Document Detail


Automatic segmentation of white matter hyperintensities in the elderly using FLAIR images at 3T.
MedLine Citation:
PMID:  20512882     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
PURPOSE: To determine the precision and accuracy of an automated method for segmenting white matter hyperintensities (WMH) on fast fluid-attenuated inversion-recovery (FLAIR) images in elderly brains at 3T. MATERIALS AND METHODS: FLAIR images from 18 individuals (60-82 years, 9 females) with WMH burdens ranging from 1-80 cm(3) were used. The protocol included the removal of clearly hyperintense voxels; two-class fuzzy C-means clustering (FCM); and thresholding to segment probable WMH. Two false-positive minimization (FPM) methods using white matter templates were tested. Precision was assessed by adding synthetic hyperintense voxels to brain slices. Accuracy was validated by comparing automatic and manual segmentations. Whole-brain, voxel-wise metrics of similarity, under- and overestimation were used to evaluate both precision and accuracy. RESULTS: Precision was high, as the lowest accuracy in the synthetic datasets was 93%. Both FPM strategies successfully improved overall accuracy. Whole-brain accuracy for the FCM segmentation alone ranged from 45%-81%, which improved to 75%-85% using the FPM strategies. CONCLUSION: The method was accurate across the range of WMH burden typically seen in the elderly. Accuracy levels achieved or exceeded those of other approaches using multispectral and/or more sophisticated pattern recognition methods.
Authors:
Erin Gibson; Fuqiang Gao; Sandra E Black; Nancy J Lobaugh
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't    
Journal Detail:
Title:  Journal of magnetic resonance imaging : JMRI     Volume:  31     ISSN:  1522-2586     ISO Abbreviation:  J Magn Reson Imaging     Publication Date:  2010 Jun 
Date Detail:
Created Date:  2010-05-31     Completed Date:  2010-09-28     Revised Date:  2010-09-30    
Medline Journal Info:
Nlm Unique ID:  9105850     Medline TA:  J Magn Reson Imaging     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1311-22     Citation Subset:  IM    
Affiliation:
Cognitive Neurology, Sunnybrook Health Sciences Centre, Toronto, Ontario M4N 3M5, Canada.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Aged
Aged, 80 and over
Aging
Algorithms
Automation
Brain / pathology*
Cluster Analysis
False Positive Reactions
Female
Fuzzy Logic
Humans
Image Processing, Computer-Assisted
Magnetic Resonance Imaging / methods*
Middle Aged
Reproducibility of Results
Grant Support
ID/Acronym/Agency:
//Canadian Institutes of Health Research
Comments/Corrections

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Quantification of the magnetic resonance signal response to dynamic (C)O(2)-enhanced imaging in the ...
Next Document:  Pulse sequence and timing of contrast-enhanced MRI for assessing blood-brain barrier disruption afte...