Measured spatially variant system response for PET image reconstruction

Quantitative accuracy in PET imaging is essential for longitudinal studies and monitoring tumor response to treatment. The goal of this work is to improve the quantitative accuracy of whole-body PET imaging through the use of an accurate, measured system model. Past empirically measured system response functions used line sources positioned at various locations in the imaging field of view. Here, we present a practical method for measuring the detector blurring component of a whole-body PET system with a non-collimated point source. We employ Monte Carlo simulations to show that a non-collimated point source is acceptable for modeling the radial blurring present in a PET tomograph. And, we justify the use of a Na22 point source for collecting these measurements. We measure the system response, simplify it to a two-dimensional function, and incorporate a parameterized version of this response into a modified OSEM algorithm. Reconstructions of measured data from an image quality and tine source phantom reveal improved quantitative accuracy and resolution with the modified system model.