter Microstructure in Community-Dwelling Older Adults

Background

Poor social health has been linked to a risk of neuropsychiatric disorders. Neuroimaging studies have shown associations between social health and global white matter microstructural integrity. We aimed to identify which white matter tracts are involved in these associations.

Methods

Social health markers (loneliness, perceived social support, and partnership status) and white matter microstructural integrity of 15 white matter tracts (identified with probabilistic tractography after diffusion magnetic resonance imaging) were collected for 3352 participants (mean age 58.4 years, 54.9% female) from 2002 to 2008 in the Rotterdam Study. Cross-sectional associations were studied using multivariable linear regression.

Results

Loneliness was associated with higher mean diffusivity (MD) in the superior thalamic radiation and the parahippocampal part of the cingulum (standardized mean difference for both tracts: 0.21, 95% CI, 0.09 to 0.34). Better perceived social support was associated with lower MD in the forceps minor (standardized mean difference per point increase in social support: -0.06, 95% CI, -0.09 to -0.03), inferior fronto-occipital fasciculus, and uncinate fasciculus. In male participants, better perceived social support was associated with lower MD in the forceps minor, and not having a partner was associated with lower fractional anisotropy in the forceps minor. Loneliness was associated with higher MD in the superior thalamic radiation in female participants only.

Conclusions

Social health was associated with tract-specific white matter microstructure. Loneliness was associated with lower integrity of limbic and sensorimotor tracts, whereas better perceived social support was associated with higher integrity of association and commissural tracts, indicating that social health domains involve distinct neural pathways of the brain.

Hypoxic oligodendrocyte precursor cell-derived VEGFA is associated with blood-brain barrier impairment

Cerebral small vessel disease is characterised by decreased cerebral blood flow and blood-brain barrier impairments which play a key role in the development of white matter lesions. We hypothesised that cerebral hypoperfusion causes local hypoxia, affecting oligodendrocyte precursor cell-endothelial cell signalling leading to blood-brain barrier dysfunction as an early mechanism for the development of white matter lesions. Bilateral carotid artery stenosis was used as a mouse model for cerebral hypoperfusion. Pimonidazole, a hypoxic cell marker, was injected prior to humane sacrifice at day 7. Myelin content, vascular density, blood-brain barrier leakages, and hypoxic cell density were quantified. Primary mouse oligodendrocyte precursor cells were exposed to hypoxia and RNA sequencing was performed. Vegfa gene expression and protein secretion was examined in an oligodendrocyte precursor cell line exposed to hypoxia. Additionally, human blood plasma VEGFA levels were measured and correlated to blood-brain barrier permeability in normal-appearing white matter and white matter lesions of cerebral small vessel disease patients and controls. Cerebral blood flow was reduced in the stenosis mice, with an increase in hypoxic cell number and blood-brain barrier leakages in the cortical areas but no changes in myelin content or vascular density. Vegfa upregulation was identified in hypoxic oligodendrocyte precursor cells, which was mediated via Hif1α and Epas1. In humans, VEGFA plasma levels were increased in patients versus controls. VEGFA plasma levels were associated with increased blood-brain barrier permeability in normal appearing white matter of patients. Cerebral hypoperfusion mediates hypoxia induced VEGFA expression in oligodendrocyte precursor cells through Hif1α/Epas1 signalling. VEGFA could in turn increase BBB permeability. In humans, increased VEGFA plasma levels in cerebral small vessel disease patients were associated with increased blood-brain barrier permeability in the normal appearing white matter. Our results support a role of VEGFA expression in cerebral hypoperfusion as seen in cerebral small vessel disease.

Plasma ceramides relate to mild cognitive impairment in middle-aged men: The Maastricht Study

Introduction

There is an urgent need for biomarkers identifying individuals at risk of early-stage cognitive impairment. Using cross-sectional data from The Maastricht Study, this study included 197 individuals with mild cognitive impairment (MCI) and 200 cognitively unimpaired individuals aged 40 to 75, matched by age, sex, and educational level.

Methods

We assessed the association of plasma sphingolipid and ceramide transfer protein (CERT) levels with MCI and adjusted for potentially confounding risk factors. Furthermore, the relationship of plasma sphingolipids and CERTs with magnetic resonance imaging brain volumes was assessed and age- and sex-stratified analyses were performed.

Results

Associations of plasma ceramide species C18:0 and C24:1 and combined plasma ceramide chain lengths (ceramide risk score) with MCI were moderated by sex, but not by age, and higher levels were associated with MCI in men. No associations were found among women. In addition, higher levels of ceramide C20:0, C22:0, and C24:1, but not the ceramide risk score, were associated with larger volume of the hippocampus after controlling for covariates, independent of MCI. Although higher plasma ceramide C18:0 was related to higher plasma CERT levels, no association of CERT levels was found with MCI or brain volumes.

Discussion

Our results warrant further analysis of plasma ceramides as potential markers for MCI in middle-aged men. In contrast to previous studies, no associations of plasma sphingolipids with MCI or brain volumes were found in women, independent of age. These results highlight the importance of accounting for sex- and age-related factors when examining sphingolipid and CERT metabolism related to cognitive function.

Associations of neuroimaging markers with depressive symptoms over time in middle-aged and elderly persons

BACKGROUND

Cerebrovascular disease is regarded as a potential cause of late-life depression. Yet, evidence for associations of neuroimaging markers of vascular brain disease with depressive symptoms is inconclusive. We examined the associations of neuroimaging markers and depressive symptoms in a large population-based study of middle-aged and elderly persons over time.

