Validation of a Regression Technique for Segmentation of White Matter Hyperintensities in Alzheimer's Disease.

TitleValidation of a Regression Technique for Segmentation of White Matter Hyperintensities in Alzheimer's Disease.
Publication TypeJournal Article
Year of Publication2017
AuthorsDadar M, Pascoal TA, Manitsirikul S, Misquitta K, Tartaglia C, Breitner JCS, Rosa-Neto P, Carmichael O, DeCarli C, D Collins L
JournalIEEE Trans Med Imaging
Date Published2017 04 12
ISSN1558-254X
Abstract

Segmentation and volumetric quantification of white matter hyperintensities (WMHs) is essential in assessment and monitoring of the vascular burden in aging and Alzheimer's disease (AD), especially when considering their effect on cognition. Manually segmenting WMHs in large cohorts is technically unfeasible due to time and accuracy concerns. Automated tools that can detect WMHs robustly and with high accuracy are needed. Here we present and validate a fully automatic technique for segmentation and volumetric quantification of WMHs in aging and AD. The proposed technique combines intensity and location features from multiple magnetic resonance imaging (MRI) contrasts and manually labeled training data with a linear classifier to perform fast and robust segmentations. It provides both a continuous subject specific WMH map reflecting different levels of tissue damage and binary segmentations. The method was used to detect WMHs in 80 elderly/AD brains (ADC dataset) as well as 40 healthy subjects at risk of AD (PREVENT-AD dataset). Robustness across different scanners was validated using 10 subjects from ADNI2/GO study. Voxel-wise and volumetric agreements were evaluated using Dice similarity index (SI) and intra-class correlation (ICC), yielding ICC=0.96, SI=0.62±0.16 for ADC dataset and ICC=0.78, SI=0.51±0.15 for PREVENT-AD dataset. The proposed method was robust in the independent sample yielding SI=0.64±0.17 with ICC=0.93 for ADNI2/GO subjects. The proposed method provides fast, accurate and robust segmentations on previously unseen data from different models of scanners, making it ideal to study WMHs in large scale multi-site studies.1.

DOI10.1109/TMI.2017.2693978
Alternate JournalIEEE Trans Med Imaging
PubMed ID28422655