Emission tomography with a multi-bang assumption on attenuation
Abstract
We consider the problem of joint reconstruction of both attenuation and source density in emission tomography in two dimensions. This is sometimes called the Single Photon Emission Computed Tomography (SPECT) identification problem, or referred to as attenuation correction in SPECT. Assuming that takes only finitely many values and we are able to characterise singularities appearing in the Attenuated Radon Transform , which models emission tomography data. Using this characterisation we prove that both and can be determined in some circumstances. We also propose a numerical algorithm to jointly compute and from based on a weakly convex regularizer when only takes values from a known finite list, and show that this algorithm performs well on some synthetic examples.
Cite
@article{arxiv.2001.04190,
title = {Emission tomography with a multi-bang assumption on attenuation},
author = {Sean Holman and Philip Richardson},
journal= {arXiv preprint arXiv:2001.04190},
year = {2020}
}
Comments
23 pages, 7 figures