METHODS

A total of 4943 participants (mean age = 64.6 ± 11.1 years, 55.7% women) from the Rotterdam Study were included. At baseline, total brain volume, gray matter volume, white matter volume, white matter hyperintensities volume, cortical infarcts, lacunar infarcts, microbleeds, white matter fractional anisotropy, and mean diffusivity (MD) were measured with a brain MRI (1.5T). Depressive symptoms were assessed twice with the Center for Epidemiologic Studies Depression scale (median follow-up time: 5.5 years, IQR = 0.9). To assess temporal associations of neuroimaging markers and depressive symptoms, linear mixed models were used.

RESULTS

A smaller total brain volume (β = -0.107, 95% CI -0.192 to -0.022), larger white matter hyperintensities volume (β = 0.047, 95% CI 0.010-0.084), presence of cortical infarcts (β = 0.194, 95% CI 0.047-0.341), and higher MD levels (β = 0.060, 95% CI 0.022-0.098) were cross-sectionally associated with more depressive symptoms. Longitudinal analyses showed that small total brain volume (β = -0.091, 95% CI -0.167 to -0.015) and presence of cortical infarcts (β = 0.168, 95% CI 0.022-0.314) were associated with increasing depressive symptoms over time. After stratification on age, effect sizes were more pronounced at older ages.

CONCLUSIONS

Neuroimaging markers of white matter microstructural damage were associated with depressive symptoms longitudinally in this study of middle-aged and elderly persons. These associations were more pronounced at older ages, providing evidence for the role of white matter structure in late-life depressive symptomatology.

Assessment of the clinical feasibility of detecting subtle blood-brain barrier leakage in cerebral small vessel disease using dynamic susceptibility contrast MRI

PURPOSE

Cerebral small vessel disease (cSVD) involves several pathologies affecting the small vessels, including blood-brain barrier (BBB) impairment. Dynamic susceptibility contrast (DSC) MRI is sensitive to both blood perfusion and BBB leakage, and correction methods may be crucial for obtaining reliable perfusion measures. These methods might also be applicable to detect BBB leakage itself. This study investigated to what extent DSC-MRI can measure subtle BBB leakage in a clinical feasibility setting.

METHODS

In vivo DCE and DSC data were collected from fifteen cSVD patients (71 (±10) years, 6F/9M) and twelve elderly controls (71 (±10) years, 4F/8M). DSC-derived leakage fractions were obtained using the Boxerman-Schmainda-Weisskoff method (K2). K2 was compared with the DCE-derived leakage rate K_i, obtained from Patlak analysis. Subsequently, differences were assessed between white matter hyperintensities (WMH), cortical gray matter (CGM), and normal-appearing white matter (NAWM). Additionally, computer simulations were performed to assess the sensitivity of DSC-MRI to BBB leakage.

RESULTS

K2 showed significant differences between tissue regions (P < 0.001 for CGM-NAWM and CGM-WMH, and P = 0.001 for NAWM-WMH). Conversely, according to the computer simulations the DSC sensitivity was insufficient to measure subtle BBB leakage, as the K2 values were below the derived limit of quantification (4∙10^-3 min^-1). As expected, K_i was elevated in the WMH compared to CGM and NAWM (P < 0.001).

CONCLUSIONS

Although clinical DSC-MRI seems capable to detect subtle BBB leakage differences between WMH and normal-appearing brain tissue it is not recommended. K2 as a direct measure for subtle BBB leakage remains ambiguous as its signal effects are due to mixed T₁- and T₂^∗-weighting. Further research is warranted to better disentangle perfusion from leakage effects.

Semisupervised white matter hyperintensities segmentation on MRI

This study proposed a semisupervised loss function named level-set loss (LSLoss) for cerebral white matter hyperintensities (WMHs) segmentation on fluid-attenuated inversion recovery images. The training procedure did not require manually labeled WMH masks. Our image preprocessing steps included biased field correction, skull stripping, and white matter segmentation. With the proposed LSLoss, we trained a V-Net using the MRI images from both local and public databases. Local databases were the small vessel disease cohort (HKU-SVD, n = 360) and the multiple sclerosis cohort (HKU-MS, n = 20) from our institutional imaging center. Public databases were the Medical Image Computing Computer-assisted Intervention (MICCAI) WMH challenge database (MICCAI-WMH, n = 60) and the normal control cohort of the Alzheimer's Disease Neuroimaging Initiative database (ADNI-CN, n = 15). We achieved an overall dice similarity coefficient (DSC) of 0.81 on the HKU-SVD testing set (n = 20), DSC = 0.77 on the HKU-MS testing set (n = 5), and DSC = 0.78 on MICCAI-WMH testing set (n = 30). The segmentation results obtained by our semisupervised V-Net were comparable with the supervised methods and outperformed the unsupervised methods in the literature.

Intracranial Carotid Arteriosclerosis Mediates the Association Between Blood Pressure and Cerebral Small Vessel Disease

BACKGROUND

Intracranial arteriosclerosis could explain the association between blood pressure (BP) and cerebral small vessel disease (CSVD). Therefore, we tested whether intracranial carotid artery calcification (ICAC) mediates the association between BP and CSVD and determined pathophysiological mechanisms based on ICAC subtypes.

METHODS

One thousand four hundred fifty-eight stroke-free participants from the Rotterdam Study (mean age, 68 years; 52% women) underwent nonenhanced computed tomography scans to quantify ICAC volume (mm³) between 2003 and 2015. ICAC was categorized into intimal and internal elastic lamina calcifications. CSVD included white matter hyperintensities volume, the presence of lacunes, and cerebral microbleeds visualized on magnetic resonance imaging. Office BP included systolic BP, diastolic BP, pulse pressure, and mean arterial pressure. Mediation analysis included a 2-way decomposition to determine the direct association between BP and CSVD and the indirect or mediated effect (negative or positive mediations expressed in %) of log-ICAC volume on such association.

