Uncertainty Analysis in SPECT Reconstruction based on Probabilistic Programming

원문 초록

Single Photon Emission Computed Tomography (SPECT) is one of the nuclear medicine imaging modalities used for functional analysis of animal or human organs. Gamma rays emitted from the scanned body are filtered with collimators and detected by the SPECT head. The conventional reconstruction algorithms do not consider the uncertainty in the process introduced by the field of view of the collimators. In this paper, we incorporate the probabilistic programming approach for SPECT image reconstruction. No-U-Turn Sampler (NUTS) is used to estimate the scanned object system by considering uncertainty. Results indicate that the current work can include uncertainty in reconstruction compared to conventional approaches like MLEM and MAP. However, reconstruction time need to be improved for phantom sizes of 128x128x128 voxels and higher.