An EM algorithm for estimating SPECT emission and transmission parameters from emission data only

A maximum-likelihood (MI,) expectation-maximization (EM) algorithm (called EM-IntraSPECT) is presented for simultaneously estimating single photon emi ssion computed tomography (SPECT) emission and attenuation parameters from emission data alone, The algorithm uses the activity within the patient as transmission tomography sources, with which attenuation coefficients can be estimated. For this initial study, EM-IntraSPECT was tested on computer-si mulated attenuation and emission maps representing a simplified human thora x as well as on SPECT data obtained from a physical phantom, Two evaluation s were performed, First, to corroborate the idea of reconstructing attenuat ion parameters from emission data, attenuation parameters (mu) were estimat ed with the emission intensities (lambda) fixed at their true values, Accur ate reconstructions of attenuation parameters were obtained. Second, emissi on parameters lambda and attenuation parameters mu were simultaneously esti mated from the emission data alone. In this case there was crosstalk betwee n estimates of lambda and mu and final estimates of X and mu depended on in itial values. Estimates degraded significantly as the support extended out farther from the body, and an explanation for this is proposed. in the EM-I ntraSPECT reconstructed attenuation images, the lungs, spine, and soft tiss ue were readily distinguished and had approximately correct shapes and size s. As compared with standard EM reconstruction assuming a fix uniform atten uation map, EM-IntraSPECT provided more uniform estimates of cardiac activi ty in the physical phantom study and in the simulation study with tight sup port, but less uniform estimates with a broad support. The new EM algorithm derived here has additional applications, including reconstructing emissio n and transmission projection data under a unified statistical model.