RESULTS

BP and log-ICAC were correlated and were also associated with CSVD. In all participants, total log-ICAC volume mediated the association of diastolic BP (-14.5%) and pulse pressure (16.5%) with log-white matter hyperintensities. Internal elastic lamina log-ICAC volume mediated -19.5% of the association between diastolic BP and log-white matter hyperintensities; intimal log-ICAC volume did not mediate associations. For lacunes, total and internal elastic lamina log-ICAC volume mediated the association of diastolic BP (-40% and -45.8%) and pulse pressure (26.9% and 18.2%). We did not observe mediations for cerebral microbleeds.

CONCLUSIONS

Intracranial arteriosclerosis mediates the association between BP and CSVD. Internal elastic lamina calcification, considered a proxy of arterial stiffness, is the leading mechanism explaining the link between BP and CSVD.

Efficient optimization based thresholding technique for analysis of alzheimer MRIs

Purpose study: Alzheimer is a type of dementia that usually affects older adults by creating memory loss due to damaged brain cells. The damaged brain cells lead to shrinkage in the size of the brain and it is very difficult to extract the grey matter (GM) and white matter (WM). The segmentation of GM and WM is a challenging task due to its homogeneous nature between the neighborhood tissues. In this proposed system, an attempt has been made to extract GM and WM tissues using optimization-based segmentation techniques.Materials and methods: The optimization method is considered for the classification of normal and alzheimer disease (ad) through magnetic resonance images (mri) using a modified cuckoo search algorithm. Gray Level Co-Occurrence Matrix (GLCM) features are calculated from the extracted GM and WM. Principal Component Analysis (PCA) is adopted for selecting the best features from the GLCM features. Support Vector Machine (SVM) is a classifier which is used to classify the normal and abnormal images. Results: The proposed optimization algorithm provides most promising and efficient level of image segmentation compared to fuzzy c means (fcm), otsu, particle swarm optimization (pso) and cuckoo search (cs). The modified cuckoo yields high accuracy of 96%, sensitivity of 97% and specificity of 94% than other methods due to its powerful searching potential for the proper identification of gray and WM tissues.Conclusions: The results of the classification process proved the effectiveness of the proposed technique in identifying alzheimer affected patients due to its very strong optimization ability. The proposed pipeline helps to diagnose early detection of AD and better assessment of the neuroprotective effect of a therapy.

Determinants of Perivascular Spaces in the General Population: A Pooled Cohort Analysis of Individual Participant Data

BACKGROUND AND OBJECTIVES

Perivascular spaces (PVS) are emerging markers of cerebral small vessel disease (CSVD), but research on their determinants has been hampered by conflicting results from small single studies using heterogeneous rating methods. In this study, we therefore aimed to identify determinants of PVS burden in a pooled analysis of multiple cohort studies using 1 harmonized PVS rating method.

METHODS

Individuals from 10 population-based cohort studies with adult participants from the Uniform Neuro-Imaging of Virchow-Robin Spaces Enlargement consortium and the UK Biobank were included. On MRI scans, we counted PVS in 4 brain regions (mesencephalon, hippocampus, basal ganglia, and centrum semiovale) according to a uniform and validated rating protocol, both manually and automated using a deep learning algorithm. As potential determinants, we considered demographics, cardiovascular risk factors, APOE genotypes, and other imaging markers of CSVD. Negative binomial regression models were used to examine the association between these determinants and PVS counts.

RESULTS

In total, 39,976 individuals were included (age range 20-96 years). The average count of PVS in the 4 regions increased from the age 20 years (0-1 PVS) to 90 years (2-7 PVS). Men had more mesencephalic PVS (OR [95% CI] = 1.13 [1.08-1.18] compared with women), but less hippocampal PVS (0.82 [0.81-0.83]). Higher blood pressure, particularly diastolic pressure, was associated with more PVS in all regions (ORs between 1.04-1.05). Hippocampal PVS showed higher counts with higher high-density lipoprotein cholesterol levels (1.02 [1.01-1.02]), glucose levels (1.02 [1.01-1.03]), and APOE ε4-alleles (1.02 [1.01-1.04]). Furthermore, white matter hyperintensity volume and presence of lacunes were associated with PVS in multiple regions, but most strongly with the basal ganglia (1.13 [1.12-1.14] and 1.10 [1.09-1.12], respectively).

DISCUSSION

Various factors are associated with the burden of PVS, in part regionally specific, which points toward a multifactorial origin beyond what can be expected from PVS-related risk factor profiles. This study highlights the power of collaborative efforts in population neuroimaging research.

Twenty-four hour blood pressure variability and the prevalence and the progression of cerebral white matter hyperintensities

Blood pressure variability (BPV) is related to cerebral white matter hyperintensities (WMH), but longitudinal studies assessing WMH progression are scarce. Patients with cardiovascular disease and control participants of the Heart-Brain Connection Study underwent 24-hour ambulatory blood pressure monitoring and repeated brain MRI at baseline and after 2 years. Using linear regression, we determined whether different measures of BPV (standard deviation, coefficient of variation, average real variability (ARV), variability independent of the mean) and nocturnal dipping were associated with WMH and whether this association was mediated or moderated by baseline cerebral perfusion. Among 177 participants (mean age: 65.9 ± 8.1 years, 33.9% female), the absence of diastolic nocturnal dipping was associated with higher WMH volume at baseline (β = 0.208, 95%CI: 0.025-0.392), but not with WMH progression among 91 participants with follow-up imaging. None of the BPV measures were associated with baseline WMH. Only 24-hour diastolic ARV was significantly associated with WMH progression (β = 0.144, 95%CI: 0.030-0.258), most profound in participants with low cerebral perfusion at baseline (p-interaction = 0.042). In conclusion, absent diastolic nocturnal dipping and 24-hour diastolic ARV were associated with higher WMH volume. Whilst requiring replication, these findings suggest that blood pressure patterns and variability may be a target for prevention of small vessel disease.

