Format

Send to:

Choose Destination
See comment in PubMed Commons below
Psychiatry Res. 2011 Aug 30;193(2):101-6. doi: 10.1016/j.pscychresns.2011.03.007. Epub 2011 Jun 16.

Quantitative approaches for assessment of white matter hyperintensities in elderly populations.

Author information

  • 1Taub Institute for Research on Alzheimer's Disease and the Aging Brain, College of Physicians and Surgeons, Columbia University, New York, NY 10032, USA. amb2139@columbia.edu

Abstract

White matter hyperintensities (WMH) are areas of increased signal on T2-weighted magnetic resonance imaging (MRI), including fluid attenuated inverse recovery sequences. Total and regional WMH burden (i.e., volume or severity) has been associated with myriad cognitive, neurological, and psychiatric conditions among older adults. In the current report, we illustrate two approaches to quantify periventricular, deep, and total WMH and examine their reliability and criterion validity among 28 elderly patients enrolled in a depression treatment trial. The first approach, an operator-driven quantitative approach, involves visual inspection of individual MRI scans and manual labeling using a three-step series of procedures. The second approach, a fully automated quantitative approach, uses a processing stream that involves image segmentation, voxel intensity thresholding, and seed growing to label WMH and calculate their volume automatically. There was good agreement in WMH quantification between the two approaches (Cronbach's alpha values from 0.835 to 0.968). Further, severity of WMH was significantly associated with worse depression and increased age, and these associations did not differ significantly between the two quantification approaches. We provide evidence for good reliability and criterion validity for two approaches for WMH volume determination. The operator-driven approach may be better suited for smaller studies with highly trained raters, whereas the fully automated quantitative approach may be more appropriate for larger, high-throughput studies.

Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.

PMID:
21680159
[PubMed - indexed for MEDLINE]
PMCID:
PMC3164869
Free PMC Article

Publication Types, MeSH Terms, Substances, Grant Support

Publication Types

MeSH Terms

Substances

Grant Support

PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science Icon for PubMed Central
    Loading ...
    Write to the Help Desk