Enlarged perivascular spaces in brain MRI: Automated quantification in four regions

Effects of one-night partial sleep deprivation on perivascular space volume fraction: Findings from the Stockholm Sleepy Brain Study

Increased waste clearance in the brain is thought to occur most readily during deep sleep (stage N3). Sleep deprivation disrupts time spent in deeper sleep stages, fragmenting the clearance process. Here, we have utilized the publicly available Stockholm Sleepy Brain Study to investigate whether various sleep-related measures are associated with changes in perivascular space (PVS) volume fraction following a late-night short-sleep experiment. The study sample consisted of 60 participants divided into old (65-75 years) and young (20-30 years) age groups. We found that partial sleep deprivation was not significantly associated with major PVS changes. In our centrum semiovale models, we observed an interaction between percentage of total sleep time spent in N3 and sleep deprivation status on PVS volume fraction. In our basal ganglia models, we saw an interaction between N2 (both percentage of total sleep time and absolute time in minutes) and sleep deprivation status. However, the significance of these findings did not survive multiple comparisons corrections. This work highlights the need for future longitudinal studies of PVS and sleep, allowing for quantification of within-subject morphological changes occurring in PVS due to patterns of poor sleep. Our findings here provide insight on the impact that a single night of late-night short-sleep has on the perivascular waste clearance system.

Emerging non-invasive MRI techniques for glymphatic system assessment in neurodegenerative disease

The discovery of the glymphatic system has revolutionized our understanding of brain physiology, particularly in waste clearance and fluid dynamics within the central nervous system. This pathway, essential for nutrient distribution and waste removal, operates predominantly during sleep and has been implicated in neurodegenerative diseases like Alzheimer's and Parkinson's. Recent advances in non-invasive MRI techniques, including diffusion tensor imaging along the perivascular space (DTI-ALPS), perivascular space (PVS) analysis, and free water (FW) indices, have improved our ability to study glymphatic function and its alterations in disease states. This review discusses the glymphatic system's ultrastructure, physiology, and the latest imaging methods to assess this critical pathway. We highlight how these non-invasive MRI techniques can enhance the understanding of glymphatic function in health and disease, with a focus on neurodegenerative conditions. By integrating insights from current research, this review underscores the diagnostic and therapeutic implications of glymphatic dysfunction. Understanding these mechanisms can pave the way for novel strategies to enhance waste clearance and improve neurological health, offering potential benefits for early diagnosis and intervention in neurodegenerative diseases.

A window into the brain: multimodal MRI assessment of vascular cognitive impairment

Vascular cognitive impairment (VCI) encompasses a diverse range of syndromes, including mild cognitive impairment and vascular dementia (VaD), primarily attributed to cerebrovascular lesions and vascular risk factors. Its prevalence ranks second only to Alzheimer's disease (AD) in neuro diseases. The advancement of medical imaging technology, particularly magnetic resonance imaging (MRI), has enabled the early detection of structural, functional, metabolic, and cerebral connectivity alterations in individuals with VCI. This paper examines the utility of multimodal MRI in evaluating structural changes in the cerebral cortex, integrity of white matter fiber tracts, alterations in the blood-brain barrier (BBB) and glymphatic system (GS) activity, alteration of neurovascular coupling function, assessment of brain connectivity, and assessment of metabolic changes in patients with VCI.

The perivascular spaces in young and middle-aged stroke: A single-center analysis integrating clinical and Doppler ultrasound findings

Purpose: This research seeks to investigate correlations between enlarged Perivascular Spaces (PVSs) and clinical/imaging data, such as information obtained through Doppler analysis, in a population with young and middle-age stroke. Materials and methods: We retrospectively reviewed demographics, clinical and MRI data, of 163 patients, with MRI confirmed stroke. All patients underwent ECD TSA (Eco-Color-Doppler of the Supra-aortic Trunks) and TCCD (Transcranial Eco-color-Doppler), to study extra or intracranial stenosis, presence and composition of plaques. Severity of PVS was evaluated on T2-weighted images according to the Potter scale. To identify potential predictors of PVSs, an exploratory backward stepwise ordinal regression model was developed, including clinical and Doppler US variables. Results: In the stepwise ordered logistic regression analysis with PVSs at BG (Basal Ganglia) as the outcome, hemodynamically significant stenosis in any vessel ipsilateral to the ischemic lesion displayed a significant positive association with a higher outcome value. Similar results were observed for ESUS (Embolic Stroke of Undetermined Source). Fibrolipid plaques in any vase exhibited a significant negative association. At MB (Midbrain), male patients and subjects with hypertension exhibited a higher value of PVSs. Dyslipidemia demonstrated a significant negative effect. When PVSs were investigated in the CS (Centrum Semiovale), no statistically significant association with the extent of PVSs emerged. Conclusion: These insights not only enhance our understanding of the role of PVSs in cerebrovascular health in a young and middle-age population but also highlight the potential of PVSs as a biomarker in neuroimaging studies, warranting further research to elucidate their clinical implications and underlying pathophysiological mechanisms.

Enlarged perivascular spaces as vascular health indicators: correlations with clinical and imaging data in young and middle-age stroke patients

PURPOSE

To understand the role of enlarged Perivascular Spaces (PVSs) in a population with young middle age stroke and to identify predictors of PVSs enlargment using clinical and imaging data.

MATERIALS/METHODS

Retrospective revision of demographics, clinical and MRI data, of 163 patients, with MRI confirmed stroke. Ischemic area and WMH were semi-automatically segmented on DWI images and FLAIR images. Severity of PVS was evaluated on T2-weighted images according to the Potter scale. To identify potential predictors of the extent of PVSs, an exploratory backward stepwise ordinal regression model was developed, including all measured variables.

RESULTS

With the extent of PVSs at Basal Ganglia as the dependent variable, the logarithm of WMH demonstrated a significant positive association with the outcome. ESUS exhibited a positive relationship, underscoring its potential role as a predictor of the outcome. In PVSs in Mid Brain, dyslipidemia displayed a significant negative effect, signifying a reduced likelihood of the outcome in its presence. Hypertension emerged as a statistically significant and notably positive predictor of PVSs.

CONCLUSION

Significant associations between PVSs, WMH volume, and vascular features suggest their potential as vascular health indicators. These findings underscore the potentiality of PVSs as a biomarker for further investigation in stroke research. However, given the cross-sectional nature of our data, the relationship between PVS alterations and stroke requires further longitudinal studies to clarify their role and temporal association and eventually refining diagnostic and therapeutic approaches and mitigating stroke risks for younger stroke populations.

Enlarged perivascular spaces in the basal ganglia are associated with arteries not veins

Enlarged perivascular spaces (EPVS) are common in cerebral small vessel disease (CSVD) and have been identified as a marker of dysfunctional brain clearance. However, it remains unknown if the enlargement occurs predominantly around arteries or veins. We combined in vivo ultra-high-resolution MRI and histopathology to investigate the spatial relationship of veins and arteries with EPVS within the basal ganglia (BG). Furthermore, we assessed the relationship between the EPVS and measures of blood-flow (blood-flow velocity, pulsatility index) in the small arteries of the BG. Twenty-four healthy controls, twelve non-CAA CSVD patients, and five probable CAA patients underwent a 3 tesla [T] and 7T MRI-scan, and EPVS, arteries, and veins within the BG were manually segmented. Furthermore, the scans were co-registered. Six autopsy-cases were also assessed. In the BG, EPVS were significantly closer to and overlapped more frequently with arteries than with veins. Histological analysis showed a higher proportion of BG EPVS surrounding arteries than veins. Finally, the pulsatility index of BG arteries correlated with EPVS volume. Our results are in line with previous works and establish a pathophysiological relationship between arteries and EPVS, contributing to elucidating perivascular clearance routes in the human brain.

Cerebral small vessel disease and glymphatic system dysfunction in multiple sclerosis: A narrative review

As the multiple sclerosis (MS) population ages, the prevalence of vascular comorbidities increases, potentially accelerating disease progression and brain atrophy. Recent studies highlight the prevalence of cerebral small vessel disease (CSVD) in MS, suggesting a potential link between vascular comorbidities and accelerated disability. CSVD affects the brain's small vessels, often leading to identifiable markers on MRI such as enlarged perivascular spaces (EPVS). EPVS are increasingly recognized also in MS and have been associated with vascular comorbidities, lower percentage of MS-specific perivenular lesions, brain atrophy and aging. The exact sequence of event leading to MRI visible EPVS is yet to be determined, but an impaired perivascular brain fluid drainage appears a possible physiopathological explanation for EPVS in both CSVD and MS. In this context, a dysfunction of the brain fluid clearance system - also known as "glymphatic system" - appears associated in MS to aging, neuroinflammation, and vascular dysfunction. Advanced imaging techniques show an impaired glymphatic function in both MS and CSVD. Additionally, lifestyle factors such as physical exercise, diet, and sleep quality appear to influence glymphatic function, potentially revealing novel therapeutic strategies to mitigate microangiopathy and neuroinflammation in MS. This review underscores the potential role of glymphatic dysfunction in the complex and not-yet elucidated interplay between neuroinflammation and CSVD in MS.