The role of cognitive and brain reserve in late-life depressive events: The Rotterdam Study

BACKGROUND

Cognitive and brain reserve aim to explain individual differences in susceptibility to dementia and may also affect the risk of late-life depressive events. We assessed whether higher cognitive and brain reserve are associated with lower risk of a late-life depressive event.

METHODS

This study included 4509 participants from the population-based Rotterdam Study (mean age: 63.4 ± 10.2 years, 55 % women) between 2005 and 2019. Participants completed cognitive testing and brain-MRI at baseline. Cognitive reserve was defined as the common variance across cognitive tests, while adjusting for demographic factors and brain MRI-markers. Brain reserve was defined as total brain volume divided by intracranial volume. Depressive events (depressive symptoms/depressive syndrome/major depressive disorder) were repeatedly measured (follow-up: 6.6 ± 3.9 years) with validated questionnaires, clinical interviews, and follow-up of medical records. Hazard ratios (HR) with 95 % confidence intervals (CI) were estimated using Cox-regressions.

RESULTS

Higher cognitive (HR: 0.91/SD, 95%CI: 0.84; 1.00) and brain reserve (HR: 0.88/SD, 95%CI: 0.77; 1.00) were associated with a lower risk of a depressive event after adjustment for baseline depressive symptoms. These associations attenuated when participants with clinically relevant depressive symptoms at baseline were excluded (HR: 0.95/SD, 95%CI: 0.86; 1.05, HR: 0.89/SD, 95%CI: 0.76; 1.03, respectively).

LIMITATIONS

No data was available on depression in early-life and the number of participants with major depressive disorder was relatively low (n = 105).

CONCLUSIONS

Higher cognitive and brain reserve may be protective factors for late-life depressive events, particularly in those who have experienced clinical relevant depressive symptoms before. Further research is needed to determine whether cognitive and brain reserve could be used as targets for the prevention of late-life depression.

Cognitive resilience depends on white matter connectivity: The Maastricht Study

INTRODUCTION

Differences in brain network connectivity may reflect the capability of the neurological substrate to compensate for brain damage and preserve cognitive function (cognitive reserve). We examined the associations between white matter connectivity, brain damage markers, and cognition in a population sample of middle-aged individuals.

METHODS

A total of 4759 participants from The Maastricht Study (mean age = 59.2, SD = 8.7, 50.2% male) underwent cognitive testing and diffusion magnetic resonance imaging (dMRI), from which brain volume, structural connectivity, and vascular damage were quantified. Multivariable linear regression was used to investigate whether connectivity modified the association between brain damage and cognition, adjusted for demographic and cardiometabolic risk factors.

RESULTS

More atrophic and vascular brain damage was associated with worse cognition scores. Increasing connectivity moderated the negative association between damage and cognition (χ² = 8.64, df = 3, p ≤ 0.001); individuals with high damage but strong connectivity showed normal cognition.

DISCUSSION

Findings support the reserve hypothesis by showing that brain connectivity is associated with cognitive resilience.

A Fuzzy Consensus Clustering Algorithm for MRI Brain Tissue Segmentation

Brain tissue segmentation is an important component of the clinical diagnosis of brain diseases using multi-modal magnetic resonance imaging (MR). Brain tissue segmentation has been developed by many unsupervised methods in the literature. The most commonly used unsupervised methods are K-Means, Expectation-Maximization, and Fuzzy Clustering. Fuzzy clustering methods offer considerable benefits compared with the aforementioned methods as they are capable of handling brain images that are complex, largely uncertain, and imprecise. However, this approach suffers from the intrinsic noise and intensity inhomogeneity (IIH) in the data resulting from the acquisition process. To resolve these issues, we propose a fuzzy consensus clustering algorithm that defines a membership function resulting from a voting schema to cluster the pixels. In particular, we first pre-process the MRI data and employ several segmentation techniques based on traditional fuzzy sets and intuitionistic sets. Then, we adopted a voting schema to fuse the results of the applied clustering methods. Finally, to evaluate the proposed method, we used the well-known performance measures (boundary measure, overlap measure, and volume measure) on two publicly available datasets (OASIS and IBSR18). The experimental results show the superior performance of the proposed method in comparison with the recent state of the art. The performance of the proposed method is also presented using a real-world Autism Spectrum Disorder Detection problem with better accuracy compared to other existing methods.

A cluster of blood-based protein biomarkers reflecting coagulation relates to the burden of cerebral small vessel disease

Biological processes underlying cerebral small vessel disease (cSVD) are largely unknown. We hypothesized that identification of clusters of inter-related bood-based biomarkers that are associated with the burden of cSVD provides leads on underlying biological processes. In 494 participants (mean age 67.6 ± 8.7 years; 36% female; 75% cardiovascular diseases; 25% reference participants) we assessed the relation between 92 blood-based biomarkers from the OLINK cardiovascular III panel and cSVD, using cluster-based analyses. We focused particularly on white matter hyperintensities (WMH). Nineteen biomarkers individually correlated with WMH ratio (r range: 0.16-0.27, Bonferroni corrected p-values <0.05), of which sixteen biomarkers formed one biomarker cluster. Pathway analysis showed that this biomarker cluster predominantly reflected coagulation processes. This cluster related also significantly to other cSVD manifestations (lacunar infarcts, microbleeds, and enlarged perivascular spaces), which supports generalizability beyond WMHs. To study possible causal effects of biological processes reflected by the cluster we performed a mediation analysis that showed a mediation effect of the cluster on the relation between age and WMH ratio (proportion mediated 17%), and hypertension and WMH-volume (proportion mediated 21%). In conclusion, we identified a cluster of blood-based biomarkers reflecting coagulation, that is related to manifestations of cSVD, corroborating involvement of coagulation abnormalities in the etiology of cSVD.

