Abstract: The image reconstruction processing in Single Photon Emission Computed Tomography (SPECT)
is a very ill-posed inverse problem. Due to its practical importance for medical diagnostics, this problem has been addressed various times recently. In this paper, the authors discuss theoretical settings of SPECT and propose new numerical strategies based on Tikhonov-type regularization for solving it. Moreover, multilevel continuation mechanism for large-scale problems is designed. Numerical results and conclusion follow.

Key words: SPECT, Tikhonov Regularization, Huber Norm, Multilevel Continuation.