partial volume effect?



Ph.D. graduate from Universite Claude Bernard (Lyon, France), currently working at the McConnell Brain Imaging Center, affiliated laboratory of the Montreal Neurological Institute and McGill University (Montreal, Quebec, Canada).
E-mail:
<olivier@OlivierRousset.com>
Office phone:
(410) 955-2900
Office fax:
(410) 955-0696
Positron Emission Tomography is a unique imaging device that allows to investigate fundamental brain functions. The accuracy of such device is limited by the relatively poor resolution compared to anatomy-oriented systems such as Magnetic Resonance Imaging (MRI) or Computerized Tomography (CT) scanners. Resolution effects are usually referred to as Partial Volume Effects (PVE's). If you want to "see" the first aspect (partial volume) of the effects of limited PET resolution, you can click here to get images of spheres of various size filled with the same activity concentration, but exhibiting different signal recovery. Those images were obtained from both real and simulated data, the latter being generated by a simulation package developed by Yilong Ma, which allows to simulate brain-scanner interaction.

Correction for PVE's must account for both the signal loss from the limited extend of the region observed compared to the spatial resolution of the tomographic system, and the "spillover" or signal contamination from surrounding tissue with different tracer uptake and kinetics. Since 1993, I have proposed to correct for partial volume effects by computing self-recovery and spillover fractions directly for the given set of regions-of-interest (ROI's), allowing for independence of the method towards any tracer level in any region that can be identified in the brain. A complete description and validation of this method has been published in the Journal of Nuclear Medicine, May 1998.

Projects and Responsibilities