A general skull stripping of multiparametric brain MRIs using 3D convolutional neural network

Accurate skull stripping facilitates following neuro-image analysis. For computer-aided methods, the presence of brain skull in structural magnetic resonance imaging (MRI) impacts brain tissue identification, which could result in serious misjudgments, specifically for patients with brain tumors. Though there are several existing works on skull stripping in literature, most of them either focus on healthy brain MRIs or only apply for a single image modality. These methods may be not optimal for multiparametric MRI scans. In the paper, we propose an ensemble neural network (EnNet), a 3D convolutional neural network (3DCNN) based method, for brain extraction on multiparametric MRI scans (mpMRIs). We comprehensively investigate the skull stripping performance by using the proposed method on a total of 15 image modality combinations. The comparison shows that utilizing all modalities provides the best performance on skull stripping. We have collected a retrospective dataset of 815 cases with/without glioblastoma multiforme (GBM) at the University of Pittsburgh Medical Center (UPMC) and The Cancer Imaging Archive (TCIA). The ground truths of the skull stripping are verified by at least one qualified radiologist. The quantitative evaluation gives an average dice score coefficient and Hausdorff distance at the 95th percentile, respectively. We also compare the performance to the state-of-the-art methods/tools. The proposed method offers the best performance.The contributions of the work have five folds: first, the proposed method is a fully automatic end-to-end for skull stripping using a 3D deep learning method. Second, it is applicable for mpMRIs and is also easy to customize for any MRI modality combination. Third, the proposed method not only works for healthy brain mpMRIs but also pre-/post-operative brain mpMRIs with GBM. Fourth, the proposed method handles multicenter data. Finally, to the best of our knowledge, we are the first group to quantitatively compare the skull stripping performance using different modalities. All code and pre-trained model are available at: https://github.com/plmoer/skull_stripping_code_SR .

Circulating Metabolome and White Matter Hyperintensities in Women and Men

BACKGROUND

White matter hyperintensities (WMH), identified on T2-weighted magnetic resonance images of the human brain as areas of enhanced brightness, are a major risk factor of stroke, dementia, and death. There are no large-scale studies testing associations between WMH and circulating metabolites.

METHODS

We studied up to 9290 individuals (50.7% female, average age 61 years) from 15 populations of 8 community-based cohorts. WMH volume was quantified from T2-weighted or fluid-attenuated inversion recovery images or as hypointensities on T1-weighted images. Circulating metabolomic measures were assessed with mass spectrometry and nuclear magnetic resonance spectroscopy. Associations between WMH and metabolomic measures were tested by fitting linear regression models in the pooled sample and in sex-stratified and statin treatment-stratified subsamples. Our basic models were adjusted for age, sex, age×sex, and technical covariates, and our fully adjusted models were also adjusted for statin treatment, hypertension, type 2 diabetes, smoking, body mass index, and estimated glomerular filtration rate. Population-specific results were meta-analyzed using the fixed-effect inverse variance-weighted method. Associations with false discovery rate (FDR)-adjusted P values (PFDR)<0.05 were considered significant.

RESULTS

In the meta-analysis of results from the basic models, we identified 30 metabolomic measures associated with WMH (PFDR<0.05), 7 of which remained significant in the fully adjusted models. The most significant association was with higher level of hydroxyphenylpyruvate in men (PFDR.full.adj=1.40×10-7) and in both the pooled sample (PFDR.full.adj=1.66×10-4) and statin-untreated (PFDR.full.adj=1.65×10-6) subsample. In men, hydroxyphenylpyruvate explained 3% to 14% of variance in WMH. In men and the pooled sample, WMH were also associated with lower levels of lysophosphatidylcholines and hydroxysphingomyelins and a larger diameter of low-density lipoprotein particles, likely arising from higher triglyceride to total lipids and lower cholesteryl ester to total lipids ratios within these particles. In women, the only significant association was with higher level of glucuronate (PFDR=0.047).

CONCLUSIONS

Circulating metabolomic measures, including multiple lipid measures (eg, lysophosphatidylcholines, hydroxysphingomyelins, low-density lipoprotein size and composition) and nonlipid metabolites (eg, hydroxyphenylpyruvate, glucuronate), associate with WMH in a general population of middle-aged and older adults. Some metabolomic measures show marked sex specificities and explain a sizable proportion of WMH variance.

Detection of subtle white matter lesions in MRI through texture feature extraction and boundary delineation using an embedded clustering strategy

White matter lesions (WML) underlie multiple brain disorders, and automatic WML segmentation is crucial to evaluate the natural disease course and effectiveness of clinical interventions, including drug discovery. Although recent research has achieved tremendous progress in WML segmentation, accurate detection of subtle WML present early in the disease course remains particularly challenging. Here we propose an approach to automatic WML segmentation of mild WML loads using an intensity standardisation technique, gray level co-occurrence matrix (GLCM) embedded clustering technique, and random forest (RF) classifier to extract texture features and identify morphology specific to true WML. We precisely define their boundaries through a local outlier factor (LOF) algorithm that identifies edge pixels by local density deviation relative to its neighbors. The automated approach was validated on 32 human subjects, demonstrating strong agreement and correlation (excluding one outlier) with manual delineation by a neuroradiologist through Intra-Class Correlation (ICC = 0.881, 95% CI 0.769, 0.941) and Pearson correlation (r = 0.895, p-value < 0.001), respectively, and outperforming three leading algorithms (Trimmed Mean Outlier Detection, Lesion Prediction Algorithm, and SALEM-LS) in five of the six established key metrics defined in the MICCAI Grand Challenge. By facilitating more accurate segmentation of subtle WML, this approach may enable earlier diagnosis and intervention.

Qualitative and Quantitative Comparison of Hippocampal Volumetric Software Applications: Do All Roads Lead to Rome?