Perivascular space enlargement accelerates in ageing and Alzheimer's disease pathology: evidence from a three-year longitudinal multicentre study

BACKGROUND

Perivascular space (PVS) enlargement in ageing and Alzheimer's disease (AD) and the drivers of such a structural change in humans require longitudinal investigation. Elucidating the effects of demographic factors, hypertension, cerebrovascular dysfunction, and AD pathology on PVS dynamics could inform the role of PVS in brain health function as well as the complex pathophysiology of AD.

METHODS

We studied PVS in centrum semiovale (CSO) and basal ganglia (BG) computationally over three to four annual visits in 503 participants (255 females; meanage = 70.78 ± 5.78) of the ongoing observational multicentre "DZNE Longitudinal Cognitive Impairment and Dementia Study" (DELCODE) cohort. We analysed data from subjects who were cognitively unimpaired (n = 401), had amnestic mild cognitive impairment (n = 71), or had AD (n = 31). We used linear mixed-effects modelling to test for changes of PVS volumes in relation to cross-sectional and longitudinal age, as well as sex, years of education, hypertension, white matter hyperintensities, AD diagnosis, and cerebrospinal-fluid-derived amyloid (A) and tau (T) status (available for 46.71%; A-T-/A + T-/A + T + n = 143/48/39).

RESULTS

PVS volumes increased significantly over follow-ups (CSO: B = 0.03 [0.02, 0.05], p < 0.001; BG: B = 0.05 [0.03, 0.07], p < 0.001). PVS enlargement rates varied substantially across subjects and depended on the participant's age, white matter hyperintensities volumes, and amyloid and tau status. PVS volumes were higher across elderly participants, regardless of region of interest (CSO: B = 0.12 [0.02, 0.21], p = 0.017; BG: B = 0.19 [0.09, 0.28], p < 0.001). Faster BG-PVS enlargement related to lower baseline white matter hyperintensities volumes (ρspearman = -0.17, pFDR = 0.001) and was more pronounced in individuals who presented with combined amyloid and tau positivity versus negativity (A + T +  > A-T-, pFDR = 0.004) or who were amyloid positive but tau negative (A + T +  > A + T-, pFDR = 0.07). CSO-PVS volumes increased at a faster rate with amyloid positivity as compared to amyloid negativity (A + T-/A + T +  > A-T-, pFDR = 0.021).

CONCLUSION

Our longitudinal evidence supports the relevance of PVS enlargement in presumably healthy ageing as well as in AD pathology. We further discuss the region-specific involvement of white matter hyperintensities and neurotoxic waste accumulation in PVS enlargement and the possibility of additional factors contributing to PVS progression. A comprehensive understanding of PVS dynamics could facilitate the understanding of pathological cascades and might inform targeted treatment strategies.

TRIAL REGISTRATION

German Clinical Trials Register DRKS00007966. Registered 04.05.2015 - retrospectively registered, https://drks.de/search/en/trial/DRKS00007966 .

Effects of one-night partial sleep deprivation on perivascular space volume fraction: Findings from the Stockholm Sleepy Brain Study

Increased waste clearance in the brain is thought to occur most readily during late-stage sleep (stage N3). Sleep deprivation disrupts time spent in deeper sleep stages, fragmenting the clearance process. Here, we have utilized the publicly available Stockholm Sleepy Brain Study to investigate whether various sleep-related measures are associated with changes in perivascular space (PVS) volume fraction following a late-night short-sleep experiment. Our sample consisted of 60 participants divided into old (65-75 years) and young (20-30 years) age groups. We found that partial sleep deprivation was not significantly associated with major PVS changes. In the centrum semiovale, we observed an interaction between percentage of total sleep time spent in N3 and sleep deprivation status on PVS volume fraction. In the basal ganglia, we saw an interaction between N2 (both percentage of total sleep time and absolute time in minutes) and sleep deprivation status. However, the significance of these findings did not survive multiple comparisons corrections. This work highlights the need for future longitudinal studies of PVS and sleep, allowing for quantification of within-subject morphological changes occurring in PVS due to patterns of poor sleep. Our findings here provide insight on the impacts that a single night of late-night short-sleep has on the perivascular waste clearance system.

A bibliometric analysis of cerebral small vessel disease

Background

Cerebral small vessel disease (CSVD) is a significant contributor to both stroke and dementia. While numerous studies on CSVD have been published, herein, we have conducted a bibliometric examination of the literature on CSVD, revealing its hot spots and emerging patterns.

Methods

We used the Web of Science Core Collection as our primary database and conducted a literature search from January 2008 to January 2023. CiteSpace, VOSviewer, online bibliometric platform, and R-bibliometrix were employed to conduct bibliometric analysis and network visualization, including the number of publications, countries, institutions, journals, citations, authors, references, and keywords.

Results

A total of 4891 publications on CSVD were published in 790 journals by 19,066 authors at 3,862 institutions from 84 countries. The United States produced the most written works and had a significant impact in this field of study. The University of Edinburgh had the highest publication count overall. The journal with the most publications and co-citations was Stroke. Wardlaw, Joanna was the most prolific author and commonly cited in the field. The current areas of research interest revolved around "MRI segmentation" and "Enlarged perivascular spaces in the basal ganglia."

Conclusion

We conducted a bibliometric analysis to examine the advancements, focal points, and cutting-edge areas in the field of CSVD to reveal potential future research opportunities. Research on CSVD is currently rapidly advancing, with a consistent rise in publications on the topic since 2008. At the same time, we identified leading countries, institutions, and leading scholars in the field and analyzed journals and representative literature. Keyword co-occurrence analysis and burst graph emergence detection identified MRI segmentation and Basal ganglia enlarged perivascular spaces as the most recent areas of research interest.

Perivascular spaces and where to find them - MR imaging and evaluation methods

BACKGROUND

Perivascular spaces (synonym: Virchow-Robin spaces) were first described over 150 years ago. They are defined as the fluid-filled spaces surrounding the small penetrating cerebral vessels. They gained growing scientific interest especially with the postulation of the so-called glymphatic system and their possible role in neurodegenerative and neuroinflammatory diseases.

METHODS

PubMed was used for a systematic search with a focus on literature regarding MRI imaging and evaluation methods of perivascular spaces. Studies on human in-vivo imaging were included with a focus on studies involving healthy populations. No time frame was set. The nomenclature in the literature is very heterogeneous with terms like "large", "dilated", "enlarged" perivascular spaces whereas borders and definitions often remain unclear. This work generally talks about perivascular spaces.

RESULTS

This review article discusses the morphologic MRI characteristics in different sequences. With the continual improvement of image quality, more and tinier structures can be depicted in detail. Visual analysis and semi or fully automated segmentation methods are briefly discussed.

CONCLUSION

If they are looked for, perivascular spaces are apparent in basically every cranial MRI examination. Their physiologic or pathologic value is still under debate.

KEY POINTS

· Perivascular spaces can be seen in basically every cranial MRI examination.. · Primarily T2-weighend sequences are used for visual analysis. Additional sequences are helpful for distinction from their differential diagnoses.. · There are promising approaches for the semi or fully automated segmentation of perivascular spaces with the possibility to collect more quantitative parameters..

CITATION FORMAT

· Seehafer S, Larsen N, Aludin S et al. Perivascular spaces and where to find them - MRI imaging and evaluation methods. Fortschr Röntgenstr 2024; 196: 1029 - 1036.

A Survey on Explainable Artificial Intelligence (XAI) Techniques for Visualizing Deep Learning Models in Medical Imaging

The combination of medical imaging and deep learning has significantly improved diagnostic and prognostic capabilities in the healthcare domain. Nevertheless, the inherent complexity of deep learning models poses challenges in understanding their decision-making processes. Interpretability and visualization techniques have emerged as crucial tools to unravel the black-box nature of these models, providing insights into their inner workings and enhancing trust in their predictions. This survey paper comprehensively examines various interpretation and visualization techniques applied to deep learning models in medical imaging. The paper reviews methodologies, discusses their applications, and evaluates their effectiveness in enhancing the interpretability, reliability, and clinical relevance of deep learning models in medical image analysis.

Development and validation of a perivascular space segmentation method in multi-center datasets

BACKGROUND

Perivascular spaces (PVS) visible on magnetic resonance imaging (MRI) are significant markers associated with various neurological diseases. Although quantitative analysis of PVS may enhance sensitivity and improve consistency across studies, the field lacks a universally validated method for analyzing images from multi-center studies.

METHODS

We annotated PVS on multi-center 3D T1-weighted (T1w) images acquired using scanners from three major vendors (Siemens, General Electric, and Philips). A neural network, mcPVS-Net (multi-center PVS segmentation network), was trained using data from 40 subjects and then tested in a separate cohort of 15 subjects. We assessed segmentation accuracy against ground truth masks tailored for each scanner vendor. Additionally, we evaluated the agreement between segmented PVS volumes and visual scores for each scanner. We also explored correlations between PVS volumes and various clinical factors such as age, hypertension, and white matter hyperintensities (WMH) in a larger sample of 1020 subjects. Furthermore, mcPVS-Net was applied to a new dataset comprising both T1w and T2-weighted (T2w) images from a United Imaging scanner to investigate if PVS volumes could discriminate between subjects with differing visual scores. We also compared the mcPVS-Net with a previously published method that segments PVS from T1 images.

