Predicting Dementia in Cerebral Small Vessel Disease using an Automatic Diffusion Tensor Image Segmentation Technique

Williams, Owen A.; Zeestraten, Eva A.; Benjamin, Philip; Lambert, Christian; Lawrence, Andrew J.; Mackinnon, Andrew D.; Morris, Robin G.; Markus, Hugh S.; Barrick, Thomas R. and Charlton, Rebecca A. 2019. Predicting Dementia in Cerebral Small Vessel Disease using an Automatic Diffusion Tensor Image Segmentation Technique. Stroke, 50(10), pp. 2775-2782. ISSN 0039-2499 [Article]

Williams_inpress_Stroke_Accepted_Manuscript.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

Download (957kB) | Preview
STROKEAHA.119.025843 (1).pdf - Published Version
Available under License Creative Commons Attribution.

Download (4MB) | Preview

Abstract or Description

Background and Purpose: Cerebral small vessel disease (SVD) is the most common cause of vascular cognitive impairment, with a significant proportion of cases going on to develop dementia. We explore the extent to which DSEG (a diffusion tensor image, DTI segmentation technique that characterizes microstructural damage across the cerebrum) predicts both degree of cognitive decline and conversion to dementia, and hence may provide a useful prognostic procedure.

Methods: 99 SVD patients (aged 43-89) underwent annual MRI scanning (for three years) and cognitive assessment (for five years). DSEG-θ was used as a whole cerebrum measure of SVD severity. Dementia diagnosis was based DSM-V criteria. Cox regression identified which DSEG measures and vascular risk factors were related to increased risk of dementia. Linear discriminant analysis was used to classify groups of stable vs. subsequent dementia diagnosis individuals.

Results: DSEG-θ was significantly related to decline in executive function and global cognition (p <.001) Eighteen (18.2%) patients converted to dementia. Baseline DSEG-θ predicted dementia with a balanced classification rate (BCR) =75.95% and area under the receiver operator curve (AUC) =0.839. The best classification model included baseline DSEG-θ, change in DSEG-θ, age, sex and premorbid IQ (BCR of 79.65%, AUC=0.903).

Conclusions: DSEG is a fully automatic technique that provides an accurate method for assessing brain microstructural damage in SVD from a single imaging modality (DTI). DSEG-θ is an important tool in identifying SVD patients at increased risk of developing dementia and has potential as a clinical marker of SVD severity.

Item Type:


Identification Number (DOI):

Additional Information:

This research was funded by UK Charity Research into Aging (Grant #374). The SCANS research study was supported by a Wellcome Trust grant (081589). Recruitment was supported by the English National Institute of Health Research (NIHR) Clinical Stroke Research Network. This paper was supported by grants from ARUK-EXT2013-2 and ARUKPG2016A-1. Prof. Markus is supported by an NIHR Senior Investigator award and the Cambridge University Hospital Comprehensive NIHR Biomedical Research Unit. The SCANS study was registered with the UK clinical research network (, study ID:4577).


Magnetic Resonance Imaging, Cerebrovascular Disease/Stroke, Cognitive Impairment, brain, cerebral small vessel disease, cerebrum, cognition, cognitive dysfunction, dementia, diffusion tensor imaging

Departments, Centres and Research Units:



17 July 2019Accepted
12 September 2019Published Online
October 2019Published

Item ID:


Date Deposited:

19 Jul 2019 13:29

Last Modified:

11 Jun 2021 23:29

Peer Reviewed:

Yes, this version has been peer-reviewed.


View statistics for this item...

Edit Record Edit Record (login required)