Brain volumetric software is increasingly suggested for clinical routine. The present study quantifies the agreement across different software applications. Ten cases with and ten gender- and age-adjusted healthy controls without hippocampal atrophy (median age: 70; 25–75% range: 64–77 years and 74; 66–78 years) were retrospectively selected from a previously published cohort of Alzheimer’s dementia patients and normal ageing controls. Hippocampal volumes were computed based on 3 Tesla T1-MPRAGE-sequences with FreeSurfer (FS), Statistical-Parametric-Mapping (SPM; Neuromorphometrics and Hammers atlases), Geodesic-Information-Flows (GIF), Similarity-and-Truth-Estimation-for-Propagated-Segmentations (STEPS), and Quantib™. MTA (medial temporal lobe atrophy) scores were manually rated. Volumetric measures of each individual were compared against the mean of all applications with intraclass correlation coefficients (ICC) and Bland–Altman plots. Comparing against the mean of all methods, moderate to low agreement was present considering categorization of hippocampal volumes into quartiles. ICCs ranged noticeably between applications (left hippocampus (LH): from 0.42 (STEPS) to 0.88 (FS); right hippocampus (RH): from 0.36 (Quantib™) to 0.86 (FS). Mean differences between individual methods and the mean of all methods [mm³] were considerable (LH: FS −209, SPM-Neuromorphometrics −820; SPM-Hammers −1474; Quantib™ −680; GIF 891; STEPS 2218; RH: FS −232, SPM-Neuromorphometrics −745; SPM-Hammers −1547; Quantib™ −723; GIF 982; STEPS 2188). In this clinically relevant sample size with large spread in data ranging from normal aging to severe atrophy, hippocampal volumes derived by well-accepted applications were quantitatively different. Thus, interchangeable use is not recommended.

Risk factors, neuroimaging correlates and prognosis of the motoric cognitive risk syndrome: a population-based comparison with mild cognitive impairment

Background and purpose

This study was undertaken to compare risk factors, neuroimaging characteristics and prognosis between two clinical prodromes of dementia, namely, the motoric cognitive risk syndrome (MCRS) and mild cognitive impairment (MCI).

Methods

Between 2009 and 2015, dementia‐free participants of the population‐based Rotterdam Study were classified with a dementia prodrome if they had subjective cognitive complaints and scored >1 SD below the population mean of gait speed (MCRS) or >1.5 SD below the population mean of cognitive test scores (MCI). Using multinomial logistic regression models, we determined cross‐sectional associations of risk factors and structural neuroimaging markers with MCRS and MCI, followed by subdistribution hazard models, to determine risk of incident dementia until 2016.

Results

Of 3025 included participants (mean age = 70.4 years, 54.7% women), 231 had MCRS (7.6%), 132 had MCI (4.4%), and 62 (2.0%) fulfilled criteria for both. Although many risk factors were shared, a higher body mass index predisposed to MCRS, whereas male sex and hypercholesterolemia were associated with MCI only. Gray matter volumes, hippocampal volumes, white matter hyperintensities, and structural white matter integrity were worse in both MCRS and MCI. During a mean follow‐up of 3.9 years, 71 individuals developed dementia and 200 died. Five‐year cumulative risk of dementia was 7.0% (2.5%–11.5%) for individuals with MCRS, versus 13.3% (5.8%–20.8%) with MCI and only 2.3% (1.5%–3.1%) in unaffected individuals.

Conclusions

MCRS is associated with imaging markers of neurodegeneration and risk of dementia, even in the absence of MCI, highlighting the potential of motor function assessment in early risk stratification for dementia.

Magnetic Resonance Imaging Markers for Cognitive Impairment in Parkinson’s Disease: Current View

Cognitive impairment (CI) ranging from mild cognitive impairment (MCI) to dementia is a common and disturbing complication in patients with Parkinson’s disease (PD). Numerous studies have focused on neuropathological mechanisms underlying CI in PD, along with the identification of specific biomarkers for CI. Magnetic resonance imaging (MRI), a promising method, has been adopted to examine the changes in the brain and identify the candidate biomarkers associated with CI. In this review, we have summarized the potential biomarkers for CI in PD which have been identified through multi-modal MRI studies. Structural MRI technology is widely used in biomarker research. Specific patterns of gray matter atrophy are promising predictors of the evolution of CI in patients with PD. Moreover, other MRI techniques, such as MRI related to small-vessel disease, neuromelanin-sensitive MRI, quantitative susceptibility mapping, MR diffusion imaging, MRI related to cerebrovascular abnormality, resting-state functional MRI, and proton magnetic resonance spectroscopy, can provide imaging features with a good degree of prediction for CI. In the future, novel combined biomarkers should be developed using the recognized analysis tools and predictive algorithms in both cross-sectional and longitudinal studies.

The Effect of Training Sample Size on the Prediction of White Matter Hyperintensity Volume in a Healthy Population Using BIANCA

Introduction: White matter hyperintensities of presumed vascular origin (WMH) are an important magnetic resonance imaging marker of cerebral small vessel disease and are associated with cognitive decline, stroke, and mortality. Their relevance in healthy individuals, however, is less clear. This is partly due to the methodological challenge of accurately measuring rare and small WMH with automated segmentation programs. In this study, we tested whether WMH volumetry with FMRIB software library v6.0 (FSL; https://fsl.fmrib.ox.ac.uk/fsl/fslwiki) Brain Intensity AbNormality Classification Algorithm (BIANCA), a customizable and trainable algorithm that quantifies WMH volume based on individual data training sets, can be optimized for a normal aging population.