RESULTS

In the test dataset, mcPVS-Net achieved a mean DICE coefficient of 0.80, with an average Precision of 0.81 and Recall of 0.79, indicating good specificity and sensitivity. The segmented PVS volumes were significantly associated with visual scores in both the basal ganglia (r = 0.541, p < 0.001) and white matter regions (r = 0.706, p < 0.001), and PVS volumes were significantly different among subjects with varying visual scores. Segmentation performance was consistent across different scanner vendors. PVS volumes exhibited significant associations with age, hypertension, and WMH. In the United Imaging scanner dataset, PVS volumes showed good associations with PVS visual scores evaluated on either T1w or T2w images. Compared to a previously published method, mcPVS-Net showed a higher accuracy and improved PVS segmentation in the basal ganglia region.

CONCLUSION

The mcPVS-Net demonstrated good accuracy for segmenting PVS from 3D T1w images. It may serve as a useful tool for future PVS research.

Machine Learning-Based Perivascular Space Volumetry in Alzheimer Disease

OBJECTIVES

Impaired perivascular clearance has been suggested as a contributing factor to the pathogenesis of Alzheimer disease (AD). However, it remains unresolved when the anatomy of the perivascular space (PVS) is altered during AD progression. Therefore, this study investigates the association between PVS volume and AD progression in cognitively unimpaired (CU) individuals, both with and without subjective cognitive decline (SCD), and in those clinically diagnosed with mild cognitive impairment (MCI) or mild AD.

MATERIALS AND METHODS

A convolutional neural network was trained using manually corrected, filter-based segmentations (n = 1000) to automatically segment the PVS in the centrum semiovale from interpolated, coronal T2-weighted magnetic resonance imaging scans (n = 894). These scans were sourced from the national German Center for Neurodegenerative Diseases Longitudinal Cognitive Impairment and Dementia Study. Convolutional neural network-based segmentations and those performed by a human rater were compared in terms of segmentation volume, identified PVS clusters, as well as Dice score. The comparison revealed good segmentation quality (Pearson correlation coefficient r = 0.70 with P < 0.0001 for PVS volume, detection rate in cluster analysis = 84.3%, and Dice score = 59.0%). Subsequent multivariate linear regression analysis, adjusted for participants' age, was performed to correlate PVS volume with clinical diagnoses, disease progression, cerebrospinal fluid biomarkers, lifestyle factors, and cognitive function. Cognitive function was assessed using the Mini-Mental State Examination, the Comprehensive Neuropsychological Test Battery, and the Cognitive Subscale of the 13-Item Alzheimer's Disease Assessment Scale.

RESULTS

Multivariate analysis, adjusted for age, revealed that participants with AD and MCI, but not those with SCD, had significantly higher PVS volumes compared with CU participants without SCD ( P = 0.001 for each group). Furthermore, CU participants who developed incident MCI within 4.5 years after the baseline assessment showed significantly higher PVS volumes at baseline compared with those who did not progress to MCI ( P = 0.03). Cognitive function was negatively correlated with PVS volume across all participant groups ( P ≤ 0.005 for each). No significant correlation was found between PVS volume and any of the following parameters: cerebrospinal fluid biomarkers, sleep quality, body mass index, nicotine consumption, or alcohol abuse.

CONCLUSIONS

The very early changes of PVS volume may suggest that alterations in PVS function are involved in the pathophysiology of AD. Overall, the volumetric assessment of centrum semiovale PVS represents a very early imaging biomarker for AD.

Systematic review and meta-analysis of automated methods for quantifying enlarged perivascular spaces in the brain

Research into magnetic resonance imaging (MRI)-visible perivascular spaces (PVS) has recently increased, as results from studies in different diseases and populations are cementing their association with sleep, disease phenotypes, and overall health indicators. With the establishment of worldwide consortia and the availability of large databases, computational methods that allow to automatically process all this wealth of information are becoming increasingly relevant. Several computational approaches have been proposed to assess PVS from MRI, and efforts have been made to summarise and appraise the most widely applied ones. We systematically reviewed and meta-analysed all publications available up to September 2023 describing the development, improvement, or application of computational PVS quantification methods from MRI. We analysed 67 approaches and 60 applications of their implementation, from 112 publications. The two most widely applied were the use of a morphological filter to enhance PVS-like structures, with Frangi being the choice preferred by most, and the use of a U-Net configuration with or without residual connections. Older adults or population studies comprising adults from 18 years old onwards were, overall, more frequent than studies using clinical samples. PVS were mainly assessed from T2-weighted MRI acquired in 1.5T and/or 3T scanners, although combinations using it with T1-weighted and FLAIR images were also abundant. Common associations researched included age, sex, hypertension, diabetes, white matter hyperintensities, sleep and cognition, with occupation-related, ethnicity, and genetic/hereditable traits being also explored. Despite promising improvements to overcome barriers such as noise and differentiation from other confounds, a need for joined efforts for a wider testing and increasing availability of the most promising methods is now paramount.

Enlarged perivascular spaces are associated with brain microangiopathy and aging in multiple sclerosis

BACKGROUND

Growing evidence links brain-MRI enlarged perivascular spaces (EPVS) and multiple sclerosis (MS), but their role remains unclear.

OBJECTIVE

This study aimed to investigate the cross-sectional associations of EPVS with several neuroinflammatory and neurodegenerative features in a large multicentric-MS cohort.

METHODS

In total, 207 patients underwent 3T axial-T2-weighted brain-MRI for EPVS assessment (EPVS dichotomized into high/low according to ⩾ 2/< 2 rating categories). MRI biomarkers included brain-predicted age and brain-predicted age difference (brain-PAD), central vein sign (CVS)-positive lesion percentage (CVS%), paramagnetic rim and cortical lesions, T2-lesion load, and brain volumetry. The variable relative importance for EPVS-category prediction was explored using a classification random forest approach.

RESULTS

High EPVS patients were older (49 vs 44 years, p = 0.003), had ⩾ 1 vascular risk factors (VRFs; p = 0.005), lower CVS% (67% vs 78%, p < 0.001), reduced brain volumes (whole brain: 0.63 vs 0.73, p = 0.01; gray matter: 0.36 vs 0.40; p = 0.002), and older brain-predicted age (58 vs 50 years, p < 0.001). No differences were found for neuroinflammatory markers. After adjusting for age and VFRs (multivariate analyses), the high EPVS category correlated with lower CVS% (odds ratio (OR) = 0.98, 95% confidence interval (CI) = 0.96-0.99; p = 0.02), lower whole brain (OR = 0.01, 95% CI = 0.0003-0.5; p = 0.02), gray matter (OR = 0.0004, 95% CI = 0.0000004-0.4; p = 0.03) volumes, and higher brain-PAD (OR = 1.05, 95% CI = 1.01-1.09; p = 0.02). Random forest identified brain-PAD as the most important predictor of high EPVS.

CONCLUSION

EPVS in MS likely reflect microangiopathic disease rather than neuroinflammation, potentially contributing to accelerated neurodegeneration.

Application of artificial intelligence-based magnetic resonance imaging in diagnosis of cerebral small vessel disease

Cerebral small vessel disease (CSVD) is an important cause of stroke, cognitive impairment, and other diseases, and its early quantitative evaluation can significantly improve patient prognosis. Magnetic resonance imaging (MRI) is an important method to evaluate the occurrence, development, and severity of CSVD. However, the diagnostic process lacks quantitative evaluation criteria and is limited by experience, which may easily lead to missed diagnoses and misdiagnoses. With the development of artificial intelligence technology based on deep learning, the extraction of high-dimensional features in imaging can assist doctors in clinical decision-making, and it has been widely used in brain function and mental disorders, and cardiovascular and cerebrovascular diseases. This paper summarizes the global research results in recent years and briefly describes the application of deep learning in evaluating CSVD signs in MRI imaging, including recent small subcortical infarcts, lacunes of presumed vascular origin, vascular white matter hyperintensity, enlarged perivascular spaces, cerebral microbleeds, brain atrophy, cortical superficial siderosis, and cortical cerebral microinfarct.

An end-to-end infant brain parcellation pipeline

Objective

Accurate infant brain parcellation is crucial for understanding early brain development; however, it is challenging due to the inherent low tissue contrast, high noise, and severe partial volume effects in infant magnetic resonance images (MRIs). The aim of this study was to develop an end-to-end pipeline that enabled accurate parcellation of infant brain MRIs.

Methods

We proposed an end-to-end pipeline that employs a two-stage global-to-local approach for accurate parcellation of infant brain MRIs. Specifically, in the global regions of interest (ROIs) localization stage, a combination of transformer and convolution operations was employed to capture both global spatial features and fine texture features, enabling an approximate localization of the ROIs across the whole brain. In the local ROIs refinement stage, leveraging the position priors from the first stage along with the raw MRIs, the boundaries o the ROIs are refined for a more accurate parcellation.

Results

We utilized the Dice ratio to evaluate the accuracy of parcellation results. Results on 263 subjects from National Database for Autism Research (NDAR), Baby Connectome Project (BCP) and Cross-site datasets demonstrated the better accuracy and robustness of our method than other competing methods.

Conclusion

Our end-to-end pipeline may be capable of accurately parcellating 6-month-old infant brain MRIs.

Detectability and accuracy of computational measurements of in-silico and physical representations of enlarged perivascular spaces from magnetic resonance images

BACKGROUND

Magnetic Resonance Imaging (MRI) visible perivascular spaces (PVS) have been associated with age, decline in cognitive abilities, interrupted sleep, and markers of small vessel disease. But the limits of validity of their quantification have not been established.

