Image reconstruction for a novel Compton scatter tomograph

Abstract

Single photon emission computed tomography(SPECT) is a widespread medical imaging technology which provides images of metabolic tracer distributions within the body by detecting gamma-ray emissions from decaying radioactive isotopes in the tracer: The Compton single photon emission tomograph (C-SPECT) is a new imaging technology under development at the University of Michigan which promises significantly higher sensitiviry than standard mechanically collimated SPECTscanners due its use offully 30 electronic collimation of Compton scattered gamma-rays. Since the C-SPECT scanner generates extremely large data sets, and since the gamma-ray emission and detection processes are govemed by the statistical physics of nuclear interactions, the theory of large scale statistical signal and image processing must play a significant role in the development of this new technology. In this paper we will summarize results on the application of estimation theoretic lower bounds and image reconstruction to a C-SPECT system.