Methods: We evaluated the effect of varying training sample sizes on the accuracy and the robustness of the predicted white matter hyperintensity volume in a population (n = 201) with a low prevalence of confluent WMH and a substantial proportion of participants without WMH. BIANCA was trained with seven different sample sizes between 10 and 40 with increments of 5. For each sample size, 100 random samples of T1w and FLAIR images were drawn and trained with manually delineated masks. For validation, we defined an internal and external validation set and compared the mean absolute error, resulting from the difference between manually delineated and predicted WMH volumes for each set. For spatial overlap, we calculated the Dice similarity index (SI) for the external validation cohort.

Results: The study population had a median WMH volume of 0.34 ml (IQR of 1.6 ml) and included n = 28 (18%) participants without any WMH. The mean absolute error of the difference between BIANCA prediction and manually delineated masks was minimized and became more robust with an increasing number of training participants. The lowest mean absolute error of 0.05 ml (SD of 0.24 ml) was identified in the external validation set with a training sample size of 35. Compared to the volumetric overlap, the spatial overlap was poor with an average Dice similarity index of 0.14 (SD 0.16) in the external cohort, driven by subjects with very low lesion volumes.

Discussion: We found that the performance of BIANCA, particularly the robustness of predictions, could be optimized for use in populations with a low WMH load by enlargement of the training sample size. Further work is needed to evaluate and potentially improve the prediction accuracy for low lesion volumes. These findings are important for current and future population-based studies with the majority of participants being normal aging people.

Long-term association of pregnancy and maternal brain structure: the Rotterdam Study

The peripartum period is the highest risk interval for the onset or exacerbation of psychiatric illness in women’s lives. Notably, pregnancy and childbirth have been associated with short-term structural and functional changes in the maternal human brain. Yet the long-term effects of pregnancy on maternal brain structure remain unknown. We investigated a large population-based cohort to examine the association between parity and brain structure. In total, 2,835 women (mean age 65.2 years; all free from dementia, stroke, and cortical brain infarcts) from the Rotterdam Study underwent magnetic resonance imaging (1.5 T) between 2005 and 2015. Associations of parity with global and lobar brain tissue volumes, white matter microstructure, and markers of vascular brain disease were examined using regression models. We found that parity was associated with a larger global gray matter volume (β = 0.14, 95% CI = 0.09–0.19), a finding that persisted following adjustment for sociodemographic factors. A non-significant dose-dependent relationship was observed between a higher number of childbirths and larger gray matter volume. The gray matter volume association with parity was globally proportional across lobes. No associations were found regarding white matter volume or integrity, nor with markers of cerebral small vessel disease. The current findings suggest that pregnancy and childbirth are associated with robust long-term changes in brain structure involving a larger global gray matter volume that persists for decades. Future studies are warranted to further investigate the mechanism and physiological relevance of these differences in brain morphology.

Automatic quantification of white matter hyperintensities on T2-weighted fluid attenuated inversion recovery magnetic resonance imaging

White matter hyperintensities (WMH) are areas of increased signal visualized on T2-weighted fluid attenuated inversion recovery (FLAIR) brain magnetic resonance imaging (MRI) sequences. They are typically attributed to small vessel cerebrovascular disease in the context of aging. Among older adults, WMH are associated with risk of cognitive decline and dementia, stroke, and various other health outcomes. There has been increasing interest in incorporating quantitative WMH measurement as outcomes in clinical trials, observational research, and clinical settings. Here, we present a novel, fully automated, unsupervised detection algorithm for WMH segmentation and quantification. The algorithm uses a robust preprocessing pipeline, including brain extraction and a sample-specific mask that incorporates spatial information for automatic false positive reduction, and a half Gaussian mixture model (HGMM). The method was evaluated in 24 participants with varying degrees of WMH (4.9-78.6 cm³) from a community-based study of aging and dementia with dice coefficient, sensitivity, specificity, correlation, and bias relative to the ground truth manual segmentation approach performed by two expert raters. Results were compared with those derived from commonly used available WMH segmentation packages, including SPM lesion probability algorithm (LPA), SPM lesion growing algorithm (LGA), and Brain Intensity AbNormality Classification Algorithm (BIANCA). The HGMM algorithm derived WMH values that had a dice score of 0.87, sensitivity of 0.89, and specificity of 0.99 compared to ground truth. White matter hyperintensity volumes derived with HGMM were strongly correlated with ground truth values (r = 0.97, p = 3.9e-16), with no observable bias (-1.1 [-2.6, 0.44], p-value = 0.16). Our novel algorithm uniquely uses a robust preprocessing pipeline and a half-Gaussian mixture model to segment WMH with high agreement with ground truth for large scale studies of brain aging.

Perforating artery flow velocity and pulsatility in patients with carotid occlusive disease. A 7 tesla MRI study

•

Patients with COD show lower perforating artery flow velocity in watershed areas.

•

Reduced perforating artery flow seems to be related to subcortical ischemic burden.

•

Perforating artery function may be relevant for clinical outcome in COD patients.

Patients with carotid occlusive disease express altered hemodynamics in the post-occlusive vasculature and lesions commonly attributed to cerebral small vessel disease (SVD). We addressed the question if cerebral perforating artery flow measures, using a novel 7T MRI technique, are altered and related to SVD lesion burden in patients with carotid occlusive disease.

21 patients were included with a uni- (18) or bilateral (3) carotid occlusion (64±7 years) and 19 controls (65±10 years). Mean flow velocity and pulsatility in the perforating arteries in the semi-oval center (CSO) and basal ganglia (BG), measured with a 2D phase contrast 7T MRI sequence, were compared between patients and controls, and between hemispheres in patients with unilateral carotid occlusive disease. In patients, relations were assessed between perforating artery flow measures and SVD burden score and white matter hyperintensity (WMH) volume.