NEW METHOD

We use a purpose-built digital reference object to construct an in-silico phantom for addressing this need, and validate it using a physical phantom. We use cylinders of different sizes as models for PVS. We also evaluate the influence of 'PVS' orientation, and different sets of parameters of the two vesselness filters that have been used for enhancing tubular structures, namely Frangi and RORPO filters, in the measurements' accuracy.

RESULTS

PVS measurements in MRI are only a proxy of their true dimensions, as the boundaries of their representation are consistently overestimated. The success in the use of the Frangi filter relies on a careful tuning of several parameters. Alpha= 0.5, beta= 0.5 and c= 500 yielded the best results. RORPO does not have these requirements and allows detecting smaller cylinders in their entirety more consistently in the absence of noise and confounding artefacts. The Frangi filter seems to be best suited for voxel sizes equal or larger than 0.4 mm-isotropic and cylinders larger than 1 mm diameter and 2 mm length. 'PVS' orientation did not affect measurements in data with isotropic voxels.

COMPARISON WITH EXISTENT METHODS

Does not apply.

CONCLUSIONS

The in-silico and physical phantoms presented are useful for establishing the validity of quantification methods of tubular small structures.

Deep learning applications in vascular dementia using neuroimaging

PURPOSE OF REVIEW

Vascular dementia (VaD) is the second common cause of dementia after Alzheimer's disease, and deep learning has emerged as a critical tool in dementia research. The aim of this article is to highlight the current deep learning applications in VaD-related imaging biomarkers and diagnosis.

RECENT FINDINGS

The main deep learning technology applied in VaD using neuroimaging data is convolutional neural networks (CNN). CNN models have been widely used for lesion detection and segmentation, such as white matter hyperintensities (WMH), cerebral microbleeds (CMBs), perivascular spaces (PVS), lacunes, cortical superficial siderosis, and brain atrophy. Applications in VaD subtypes classification also showed excellent results. CNN-based deep learning models have potential for further diagnosis and prognosis of VaD.

SUMMARY

Deep learning neural networks with neuroimaging data in VaD research represent significant promise for advancing early diagnosis and treatment strategies. Ongoing research and collaboration between clinicians, data scientists, and neuroimaging experts are essential to address challenges and unlock the full potential of deep learning in VaD diagnosis and management.

Connecting the Dots: The Cerebral Lymphatic System as a Bridge Between the Central Nervous System and Peripheral System in Health and Disease

As a recently discovered waste removal system in the brain, cerebral lymphatic system is thought to play an important role in regulating the homeostasis of the central nervous system. Currently, more and more attention is being focused on the cerebral lymphatic system. Further understanding of the structural and functional characteristics of cerebral lymphatic system is essential to better understand the pathogenesis of diseases and to explore therapeutic approaches. In this review, we summarize the structural components and functional characteristics of cerebral lymphatic system. More importantly, it is closely associated with peripheral system diseases in the gastrointestinal tract, liver, and kidney. However, there is still a gap in the study of the cerebral lymphatic system. However, we believe that it is a critical mediator of the interactions between the central nervous system and the peripheral system.

Overview of the Current Knowledge and Conventional MRI Characteristics of Peri- and Para-Vascular Spaces

Brain spaces around (perivascular spaces) and alongside (paravascular or Virchow-Robin spaces) vessels have gained significant attention in recent years due to the advancements of in vivo imaging tools and to their crucial role in maintaining brain health, contributing to the anatomic foundation of the glymphatic system. In fact, it is widely accepted that peri- and para-vascular spaces function as waste clearance pathways for the brain for materials such as ß-amyloid by allowing exchange between cerebrospinal fluid and interstitial fluid. Visible brain spaces on magnetic resonance imaging are often a normal finding, but they have also been associated with a wide range of neurological and systemic conditions, suggesting their potential as early indicators of intracranial pressure and neurofluid imbalance. Nonetheless, several aspects of these spaces are still controversial. This article offers an overview of the current knowledge and magnetic resonance imaging characteristics of peri- and para-vascular spaces, which can help in daily clinical practice image description and interpretation. This paper is organized into different sections, including the microscopic anatomy of peri- and para-vascular spaces, their associations with pathological and physiological events, and their differential diagnosis.

Dilated perivascular spaces and steno-occlusive changes in children and adults with moyamoya disease

BACKGROUND

Dilated perivascular spaces (DPVS), known as one of imaging markers in cerebral small vessel disease, may be found in patients with moyamoya disease (MMD). However, little is known about DPVS in MMD. The purpose of this study was to investigate the distribution pattern of dPVS in children and adults with MMD and determine whether it is related to steno-occlusive changes of MMD.

METHODS

DPVS was scored in basal ganglia (BG) and white matter (WM) on T2-weighted imaging, using a validated 4-point semi-quantitative score. The degree of dPVS was classified as high (score > 2) or low (score ≤ 2) grade. The steno-occlusive changes on MR angiography (MRA) was scored using a validated MRA grading. Asymmetry of DPVS and MRA grading was defined as a difference of 1 grade or higher between hemispheres.

RESULTS

Fifty-one patients with MMD (mean age 24.9 ± 21.1 years) were included. Forty-five (88.2%) patients had high WM-DPVS grade (degree 3 or 4). BG-DPVS was found in 72.5% of all patients and all were low grade (degree 1 or 2). The distribution patterns of DPVS degree in BG (P = 1.000) and WM (P = 0.767) were not different between child and adult groups. The asymmetry of WM-DPVS (26%) and MRA grade (42%) were significantly correlated to each other (Kendall's tau-b = 0.604, P < 0.001).

CONCLUSIONS

DPVS of high grade in MMD is predominantly found in WM, which was not different between children and adults. The correlation between asymmetry of WM-DPVS degree and MRA grade suggests that weak cerebral artery pulsation due to steno-occlusive changes may affect WM-DPVS in MMD.

Correlation of glymphatic system abnormalities with Parkinson's disease progression: a clinical study based on non-invasive fMRI

BACKGROUND

The glymphatic system is reportedly involved in Parkinson's disease (PD). Based on previous studies, we aimed to confirm the correlation between the glymphatic system and PD progression by combining two imaging parameters, diffusion tensor image analysis along the perivascular space (DTI-ALPS), and enlarged perivascular spaces (EPVS).

METHODS

Fifty-one PD patients and fifty healthy control (HC) were included. Based on the Hoehn-Yahr scale, the PD group was divided into early-stage and medium-to late-stage. All PD patients were scored using the Unified PD Rating Scale (UPDRS). We assessed the DTI-ALPS indices in the bilateral hemispheres and EPVS numbers in bilateral centrum semiovale (CSO), basal ganglia (BG), and midbrain.

RESULTS

The DTI-ALPS indices were significantly lower bilaterally in PD patients than in the HC group, and EPVS numbers in any of the bilateral CSO, BG, and midbrain were significantly higher, especially for the medium- to late-stage group and the BG region. In PD patients, the DTI-ALPS index was significantly negatively correlated with age, while the BG-EPVS numbers were significantly positively correlated with age. Furthermore, the DTI-ALPS index was negatively correlated with UPDRS II and III scores, while the BG-EPVS numbers were positively correlated with UPDRS II and III scores. Similarly, the correlation was more pronounced in the medium- to late-stage group.

CONCLUSION

The DTI-ALPS index and EPVS numbers (especially in the BG region) are closely related to age and PD progression and can serve as non-invasive assessments for glymphatic dysfunction and its interventions in clinical studies.

Enlarged Perivascular Space and Index for Diffusivity Along the Perivascular Space as Emerging Neuroimaging Biomarkers of Neurological Diseases

The existence of lymphatic vessels or similar clearance systems in the central nervous system (CNS) that transport nutrients and remove cellular waste is a neuroscientific question of great significance. As the brain is the most metabolically active organ in the body, there is likely to be a potential correlation between its clearance system and the pathological state of the CNS. Until recently the successive discoveries of the glymphatic system and the meningeal lymphatics solved this puzzle. This article reviews the basic anatomy and physiology of the glymphatic system. Imaging techniques to visualize the function of the glymphatic system mainly including post-contrast imaging techniques, indirect lymphatic assessment by detecting increased perivascular space, and diffusion tensor image analysis along the perivascular space (DTI-ALPS) are discussed. The pathological link between glymphatic system dysfunction and neurological disorders is the key point, focusing on the enlarged perivascular space (EPVS) and the index of diffusivity along the perivascular space (ALPS index), which may represent the activity of the glymphatic system as possible clinical neuroimaging biomarkers of neurological disorders. The pathological link between glymphatic system dysfunction and neurological disorders is the key point, focusing on the enlarged perivascular space (EPVS) and the index for of diffusivity along the perivascular space (ALPS index), which may represent the activity of the glymphatic system as possible clinical neuroimaging biomarkers of neurological disorders.

Detection and quantification of breast arterial calcifications on mammograms: a deep learning approach

OBJECTIVE

Breast arterial calcifications (BAC) are a sex-specific cardiovascular disease biomarker that might improve cardiovascular risk stratification in women. We implemented a deep convolutional neural network for automatic BAC detection and quantification.

