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

Williamson, Brady J.; Khandwala, Vivek; Wang, David; Maloney, Thomas; Sucharew, Heidi; Horn, Paul S.; Haverbusch, Mary; Alwell, Kathleen; Gangatirkar, Shantala; Mahammedi, Abdelkader; Wang, Lily L.; Tomsick, Thomas A.; Gaskill-Shipley, Mary; Cornelius, Rebecca S.; Khatri, Pooja; Kissela, Brett M; Vagal, Achala

doi:10.1038/s41598-021-04287-4

Cited by 23 publications

(12 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The FMRIB Software Library (FSL, https://fsl.fmrib.ox.ac.uk/fsl/fslwiki) is the preferred software used for sequence co-registration, occasionally for brain extraction, and in two sources (Gonzalez-Castro et al, 2016b, 2017) it is also used for generating priors of the basal ganglia region. AFNI (https://afni.nimh.nih.gov/) is used by two sources (Boespflug et al, 2018; Williamson et al, 2022) for co-registration and skull-stripping, and elastix (https://elastix.lumc.nl/) is used for co- registration by one (Spijkerman et al, 2022). Not all methods correct for b1 inhomogeneities in the magnetic field, compensate for the presence of noise, or refer to a software/algorithm for normalising the intensities of the images prior to segmentation.…”

Section: Resultsmentioning

confidence: 99%

A Systematic Review and Meta-Analysis of Automated Methods for Quantifying Enlarged Perivascular Spaces in the Brain

Waymont,

Valdés Hernández,

Bernal

et al. 2024

Preprint

View full text Add to dashboard Cite

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.

show abstract

Section: Resultsmentioning

confidence: 99%

A Systematic Review and Meta-Analysis of Automated Methods for Quantifying Enlarged Perivascular Spaces in the Brain

Waymont,

Valdés Hernández,

Bernal

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Most studies assessed MVPVS using a manual scoring and/or segmentation. However, a couple of studies used automated segmentation methods using intensity-based thresholding approaches ( Ramirez et al, 2015 ; Wang et al, 2016 ; Boespflug et al, 2018 ), vesselness filter approaches ( Ballerini et al, 2018 ; Sepehrband et al, 2019 ), combination of these two methods ( Spijkerman et al, 2022 ), or approaches based on machine-learning ( Park et al, 2016 ; Hou et al, 2017 ; Boutinaud et al, 2021 ; Williamson et al, 2022 ) [reviewed in Moses et al (2022) ].…”

Section: Resultsmentioning

confidence: 99%

The etiology and evolution of magnetic resonance imaging-visible perivascular spaces: Systematic review and meta-analysis

Okar

Shinohara

et al. 2023

Front. Neurosci.

View full text Add to dashboard Cite

ObjectivesPerivascular spaces have been involved in neuroinflammatory and neurodegenerative diseases. Upon a certain size, these spaces can become visible on magnetic resonance imaging (MRI), referred to as enlarged perivascular spaces (EPVS) or MRI-visible perivascular spaces (MVPVS). However, the lack of systematic evidence on etiology and temporal dynamics of MVPVS hampers their diagnostic utility as MRI biomarker. Thus, the goal of this systematic review was to summarize potential etiologies and evolution of MVPVS.MethodsIn a comprehensive literature search, out of 1,488 unique publications, 140 records assessing etiopathogenesis and dynamics of MVPVS were eligible for a qualitative summary. 6 records were included in a meta-analysis to assess the association between MVPVS and brain atrophy.ResultsFour overarching and partly overlapping etiologies of MVPVS have been proposed: (1) Impairment of interstitial fluid circulation, (2) Spiral elongation of arteries, (3) Brain atrophy and/or perivascular myelin loss, and (4) Immune cell accumulation in the perivascular space. The meta-analysis in patients with neuroinflammatory diseases did not support an association between MVPVS and brain volume measures [R: −0.15 (95%-CI −0.40–0.11)]. Based on few and mostly small studies in tumefactive MVPVS and in vascular and neuroinflammatory diseases, temporal evolution of MVPVS is slow.ConclusionCollectively, this study provides high-grade evidence for MVPVS etiopathogenesis and temporal dynamics. Although several potential etiologies for MVPVS emergence have been proposed, they are only partially supported by data. Advanced MRI methods should be employed to further dissect etiopathogenesis and evolution of MVPVS. This can benefit their implementation as an imaging biomarker.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/display_record.php?RecordID=346564, identifier CRD42022346564.

show abstract

“…Due to very little ePVS, coexisting extensive WMH, or the presence of lacunes, the reliability of this evaluation method is reduced [ 21 ]. Some special rating scales or three-dimensional automated grading scales can be used to improve the detection of CSVD [ 21 , 22 ]. Second, the risk factors, such as hypertension, diabetes, and hyperlipidemia, were included, while some confounding variables, such as smoking and obesity, were not included in the analysis model.…”

Section: Discussionmentioning

confidence: 99%

White Matter Hyperintensities in Young Patients from a Neurological Outpatient Clinic: Prevalence, Risk Factors, and Correlation with Enlarged Perivascular Spaces

Zou

Wang

Zhang

et al. 2023

JPM

View full text Add to dashboard Cite

(1) Background: to investigate the prevalence of white matter hyperintensities (WMH), risk factors, and correlation with enlarged perivascular spaces (ePVS) among young patients (age, 16–45 years) in a neurological outpatient clinic. (2) Methods: a total of 887 young patients who underwent a head magnetic resonance imaging (MRI)examination between 1 June 2021, and 30 November 2021, were included in this study. Paraventricular WMH (PWMH), deep WMH (DWMH), ePVS in the centrum semiovale (CSO-ePVS), and basal ganglia (BG-ePVS) were rated. Logistic regression analysis was used to identify the best predictors for the presence of WMH and, for the association of the severity of ePVS with the presence of WMH. Goodman–Kruskal gamma test was used to assess the correlation between the severity of ePVS and WMH. (3) Results: the prevalence of WMH was 37.0%, with low severity. Age, hypertension (p < 0.001), headache (p = 0.031), syncope (p = 0.012), and sleep disturbance (p = 0.003) were associated with the presence of DWMH. Age, sex (p = 0.032), hypertension (p = 0.004) and sleep disturbance (p < 0.001) were associated with the presence of PWMH. The severity of CSO-ePVS was associated with the presence and the severities of DWMH. The severity of BG-ePVS was associated with the presence and severities of DWMH and PWMH. (4) Conclusions: the prevalence of WMH was 37% and mild in young patients without specific causes. Older age, female, hypertension, headache, syncope, and sleep disturbance were associated with WMH. The severity of ePVS had an impact on the presence and severity of WMH in the corresponding brain regions.

show abstract

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

Cited by 23 publications

References 20 publications

A Systematic Review and Meta-Analysis of Automated Methods for Quantifying Enlarged Perivascular Spaces in the Brain

A Systematic Review and Meta-Analysis of Automated Methods for Quantifying Enlarged Perivascular Spaces in the Brain

The etiology and evolution of magnetic resonance imaging-visible perivascular spaces: Systematic review and meta-analysis

White Matter Hyperintensities in Young Patients from a Neurological Outpatient Clinic: Prevalence, Risk Factors, and Correlation with Enlarged Perivascular Spaces

Contact Info

Product

Resources

About