Probability Based Positron Range Modeling in Inhomogeneous Medium for PET

Abstract

This paper introduces a novel probability based positron range model accounting for the typical case of inhomogeneous medium in experimental or clinical implementations. A survival probability for given energy, medium type and thickness, is used to describe the propagation of monoenergetic positrons with increasing distance in homogeneous medium. This follows from the fact that annihilation distribution of monoenergetic positrons in homogeneous medium can be approximately expressed as an exponential distribution. Similarly, we consider the transport and annihilation of monoenergetic positrons in each layer of inhomogeneous medium as probability events, which can be easily computed using a set of survival probabilities. The ultimate annihilation distribution is a weighted summation over the energy spectrum. Monte Carlo simulation is used to validate the availability and performance of this model. Simulated and estimated annihilation distributions are presented with close agreement. The new model can be used to improve the image resolution without introducing serious artifacts due to the mismatching between models and fact.

Isotope specific resolution recovery image reconstruction in high resolution PET imaging

“…1 Such separable positron range models have been reported in the literature using either analytic or Monte Carlo methods for both uniform 5,[19][20][21][22][23] and heterogeneous media with varying properties. [24][25][26][27][28][29][30] However, these models again need experimental validation. Measuring spatially variant isotope-specific PSFs from short-lived clinically used isotopes is very challenging owing to the rapid decay and the multitude of positions that need to be sam-pled and to best of our knowledge, experimental derivation of isotope-specific PSF maps has not been reported before.…”