METHODS

In this retrospective study, four readers labelled four-view mammograms as BAC positive (BAC+) or BAC negative (BAC-) at image level. Starting from a pretrained VGG16 model, we trained a convolutional neural network to discriminate BAC+ and BAC- mammograms. Accuracy, F1 score, and area under the receiver operating characteristic curve (AUC-ROC) were used to assess the diagnostic performance. Predictions of calcified areas were generated using the generalized gradient-weighted class activation mapping (Grad-CAM++) method, and their correlation with manual measurement of BAC length in a subset of cases was assessed using Spearman ρ.

RESULTS

A total 1493 women (198 BAC+) with a median age of 59 years (interquartile range 52-68) were included and partitioned in a training set of 410 cases (1640 views, 398 BAC+), validation set of 222 cases (888 views, 89 BAC+), and test set of 229 cases (916 views, 94 BAC+). The accuracy, F1 score, and AUC-ROC were 0.94, 0.86, and 0.98 in the training set; 0.96, 0.74, and 0.96 in the validation set; and 0.97, 0.80, and 0.95 in the test set, respectively. In 112 analyzed views, the Grad-CAM++ predictions displayed a strong correlation with BAC measured length (ρ = 0.88, p < 0.001).

CONCLUSION

Our model showed promising performances in BAC detection and in quantification of BAC burden, showing a strong correlation with manual measurements.

CLINICAL RELEVANCE STATEMENT

Integrating our model to clinical practice could improve BAC reporting without increasing clinical workload, facilitating large-scale studies on the impact of BAC as a biomarker of cardiovascular risk, raising awareness on women's cardiovascular health, and leveraging mammographic screening.

KEY POINTS

• We implemented a deep convolutional neural network (CNN) for BAC detection and quantification. • Our CNN had an area under the receiving operator curve of 0.95 for BAC detection in the test set composed of 916 views, 94 of which were BAC+ . • Furthermore, our CNN showed a strong correlation with manual BAC measurements (ρ = 0.88) in a set of 112 views.

Explainable AI in medical imaging: An overview for clinical practitioners – Saliency-based XAI approaches

Since recent achievements of Artificial Intelligence (AI) have proven significant success and promising results throughout many fields of application during the last decade, AI has also become an essential part of medical research. The improving data availability, coupled with advances in high-performance computing and innovative algorithms, has increased AI's potential in various aspects. Because AI rapidly reshapes research and promotes the development of personalized clinical care, alongside its implementation arises an urgent need for a deep understanding of its inner workings, especially in high-stake domains. However, such systems can be highly complex and opaque, limiting the possibility of an immediate understanding of the system's decisions. Regarding the medical field, a high impact is attributed to these decisions as physicians and patients can only fully trust AI systems when reasonably communicating the origin of their results, simultaneously enabling the identification of errors and biases. Explainable AI (XAI), becoming an increasingly important field of research in recent years, promotes the formulation of explainability methods and provides a rationale allowing users to comprehend the results generated by AI systems. In this paper, we investigate the application of XAI in medical imaging, addressing a broad audience, especially healthcare professionals. The content focuses on definitions and taxonomies, standard methods and approaches, advantages, limitations, and examples representing the current state of research regarding XAI in medical imaging. This paper focuses on saliency-based XAI methods, where the explanation can be provided directly on the input data (image) and which naturally are of special importance in medical imaging.

Recent Advances in Explainable Artificial Intelligence for Magnetic Resonance Imaging

Advances in artificial intelligence (AI), especially deep learning (DL), have facilitated magnetic resonance imaging (MRI) data analysis, enabling AI-assisted medical image diagnoses and prognoses. However, most of the DL models are considered as "black boxes". There is an unmet need to demystify DL models so domain experts can trust these high-performance DL models. This has resulted in a sub-domain of AI research called explainable artificial intelligence (XAI). In the last decade, many experts have dedicated their efforts to developing novel XAI methods that are competent at visualizing and explaining the logic behind data-driven DL models. However, XAI techniques are still in their infancy for medical MRI image analysis. This study aims to outline the XAI applications that are able to interpret DL models for MRI data analysis. We first introduce several common MRI data modalities. Then, a brief history of DL models is discussed. Next, we highlight XAI frameworks and elaborate on the principles of multiple popular XAI methods. Moreover, studies on XAI applications in MRI image analysis are reviewed across the tissues/organs of the human body. A quantitative analysis is conducted to reveal the insights of MRI researchers on these XAI techniques. Finally, evaluations of XAI methods are discussed. This survey presents recent advances in the XAI domain for explaining the DL models that have been utilized in MRI applications.

Assessment of Risk Factors and Clinical Importance of Enlarged Perivascular Spaces by Whole-Brain Investigation in the Multi-Ethnic Study of Atherosclerosis

Importance

Enlarged perivascular spaces (ePVSs) have been associated with cerebral small-vessel disease (cSVD). Although their etiology may differ based on brain location, study of ePVSs has been limited to specific brain regions; therefore, their risk factors and significance remain uncertain.

Objective

Toperform a whole-brain investigation of ePVSs in a large community-based cohort.

Design, Setting, and Participants

This cross-sectional study analyzed data from the Atrial Fibrillation substudy of the population-based Multi-Ethnic Study of Atherosclerosis. Demographic, vascular risk, and cardiovascular disease data were collected from September 2016 to May 2018. Brain magnetic resonance imaging was performed from March 2018 to July 2019. The reported analysis was conducted between August and October 2022. A total of 1026 participants with available brain magnetic resonance imaging data and complete information on demographic characteristics and vascular risk factors were included.

Main Outcomes and Measures

Enlarged perivascular spaces were quantified using a fully automated deep learning algorithm. Quantified ePVS volumes were grouped into 6 anatomic locations: basal ganglia, thalamus, brainstem, frontoparietal, insular, and temporal regions, and were normalized for the respective regional volumes. The association of normalized regional ePVS volumes with demographic characteristics, vascular risk factors, neuroimaging indices, and prevalent cardiovascular disease was explored using generalized linear models.

Results

In the 1026 participants, mean (SD) age was 72 (8) years; 541 (53%) of the participants were women. Basal ganglia ePVS volume was positively associated with age (β = 3.59 × 10-3; 95% CI, 2.80 × 10-3 to 4.39 × 10-3), systolic blood pressure (β = 8.35 × 10-4; 95% CI, 5.19 × 10-4 to 1.15 × 10-3), use of antihypertensives (β = 3.29 × 10-2; 95% CI, 1.92 × 10-2 to 4.67 × 10-2), and negatively associated with Black race (β = -3.34 × 10-2; 95% CI, -5.08 × 10-2 to -1.59 × 10-2). Thalamic ePVS volume was positively associated with age (β = 5.57 × 10-4; 95% CI, 2.19 × 10-4 to 8.95 × 10-4) and use of antihypertensives (β = 1.19 × 10-2; 95% CI, 6.02 × 10-3 to 1.77 × 10-2). Insular region ePVS volume was positively associated with age (β = 1.18 × 10-3; 95% CI, 7.98 × 10-4 to 1.55 × 10-3). Brainstem ePVS volume was smaller in Black than in White participants (β = -5.34 × 10-3; 95% CI, -8.26 × 10-3 to -2.41 × 10-3). Frontoparietal ePVS volume was positively associated with systolic blood pressure (β = 1.14 × 10-4; 95% CI, 3.38 × 10-5 to 1.95 × 10-4) and negatively associated with age (β = -3.38 × 10-4; 95% CI, -5.40 × 10-4 to -1.36 × 10-4). Temporal region ePVS volume was negatively associated with age (β = -1.61 × 10-2; 95% CI, -2.14 × 10-2 to -1.09 × 10-2), as well as Chinese American (β = -2.35 × 10-1; 95% CI, -3.83 × 10-1 to -8.74 × 10-2) and Hispanic ethnicities (β = -1.73 × 10-1; 95% CI, -2.96 × 10-1 to -4.99 × 10-2).

Conclusions and Relevance

In this cross-sectional study of ePVSs in the whole brain, increased ePVS burden in the basal ganglia and thalamus was a surrogate marker for underlying cSVD, highlighting the clinical importance of ePVSs in these locations.

Deep Learning Based Detection of Enlarged Perivascular Spaces on Brain MRI

Deep learning has been demonstrated effective in many neuroimaging applications. However, in many scenarios, the number of imaging sequences capturing information related to small vessel disease lesions is insufficient to support data-driven techniques. Additionally, cohort-based studies may not always have the optimal or essential imaging sequences for accurate lesion detection. Therefore, it is necessary to determine which imaging sequences are crucial for precise detection. This study introduces a deep learning framework to detect enlarged perivascular spaces (ePVS) and aims to find the optimal combination of MRI sequences for deep learning-based quantification. We implemented an effective lightweight U-Net adapted for ePVS detection and comprehensively investigated different combinations of information from SWI, FLAIR, T1-weighted (T1w), and T2-weighted (T2w) MRI sequences. The experimental results showed that T2w MRI is the most important for accurate ePVS detection, and the incorporation of SWI, FLAIR and T1w MRI in the deep neural network had minor improvements in accuracy and resulted in the highest sensitivity and precision (sensitivity =0.82, precision =0.83). The proposed method achieved comparable accuracy at a minimal time cost compared to manual reading. The proposed automated pipeline enables robust and time-efficient readings of ePVS from MR scans and demonstrates the importance of T2w MRI for ePVS detection and the potential benefits of using multimodal images. Furthermore, the model provides whole-brain maps of ePVS, enabling a better understanding of their clinical correlates compared to the clinical rating methods within only a couple of brain regions.