CSO perforating artery flow velocity was lower in patients than controls, albeit non-significant (mean difference [95% confidence interval] 0.08 cm/s [0.00–0.16]; p = 0.053), but pulsatility was similar (0.07  [-0.04–0.18]; p = 0.23). BG flow velocity and pulsatility did not differ between patients and controls (velocity = 0.28 cm/s [-0.32–0.88]; p = 0.34; pulsatility = 0.00 [-0.10–0.11]; p = 0.97). Patients with unilateral carotid occlusive disease showed no significant interhemispheric flow differences. Though non-significant, within patients lower CSO (p = 0.06) and BG (p = 0.11) flow velocity related to larger WMH volume.

Our findings suggest that carotid occlusive disease may be associated with abnormal cerebral perforating artery flow and that this relates to SVD lesion burden in these patients, although our observations need corroboration in larger study populations.

Comparison of domain adaptation techniques for white matter hyperintensity segmentation in brain MR images

Robust automated segmentation of white matter hyperintensities (WMHs) in different datasets (domains) is highly challenging due to differences in acquisition (scanner, sequence), population (WMH amount and location) and limited availability of manual segmentations to train supervised algorithms. In this work we explore various domain adaptation techniques such as transfer learning and domain adversarial learning methods, including domain adversarial neural networks and domain unlearning, to improve the generalisability of our recently proposed triplanar ensemble network, which is our baseline model. We used datasets with variations in intensity profile, lesion characteristics and acquired using different scanners. For the source domain, we considered a dataset consisting of data acquired from 3 different scanners, while the target domain consisted of 2 datasets. We evaluated the domain adaptation techniques on the target domain datasets, and additionally evaluated the performance on the source domain test dataset for the adversarial techniques. For transfer learning, we also studied various training options such as minimal number of unfrozen layers and subjects required for fine-tuning in the target domain. On comparing the performance of different techniques on the target dataset, domain adversarial training of neural network gave the best performance, making the technique promising for robust WMH segmentation.

Extracerebral microvascular dysfunction is related to brain MRI markers of cerebral small vessel disease: The Maastricht Study

BACKGROUND

Cerebral small vessel disease (cSVD) is a late consequence of cerebral microvascular dysfunction (MVD). MVD is hard to measure in the brain due to its limited accessibility. Extracerebral MVD (eMVD) measures can give insights in the etiology of cerebral MVD, as MVD may be a systemic process. We aim to investigate whether a compound score consisting of several eMVD measures is associated with structural cSVD MRI markers.

METHODS

Cross-sectional data of the population-based Maastricht Study was used (n = 1872, mean age 59 ± 8 years, 49% women). Measures of eMVD included flicker light-induced retinal arteriolar and venular dilation response (retina), albuminuria and glomerular filtration rate (kidney), heat-induced skin hyperemia (skin), and plasma biomarkers of endothelial dysfunction (sICAM-1, sVCAM-1, sE-selectin, and von Willebrand factor). These measures were standardized into z scores and summarized into a compound score. Linear and logistic regression analyses were used to investigate the associations between the compound score and white matter hyperintensity (WMH) volume, and the presence of lacunes and microbleeds, as measured by brain MRI.

RESULTS

The eMVD compound score was associated with WMH volume independent of age, sex, and cardiovascular risk factors (St β 0.057 [95% CI 0.010-0.081], p value 0.01), but not with the presence of lacunes (OR 1.011 [95% CI 0.803-1.273], p value 0.92) or microbleeds (OR 1.055 [95% CI 0.896-1.242], p value 0.52).

CONCLUSION

A compound score of eMVD is associated with WMH volume. Further research is needed to expand the knowledge about the role of systemic MVD in the pathophysiology of cSVD.

Whole brain segmentation with full volume neural network

Whole brain segmentation is an important neuroimaging task that segments the whole brain volume into anatomically labeled regions-of-interest. Convolutional neural networks have demonstrated good performance in this task. Existing solutions, usually segment the brain image by classifying the voxels, or labeling the slices or the sub-volumes separately. Their representation learning is based on parts of the whole volume whereas their labeling result is produced by aggregation of partial segmentation. Learning and inference with incomplete information could lead to sub-optimal final segmentation result. To address these issues, we propose to adopt a full volume framework, which feeds the full volume brain image into the segmentation network and directly outputs the segmentation result for the whole brain volume. The framework makes use of complete information in each volume and can be implemented easily. An effective instance in this framework is given subsequently. We adopt the 3D high-resolution network (HRNet) for learning spatially fine-grained representations and the mixed precision training scheme for memory-efficient training. Extensive experiment results on a publicly available 3D MRI brain dataset show that our proposed model advances the state-of-the-art methods in terms of segmentation performance.

Triplanar ensemble U-Net model for white matter hyperintensities segmentation on MR images

White matter hyperintensities (WMHs) have been associated with various cerebrovascular and neurodegenerative diseases. Reliable quantification of WMHs is essential for understanding their clinical impact in normal and pathological populations. Automated segmentation of WMHs is highly challenging due to heterogeneity in WMH characteristics between deep and periventricular white matter, presence of artefacts and differences in the pathology and demographics of populations. In this work, we propose an ensemble triplanar network that combines the predictions from three different planes of brain MR images to provide an accurate WMH segmentation. In the loss functions the network uses anatomical information regarding WMH spatial distribution in loss functions, to improve the efficiency of segmentation and to overcome the contrast variations between deep and periventricular WMHs. We evaluated our method on 5 datasets, of which 3 are part of a publicly available dataset (training data for MICCAI WMH Segmentation Challenge 2017 - MWSC 2017) consisting of subjects from three different cohorts, and we also submitted our method to MWSC 2017 to be evaluated on the unseen test datasets. On evaluating our method separately in deep and periventricular regions, we observed robust and comparable performance in both regions. Our method performed better than most of the existing methods, including FSL BIANCA, and on par with the top ranking deep learning methods of MWSC 2017.