Premium
IC‐P‐139: Accurate automatic segmentation of white matter hyperintensities using a linear regression classifier
Author(s) -
Dadar Mahsa,
Pascoal Tharick Ali,
Manitsirikul Sarinporn,
Breitner John,
Rosa-Neto Pedro,
Collins D. Louis
Publication year - 2015
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2015.06.161
Subject(s) - fluid attenuated inversion recovery , hyperintensity , segmentation , thresholding , artificial intelligence , pattern recognition (psychology) , voxel , gold standard (test) , kappa , dementia , linear regression , nuclear medicine , medicine , computer science , magnetic resonance imaging , mathematics , radiology , pathology , machine learning , geometry , disease , image (mathematics)
features are used to segment WMHs. Using the automatic segmentations, average maps of WMH loads were calculated for each population separately. Results: Figure 1 shows 12 transverse slices of the average WMH maps for the entire PREVENT-AD and ADC populations overlapped on the ADNI template. Conclusions: WMHs have been observed in normal control subjects at risk of AD as well as NC/MCI/AD population. WMHs occur more often in the periventricular areas especially adjacent to the horns of the lateral ventricles and less often in the other areas. The difference between the two maps suggests that WMHs are more extensive in AD populations than in healthy controls at risk of AD, especially in regions such as parietal lobe suggesting the probability of an inhomogeneous spread of the WMHs in different regions. (Funding: CIHR MOP-111169, P30 AG010129, Pfizer Canada, FRQ-S, and Douglas Hospital Research Centre) IC-P-139 ACCURATE AUTOMATIC SEGMENTATION OF WHITE MATTER HYPERINTENSITIES USING A LINEAR REGRESSION CLASSIFIER