Role of Enlarged Perivascular Space in the Temporal Lobe in Cerebral Amyloidosis

OBJECTIVE

Although growing evidence suggests that perivascular space (PVS) serves as a clearance route for amyloid and tau, the association between enlarged PVS (EPVS) and Alzheimer disease is highly inconsistent across studies. As the conventional visual rating systems for EPVS were insufficient to predict amyloid/tau/neurodegeneration (A/T/N) status, we developed a new rating scale for EPVS located in the temporal lobe (T-EPVS).

METHODS

EPVS located in the basal ganglia (BG-EPVS), centrum semiovale (CS-EPVS), and T-EPVS was visually rated in 272 individuals (healthy controls, n = 96; mild cognitive impairment, n = 106; dementia, n = 70) who underwent structural magnetic resonance imaging (MRI) and dual positron emission tomography scans (¹⁸ F-flortaucipir and ¹⁸ F-florbetaben). T-EPVS and BG-EPVS were defined as high degree when the counts in any hemisphere were >10, and the CS-EPVS cutoff was >20. Logistic regression models were constructed to investigate whether the regional EPVS burden was predictive of A/T/N status. The derived models were externally validated in a temporal validation cohort (n = 195) that underwent MRI studies using a different scanner.

RESULTS

Compared with those with low-degree T-EPVS (23/136, 16.9%), individuals with high-degree T-EPVS/CS-EPVS but low-degree BG-EPVS were more likely to exhibit amyloid positivity (46/56, 82.1%). High-degree T-EPVS burden (odds ratio [OR] = 7.251, 95% confidence interval [CI] = 3.296-15.952) and low-degree BG-EPVS (OR = 0.241, 95% CI = 0.109-0.530) were predictive of amyloid positivity. Although high-degree T-EPVS was associated with tau positivity, the association was no longer significant after adjusting for amyloid and neurodegeneration status.

INTERPRETATION

Investigating the burden and topographic distribution of EPVS including T-EPVS may be useful for predicting amyloid status, indicating that impaired perivascular drainage may contribute to cerebral amyloidosis. ANN NEUROL 2023.

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.

Automated Methods for Detecting and Quantitation of Enlarged Perivascular spaces on MRI

The brain's glymphatic system is a network of intracerebral vessels that function to remove "waste products" such as degraded proteins from the brain. It comprises of the vasculature, perivascular spaces (PVS), and astrocytes. Poor glymphatic function has been implicated in numerous diseases; however, its contribution is still unknown. Efforts have been made to image the glymphatic system to further assess its role in the pathogenesis of different diseases. Numerous imaging modalities have been utilized including two-photon microscopy and contrast-enhanced magnetic resonance imaging (MRI). However, these are associated with limitations for clinical use. PVS form a part of the glymphatic system and can be visualized on standard MRI sequences when enlarged. It is thought that PVS become enlarged secondary to poor glymphatic drainage of metabolites. Thus, quantitating PVS could be a good surrogate marker for glymphatic function. Numerous manual rating scales have been developed to measure the PVS number and size on MRI scans; however, these are associated with many limitations. Instead, automated methods have been created to measure PVS more accurately in different diseases. In this review, we discuss the imaging techniques currently available to visualize the glymphatic system as well as the automated methods currently available to measure PVS, and the strengths and limitations associated with each technique. EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 1.

A critical guide to the automated quantification of perivascular spaces in magnetic resonance imaging

The glymphatic system is responsible for waste clearance in the brain. It is comprised of perivascular spaces (PVS) that surround penetrating blood vessels. These spaces are filled with cerebrospinal fluid and interstitial fluid, and can be seen with magnetic resonance imaging. Various algorithms have been developed to automatically label these spaces in MRI. This has enabled volumetric and morphological analyses of PVS in healthy and disease cohorts. However, there remain inconsistencies between PVS measures reported by different methods of automated segmentation. The present review emphasizes that importance of voxel-wise evaluation of model performance, mainly with the Sørensen Dice similarity coefficient. Conventional count correlations for model validation are inadequate if the goal is to assess volumetric or morphological measures of PVS. The downside of voxel-wise evaluation is that it requires manual segmentations that require large amounts of time to produce. One possible solution is to derive these semi-automatically. Additionally, recommendations are made to facilitate rigorous development and validation of automated PVS segmentation models. In the application of automated PVS segmentation tools, publication of image quality metrics, such as the contrast-to-noise ratio, alongside descriptive statistics of PVS volumes and counts will facilitate comparability between studies. Lastly, a head-to-head comparison between two algorithms, applied to two cohorts of astronauts reveals how results can differ substantially between techniques.

Assessment of perivascular space filtering methods using a three-dimensional computational model

Growing interest surrounds the assessment of perivascular spaces (PVS) on magnetic resonance imaging (MRI) and their validation as a clinical biomarker of adverse brain health. Nonetheless, the limits of validity of current state-of-the-art segmentation methods are still unclear. Here, we propose an open-source three-dimensional computational framework comprising 3D digital reference objects and evaluate the performance of three PVS filtering methods under various spatiotemporal imaging considerations (including sampling, motion artefacts, and Rician noise). Specifically, we study the performance of the Frangi, Jerman and RORPO filters in enhancing PVS-like structures to facilitate segmentation. Our findings were three-fold. First, as long as voxels are isotropic, RORPO outperforms the other two filters, regardless of imaging quality. Unlike the Frangi and Jerman filters, RORPO's performance does not deteriorate as PVS volume increases. Second, the performance of all "vesselness" filters is heavily influenced by imaging quality, with sampling and motion artefacts being the most damaging for these types of analyses. Third, none of the filters can distinguish PVS from other hyperintense structures (e.g. white matter hyperintensities, stroke lesions, or lacunes) effectively, the area under precision-recall curve dropped substantially (Frangi: from 94.21 [IQR 91.60, 96.16] to 43.76 [IQR 25.19, 63.38]; Jerman: from 94.51 [IQR 91.90, 95.37] to 58.00 [IQR 35.68, 64.87]; RORPO: from 98.72 [IQR 95.37, 98.96] to 71.87 [IQR 57.21, 76.63] without and with other hyperintense structures, respectively). The use of our computational model enables comparing segmentation methods and identifying their advantages and disadvantages, thereby providing means for testing and optimising pipelines for ongoing and future studies.

Perivascular spaces as a marker of disease severity and neurodegeneration in patients with behavioral variant frontotemporal dementia

Background

Behavioural Variant Frontotemporal Dementia (bvFTD) is a rapidly progressing neurodegenerative proteinopathy. Perivascular spaces (PVS) form a part of the brain’s glymphatic clearance system. When enlarged due to poor glymphatic clearance of toxic proteins, PVS become larger and more conspicuous on MRI. Therefore, enlarged PVS may be a useful biomarker of disease severity and progression in neurodegenerative proteinopathies such as bvFTD. This study aimed to determine the utility of PVS as a biomarker of disease progression in patients with bvFTD.

Materials and methods

Serial baseline and week 52 MRIs acquired from ten patients with bvFTD prospectively recruited and followed in a Phase 1b open label trial of sodium selenate for bvFTD were used in this study. An automated algorithm quantified PVS on MRI, which was visually inspected and validated by a member of the study team. The number and volume of PVS were extracted and mixed models used to assess the relationship between PVS burden and other measures of disease (cognition, carer burden scale, protein biomarkers). Additional exploratory analysis investigated PVS burden in patients who appeared to not progress over the 12 months of selenate treatment (i.e., “non-progressors”).

Results

Overall, PVS cluster number (ß = −3.27, CI [−7.80 – 1.27], p = 0.267) and PVS volume (ß = −36.8, CI [−84.9 – 11.3], p = 0.171) did not change over the paired MRI scans 12 months apart. There was association between cognition total composite scores and the PVS burden (PVS cluster ß = −0.802e^–3, CI [9.45e^–3 – −6.60e^–3, p ≤ 0.001; PVS volume ß = −1.30e^–3, CI [−1.55e^–3 – −1.05e^–3], p ≤ 0.001), as well as between the change in the cognition total composite score and the change in PVS volume (ß = 4.36e^–3, CI [1.33e^–3 – 7.40e^–3], p = 0.046) over the trial period. There was a significant association between CSF t-tau and the number of PVS clusters (ß = 2.845, CI [0.630 – 5.06], p = 0.036). Additionally, there was a significant relationship between the change in CSF t-tau and the change in the number of PVS (ß = 1.54, CI [0.918 – 2.16], p < 0.001) and PVS volume (ß = 13.8, CI [6.37 – 21.1], p = 0.003) over the trial period. An association was found between the change in NfL and the change in PVS volume (ß = 1.40, CI [0.272 – 2.52], p = 0.045) over time. Within the “non-progressor” group (n = 7), there was a significant relationship between the change in the CSF total-tau (t-tau) levels and the change in the PVS burden (PVS cluster (ß = 1.46, CI [0.577 – 2.34], p = 0.014; PVS volume ß = 14.6, CI [3.86 – 25.4], p = 0.032) over the trial period. Additionally, there was evidence of a significant relationship between the change in NfL levels and the change in the PVS burden over time (PVS cluster ß = 0.296, CI [0.229 – 0.361], p ≤ 0.001; PVS volume ß = 3.67, CI [2.42 – 4.92], p = 0.002).

