Perivascular Space Semi-Automatic Segmentation (PVSSAS): A Tool for Segmenting, Viewing and Editing Perivascular Spaces

Abstract

SummaryObjectiveIn this study, we validate and describe a user-friendly tool for PVS tracing that uses a Frangi-based detection algorithm; which will be made freely available to aid in future clinical and research applications. All PVS detected by the semi-automated method had a match with the manual dataset and 94% of the manual PVS had a match within the semi-automated dataset.MethodsWe deployed a Frangi-based filter using a pre-existing Matlab toolbox. The PVSSAS tool pre-processes the images and is optimized for maximum effectiveness in this application. A user-friendly GUI was developed to aid the speed and ease in marking large numbers of PVS across the entire brain at once.ResultsUsing a tolerance of 0.7 cm, 83% of all PVSs detected by the semi-automated method had a match with the manual dataset and 94% of the manual PVS had a match within the semi-automated dataset. As shown in figure 3, there was generally excellent agreement between the manual and semi-automated markings in any given slice.SignificanceThe primary benefit of PVSSAS will be time saved marking PVS. Clinical MRI use is likely to become more widespread in the diagnosis, treatment, and study of MS and other degenerative neurological conditions in the coming years. Tools like the one presented here will be invaluable in ensuring that the tracing and quantitative analysis of these PVS does not act as a bottle neck to treatment and further research.

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

“…Many of the computational methods used in the application studies have been described in section 3.3, and have been applied directly or used as the basis of PVS quantification approaches in application studies. The most frequently applied method was thresholding the response of the Frangi filter, as described by Ballerini et al’s (2016, 2018) (12 studies) or in the frameworks proposed by Sepehrband et al, (2019) (10 studies), Smith et al, 2020 (1 study), or Ranti et al, (2022) (1 study), which in total make up for 54.24% of the application studies. Ten studies use CNN configurations, mainly U-Net and U-ResNet (16.95%), to assess PVS burden, and four studies use this approach to post-process the output from the Frangi filter.…”

“…It is worth noting that the four methods showcased by Sudre et al (2022) submitted as part of the challenge organised by the publication authors were, at the time this review was conducted, in a preprint repository. Despite the semi-automatic method proposed by Smith et al (2020) being also at a preprint repository, this method per-se has already been applied by a clinical study (Langan et al, 2022). Segmentation performance described qualitatively, with illustration.…”

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

“…Thus, manual delineation of 3D PVS is laborious and time-consuming. Despite this, the manual segmentation of perivascular spaces is a worthy endeavor for which multiple research groups have dedicated time and resources, in order to develop high quality algorithms ( Park et al, 2016 ; Zong et al, 2016 ; Zhang et al, 2017 ; Boespflug et al, 2018 ; Lian et al, 2018 ; Schwartz et al, 2019 ; Smith et al, 2020 , Preprint; Sepehrband et al, 2021 ; Lynch et al, 2022 , Preprint). The goal of automatic segmentation algorithms is to label such structures, eliminating the need for manual labeling, thereby expediting detailed analyses of PVS.…”

“…However, these studies are based on algorithms with certain disadvantages. The conventional methods relying on image processing techniques require further parameter optimization for different datasets ( Ballerini et al, 2016 ; Smith et al, 2020 , Preprint; Boutinaud et al, 2021 ; Bernal et al, 2022 ). Currently, only one fully automatic segmentation pipeline is freely available, making it difficult to replicate previous methods ( Boutinaud et al, 2021 ).…”