A tool for comparison of PET and fMRI methods: calculation of the uncertainty in the location of an activation site in a PET image

Abstract

A technique for calculating the uncertainty in the location of an activation site in a PET image, without performing repeated measures, is presented. With the development of new fMRI methods for measuring cerebral hemodynamics, demonstration of the efficacy of these techniques will be critical to establish clinical utility. Comparisons with PET are a powerful tool for validating these new fMRI techniques. In addition to the fact that PET techniques are well-established methods for making physiological measurements in vivo, PET methods are also free of the geometric distortions and nonuniform signal-to-noise artifacts (due to signal dropout) common in fMRI techniques. Comparisons reported previously have been limited by the large number of trials acquired in single-subject fMRI studies and the small number of trials in a PET study (due to the radiation dose to the patient or the interscan delays for tracer decay). Our method calculates both the center of mass (CM) of a predefined region of interest and the uncertainty in the location of the CM using the preimage PET data (sinograms). Results of phantom studies demonstrate that our method is an unbiased measurement equivalent to that of repeated measures with a large number of images. Extension of this technique to estimate the uncertainty in the location of an activation site in a PET statistical parametric map will permit precise rigorous comparisons of PET and fMRI methods in single subjects without the constraints imposed by the relatively small number of PET measurements.