Conclusion

Analysis of serial MRI scans 12 months apart in patients with bvFTD demonstrated a relationship between PVS burden and disease severity as measured by the total cognitive composite score and CSF t-tau. Further studies are needed to confirm PVS as a robust marker of neurodegeneration in proteinopathies.

Widespread white matter oedema in subacute COVID-19 patients with neurological symptoms

While neuropathological examinations in patients who died from COVID-19 revealed inflammatory changes in cerebral white matter, cerebral MRI frequently fails to detect abnormalities even in the presence of neurological symptoms. Application of multi-compartment diffusion microstructure imaging (DMI), that detects even small volume shifts between the compartments (intra-axonal, extra-axonal and free water/CSF) of a white matter model, is a promising approach to overcome this discrepancy. In this monocentric prospective study, a cohort of 20 COVID-19 inpatients (57.3 ± 17.1 years) with neurological symptoms (e.g. delirium, cranial nerve palsies) and cognitive impairments measured by the Montreal Cognitive Assessment (MoCA test; 22.4 ± 4.9; 70% below the cut-off value <26/30 points) underwent DMI in the subacute stage of the disease (29.3 ± 14.8 days after positive PCR). A comparison of whole-brain white matter DMI parameters with a matched healthy control group (n = 35) revealed a volume shift from the intra- and extra-axonal space into the free water fraction (V-CSF). This widespread COVID-related V-CSF increase affected the entire supratentorial white matter with maxima in frontal and parietal regions. Streamline-wise comparisons between COVID-19 patients and controls further revealed a network of most affected white matter fibres connecting widespread cortical regions in all cerebral lobes. The magnitude of these white matter changes (V-CSF) was associated with cognitive impairment measured by the MoCA test (r = -0.64, P = 0.006) but not with olfactory performance (r = 0.29, P = 0.12). Furthermore, a non-significant trend for an association between V-CSF and interleukin-6 emerged (r = 0.48, P = 0.068), a prominent marker of the COVID-19 related inflammatory response. In 14/20 patients who also received cerebral 18F-FDG PET, V-CSF increase was associated with the expression of the previously defined COVID-19-related metabolic spatial covariance pattern (r = 0.57; P = 0.039). In addition, the frontoparietal-dominant pattern of neocortical glucose hypometabolism matched well to the frontal and parietal focus of V-CSF increase. In summary, DMI in subacute COVID-19 patients revealed widespread volume shifts compatible with vasogenic oedema, affecting various supratentorial white matter tracts. These changes were associated with cognitive impairment and COVID-19 related changes in 18F-FDG PET imaging.

Effect of cerebral small vessel disease on cognitive impairment in Parkinson's disease

OBJECTIVES

To explore the association between cerebral small vessel disease (cSVD) and cognitive impairment (CI) in Parkinson's disease (PD).

METHODS

81 PD patients were recruited into the study from September 2018 to December 2020. The demographic characteristics and radiologic and laboratory data were collected. Cognitive assessments were carried out using the Montreal Cognitive Assessment. The association between cSVD and cognitive impairment was analyzed using univariate and binary logistic regression analysis.

RESULTS

The binary logistic regression analysis showed that, after correcting for age, educational years, hyperhomocysteinemia, hypertension, and diabetes mellitus, total cSVD scores (OR 1.55, 95% CI 1.07-2.27, P = 0.02), the presence of paraventricular white matter hyperintensity (PVH) (OR 11.78, 95% CI 3.08-45.01, P < 0.001), white matter hyperintensity (WMH) (OR 7.95, 95% CI 2.28-27.79, P = 0.001), and perivascular space (PVS) (OR 6.66, 95% CI 2.08-21.40, P = 0.001) were independent risk factors for PD-CI.

CONCLUSION

The presence of cSVD was associated with cognitive dysfunction in patients with PD. It may be beneficial to manage cSVD to prevent the progression of cognitive impairment in patients with PD.

Topological relationships between perivascular spaces and progression of white matter hyperintensities: A pilot study in a sample of the Lothian Birth Cohort 1936

Enlarged perivascular spaces (PVS) and white matter hyperintensities (WMH) are features of cerebral small vessel disease which can be seen in brain magnetic resonance imaging (MRI). Given the associations and proposed mechanistic link between PVS and WMH, they are hypothesized to also have topological proximity. However, this and the influence of their spatial proximity on WMH progression are unknown. We analyzed longitudinal MRI data from 29 out of 32 participants (mean age at baseline = 71.9 years) in a longitudinal study of cognitive aging, from three waves of data collection at 3-year intervals, alongside semi-automatic segmentation masks for PVS and WMH, to assess relationships. The majority of deep WMH clusters were found adjacent to or enclosing PVS (waves-1: 77%; 2: 76%; 3: 69%), especially in frontal, parietal, and temporal regions. Of the WMH clusters in the deep white matter that increased between waves, most increased around PVS (waves-1-2: 73%; 2-3: 72%). Formal statistical comparisons of severity of each of these two SVD markers yielded no associations between deep WMH progression and PVS proximity. These findings may suggest some deep WMH clusters may form and grow around PVS, possibly reflecting the consequences of impaired interstitial fluid drainage via PVS. The utility of these relationships as predictors of WMH progression remains unclear.

Imaging perivascular space structure and function using brain MRI

In this article, we provide an overview of current neuroimaging methods for studying perivascular spaces (PVS) in humans using brain MRI. In recent years, an increasing number of studies highlighted the role of PVS in cerebrospinal/interstial fluid circulation and clearance of cerebral waste products and their association with neurological diseases. Novel strategies and techniques have been introduced to improve the quantification of PVS and to investigate their function and morphological features in physiological and pathological conditions. After a brief introduction on the anatomy and physiology of PVS, we examine the latest technological developments to quantitatively analyze the structure and function of PVS in humans with MRI. We describe the applications, advantages, and limitations of these methods, providing guidance and suggestions on the acquisition protocols and analysis techniques that can be applied to study PVS in vivo. Finally, we review the human neuroimaging studies on PVS across the normative lifespan and in the context of neurological disorders.

The association of enlarged perivascular space with microglia-related inflammation and Alzheimer's pathology in cognitively normal elderly

BACKGROUND

Glymphatic dysfunction may contribute to the accumulation of Alzheimer's disease (AD) pathologies. Conversely, AD pathologic change might also cause neuroinflammation and aggravate glymphatic dysfunction, forming a loop that accelerates AD progression. In vivo validations are needed to confirm their relationships.

METHODS

In this study, we included 144 cognitively normal participants with AD pathological biomarker data (baseline CSF Aβ1-42, T-Tau, P-Tau181; plasma P-Tau181 at baseline and at least one follow-up) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Each subject had completed structural MRI scans. Among them, 117 subjects have available neuroinflammatory biomarker (soluble triggering receptor expressed on myeloid cells 2 (sTREM2), and 123 subjects have completed two times [18F]-florbetapir PET. The enlarged PVS (EPVS) visual rating scores in basal ganglia (BG) and centrum semiovale (CS) were assessed on T1-weighted images to reflect glymphatic dysfunction. Intracranial volume and white matter hyperintensities (WMH) volume were also calculated for further analysis. We performed stepwise linear regression models and mediation analyses to estimate the association between EPVS severity, sTREM2, and AD biomarkers.

RESULTS

CS-EPVS degree was associated with CSF sTREM2, annual change of plasma P-tau181 and total WMH volume, whereas BG-EPVS severity was associated with age, gender and intracranial volume. The sTREM2 mediated the association between CSF P-tau181 and CS-EPVS.

CONCLUSION

Impaired glymphatic dysfunction could contribute to the accumulation of pathological tau protein. The association between tauopathy and glymphatic dysfunction was mediated by the microglia inflammatory process. These findings may provide evidence for novel treatment strategies of anti-neuroinflammation therapy in the early stage.

Computer-Aided Extraction of Select MRI Markers of Cerebral Small Vessel Disease: A Systematic Review

Cerebral small vessel disease (CSVD) is a major vascular contributor to cognitive impairment in ageing, including dementias. Imaging remains the most promising method for in vivo studies of CSVD. To replace the subjective and laborious visual rating approaches, emerging studies have applied state-of-the-art artificial intelligence to extract imaging biomarkers of CSVD from MRI scans. We aimed to summarise published computer-aided methods for the examination of three imaging biomarkers of CSVD, namely cerebral microbleeds (CMB), dilated perivascular spaces (PVS), and lacunes of presumed vascular origin. Seventy classical image processing, classical machine learning, and deep learning studies were identified. Transfer learning and weak supervision techniques have been applied to accommodate the limitations in the training data. While good performance metrics were achieved in local datasets, there have not been generalisable pipelines validated in different research and/or clinical cohorts. Future studies could consider pooling data from multiple sources to increase data size and diversity, and evaluating performance using both image processing metrics and associations with clinical measures.

Enlarged perivascular spaces and white matter hyperintensities in patients with frontotemporal lobar degeneration syndromes

Objective

The aim of this study was to investigate the distribution characteristics of enlarged perivascular spaces (EPVS) and white matter hyperintensities (WMH) and their associations with disease severity across the frontotemporal lobar degeneration (FTLD) syndromes spectrum.

