Reliability of tissue volumes and their spatial distribution for segmentedmagnetic resonance images

Before using MRI tissue segmentation in clinical studies as a dependent var iable or as a means to correct functional data for differential tissue cont ribution, we must first establish the volume reliability and spatial distri bution reproducibility of the segmentation method. Although several reports of volume reliability can be found in the literature, there are no article s assessing the reproducibility of the spatial distribution of tissue. In t his report, we examine the validity, volume reliability, and spatial distri bution reproducibility for our K-means cluster segmentation. Validation was examined by classifying gray matter, white matter, and CST: on images cons tructed using an MRI simulator and digital brain phantom, with percentage v olume differences of less than 5% and spatial distribution overlaps greater than 0.94 (1.0 is perfect). We also segmented repeat scan MRIs from 10 hea lthy subjects, with intraclass correlation coefficients greater than 0.92 f or cortical gray matter, white matter, sulcal CSF, and ventricular CSF. The original scans were also coregistered to the repeat scan of the same subje ct, and the spatial overlap for each tissue was then computed. Our overlaps ranged from 0.75 to 0.86 fur these tissues. Our results support the use of K-means cluster segmentation, and the use of segmented structural MRIs to guide the analysis of functional and other images. (C) 2001 Elsevier Scienc e Ireland Ltd. All rights reserved.