Development and Reliability of a Method for Using Magnetic Resonance Imaging for the Definition of Regions of Interest for Positron Emission Tomography.

Abstract

Although positron emission tomography (PET) provides important physiological information, a major limitation of this technique is poor anatomical resolution. Combining the superior anatomical resolution of magnetic resonance imaging (MRI) with the physiologic information of PET could be an important advance for clinical and research applications of PET. The purpose of this study was to develop reliable methods for using MRI in the definition of brain regions of interest (ROIs) for application to coregistered PET brain images. First, specific criteria were developed for definition of ROIs on coregistered MRI using anatomical landmarks from a standard anatomical atlas. ROIs were then drawn by three independent raters using the criteria on 11 MRI scans, which had been coregistered to PET [(18)F]2-fluoro-2-deoxyglucose (FDG) brain scans. These MRI-based templates were used to determine activity in various brain regions on the PET scans, which were used in the determination of regional brain metabolic rates. There was a high level of agreement between raters for the measurement of regional metabolic rates, with intra-class correlation coefficients ranging from 0.63-0.98. These findings suggest that specific ROI criteria based on coregistered MRI are a reliable method of measuring activity from PET images. The use of a standard and widely available atlas and readily applicable criteria for ROIs should facilitate a standard method of measurement which can be applied in a similar manner at different PET sites.