Semi-Automated Method for Quantifying and Localizing White Matter Hyperintensities on MR Images
Wu M, Rosano C, Butters M, Whyte E, Meltzer C, Crooks R, Aizenstein H, Reynolds III C
University of Pittsburgh
White matter hyperintensities (WMH), commonly found on T2 and FLAIR brain MR images in the elderly, are associated with a number of neuropsychiatric disorders. In the current study we implemented and validated a semi-automated method for quantifying and localizing WMHs. We adapted a fuzzy connected algorithm to automate the segmentation of WMHs and used a demons-based image registration to automate the anatomic localization of WMHs. The volumes of automated WMHs were significantly correlated to the accepted gold standard (manual ratings, p=0.0003; correlation coefficient 0.907). As predicted, the WMHs in late-life depression (LLD) clustered in the prefrontal cortex and subcortical regions.