Methods

This study included 73 controls, 39 progressive supranuclear palsy Richardson’s syndrome (PSP-RS), 31 corticobasal syndrome (CBS), 47 behavioral variant frontotemporal dementia (bvFTD), 36 non-fluent variant primary progressive aphasia (nfvPPA), and 50 semantic variant primary progressive aphasia (svPPA). All subjects had brain magnetic resonance imaging (MRI) and neuropsychological tests, including progressive supranuclear palsy rating scale (PSPRS) and FTLD modified clinical dementia rating sum of boxes (FTLD-CDR). EPVS number and grade were rated on MRI in the centrum semiovale (CSO-EPVS), basal ganglia (BG-EPVS), and brain stem (BS-EPVS). Periventricular (PWMH) and deep (DWMH) were also graded on MRI. The distribution characteristics of EPVS and WMH were compared between control and disease groups. Multivariable linear regression analysis was performed to evaluate the association of EPVS and WMH with disease severity.

Results

Compared with control subjects, PSP-RS and CBS had more BS-EPVS; CBS, bvFTD, and nfvPPA had less CSO-EPVS; all disease groups except CBS had higher PWMH (p < 0.05). BS-EPVS was associated with PSPRS in PSP-RS (β = 2.395, 95% CI 0.888–3.901) and CBS (β = 3.115, 95% CI 1.584–4.647). PWMH was associated with FTLD-CDR in bvFTD (β = 1.823, 95% CI 0.752–2.895), nfvPPA (β = 0.971, 95% CI 0.030–1.912), and svPPA (OR: 1.330, 95% CI 0.457–2.204).

Conclusion

BS-EPVS could be a promising indicator of disease severity in PSP-RS and CBS, while PWMH could reflect the severity of bvFTD, nfvPPA, and svPPA.

Risk factors for hippocampal cavities in a marginally housed population

Cavities in the hippocampus are morphological variants of uncertain significance. Aberrant neurodevelopment along with vascular and inflammatory etiologies have been proposed. We sought to characterize these cavities and their potential risk factors in a marginally housed population, with high rates of viral infection, addiction, and mental illness. (1) The volume of hippocampal cavities (HCavs) is greater in this highly multimorbid population compared to the general population. (2) Conventional vascular risk factors such as greater age and systolic blood pressure are associated with higher HCav volume. (3) Nonprescribed substance-related risk factors such as stimulant use or dependence, and smoking are associated with increased HCav volume independent of vascular risk factors. This is a retrospective analysis of an ongoing prospective study. We analyzed baseline data, including medical history, physical exam, psychiatric diagnosis, and MRI from a total of 375 participants. Hippocampal cavities were defined as spaces isointense to CSF on T1 MRI sequences, bounded on all sides by hippocampal tissue, with a volume of at least 1 mm³ . Risk factors were evaluated using negative binomial multiple regression. Stimulant use was reported by 87.3% of participants, with stimulant dependence diagnosed in 83.3% of participants. Prevalence of cavities was 71.6%, with a mean total bilateral HCav volume of 13.89 mm³ . On average, a 1 mmHg greater systolic blood pressure was associated with a 2.17% greater total HCav volume (95% CI = [0.57%, 3.79%], p = .0076), while each cigarette smoked per day trended toward a 2.69% greater total HCav volume (95% CI = [-0.87%, 5.54%], p = .058). A diagnosis of stimulant dependence was associated with a 95.6% greater total HCav volume (95% CI = [5.39%, 263.19%], p = .0335). Hypertension and diagnosis of stimulant dependence were associated with a greater total volume of HCav.

Small Vessel Disease, a Marker of Brain Health: What the Radiologist Needs to Know

Small vessel disease, a disorder of cerebral microvessels, is an expanding epidemic and a common cause of stroke and dementia. Despite being almost ubiquitous in brain imaging, the clinicoradiologic association of small vessel disease is weak, and the underlying pathogenesis is poorly understood. The STandards for ReportIng Vascular changes on nEuroimaging (STRIVE) criteria have standardized the nomenclature. These include white matter hyperintensities of presumed vascular origin, recent small subcortical infarcts, lacunes of presumed vascular origin, prominent perivascular spaces, cerebral microbleeds, superficial siderosis, cortical microinfarcts, and brain atrophy. Recently, the rigid categories among cognitive impairment, vascular dementia, stroke, and small vessel disease have become outdated, with a greater emphasis on brain health. Conventional and advanced small vessel disease imaging markers allow a comprehensive assessment of global brain heath. In this review, we discuss the pathophysiology of small vessel disease neuroimaging nomenclature by means of the STRIVE criteria, clinical implications, the role of advanced imaging, and future directions.

Perivascular space dilation is associated with vascular amyloid-β accumulation in the overlying cortex

Perivascular spaces (PVS) are compartments surrounding cerebral blood vessels that become visible on MRI when enlarged. Enlarged PVS (EPVS) are commonly seen in patients with cerebral small vessel disease (CSVD) and have been suggested to reflect dysfunctional perivascular clearance of soluble waste products from the brain. In this study, we investigated histopathological correlates of EPVS and how they relate to vascular amyloid-β (Aβ) in cerebral amyloid angiopathy (CAA), a form of CSVD that commonly co-exists with Alzheimer's disease (AD) pathology. We used ex vivo MRI, semi-automatic segmentation and validated deep-learning-based models to quantify EPVS and associated histopathological abnormalities. Severity of MRI-visible PVS during life was significantly associated with severity of MRI-visible PVS on ex vivo MRI in formalin fixed intact hemispheres and corresponded with PVS enlargement on histopathology in the same areas. EPVS were located mainly around the white matter portion of perforating cortical arterioles and their burden was associated with CAA severity in the overlying cortex. Furthermore, we observed markedly reduced smooth muscle cells and increased vascular Aβ accumulation, extending into the WM, in individually affected vessels with an EPVS. Overall, these findings are consistent with the notion that EPVS reflect impaired outward flow along arterioles and have implications for our understanding of perivascular clearance mechanisms, which play an important role in the pathophysiology of CAA and AD.

Lesion Volume in Relapsing Multiple Sclerosis is Associated with Perivascular Space Enlargement at the Level of the Basal Ganglia

BACKGROUND AND PURPOSE

Perivascular spaces surround the blood vessels of the brain and are involved in neuroimmune functions and clearance of metabolites via the glymphatic system of the brain. Enlarged perivascular spaces could be a marker of dysfunction in these processes and, therefore, are highly relevant to monitoring disease activity in MS. This study aimed to compare the number of enlarged perivascular spaces in people with relapsing MS with MR imaging markers of inflammation and brain atrophy.

MATERIALS AND METHODS

Fifty-nine patients (18 with clinically isolated syndrome, 22 with early and 19 with late relapsing-remitting MS) were scanned longitudinally (mean follow-up duration = 19.6 [SD, 0.5] months) using T2-weighted, T1-weighted, and FLAIR MR imaging. Two expert raters identified and counted enlarged perivascular spaces on T2-weighted MR images from 3 ROIs (the centrum semiovale, basal ganglia, and midbrain). Baseline and change with time in the number of enlarged perivascular spaces were correlated with demographics and lesion and brain volumes.

RESULTS

Late relapsing-remitting MS had a greater average number of enlarged perivascular spaces at baseline at the level of the basal ganglia (72.3) compared with early relapsing-remitting MS (60.5) and clinically isolated syndrome (54.7) (F = 3.4, P = .042), and this finding correlated with lesion volume (R = 0.44, P = .0004) but not brain atrophy (R = -0.16). Enlarged perivascular spaces increased in number with time in all regions, and the rate of increase did not differ among clinical groups.

CONCLUSIONS

Enlarged perivascular spaces at the level of the basal ganglia are associated with greater neuroinflammatory burden, and the rate of enlargement appears constant in patients with relapsing-remitting disease phenotypes.

Automated grading of enlarged perivascular spaces in clinical imaging data of an acute stroke cohort using an interpretable, 3D deep learning framework

Enlarged perivascular spaces (EPVS), specifically in stroke patients, has been shown to strongly correlate with other measures of small vessel disease and cognitive impairment at 1 year follow-up. Typical grading of EPVS is often challenging and time consuming and is usually based on a subjective visual rating scale. The purpose of the current study was to develop an interpretable, 3D neural network for grading enlarged perivascular spaces (EPVS) severity at the level of the basal ganglia using clinical-grade imaging in a heterogenous acute stroke cohort, in the context of total cerebral small vessel disease (CSVD) burden. T2-weighted images from a retrospective cohort of 262 acute stroke patients, collected in 2015 from 5 regional medical centers, were used for analyses. Patients were given a label of 0 for none-to-mild EPVS (< 10) and 1 for moderate-to-severe EPVS (≥ 10). A three-dimensional residual network of 152 layers (3D-ResNet-152) was created to predict EPVS severity and 3D gradient class activation mapping (3DGradCAM) was used for visual interpretation of results. Our model achieved an accuracy 0.897 and area-under-the-curve of 0.879 on a hold-out test set of 15% of the total cohort (n = 39). 3DGradCAM showed areas of focus that were in physiologically valid locations, including other prevalent areas for EPVS. These maps also suggested that distribution of class activation values is indicative of the confidence in the model's decision. Potential clinical implications of our results include: (1) support for feasibility of automated of EPVS scoring using clinical-grade neuroimaging data, potentially alleviating rater subjectivity and improving confidence of visual rating scales, and (2) demonstration that explainable models are critical for clinical translation.