BAYESIAN RECONSTRUCTION AND USE OF ANATOMICAL A-PRIORI INFORMATION FOR EMISSION TOMOGRAPHY

A Bayesian method is presented for simultaneously segmenting and recon structing emission computed tomography (ECT) images and for incorporat ing high-resolution, anatomical information into those reconstructions . The anatomical information is often available from other imaging mod alities such as computed tomography (CT) or magnetic resonance imaging (MRI). The Bayesian procedure models the ECT radiopharmaceutical dist ribution as consisting of regions, such that radiopharmaceutical activ ity is similar throughout each region. It estimates the number of regi ons, the mean activity of each region, and the region classification a nd mean activity of each voxel. Anatomical information is incorporated by assigning higher prior probabilities to ECT segmentations in which each ECT region stays within a single anatomical region. This approac h is effective because anatomical tissue type often strongly influence s radiopharmaceutical uptake. The Bayesian procedure is evaluated usin g physically acquired single-photon emission computed tomography (SPEC T) projection data and MRI for the three-dimensional (3-D) Hoffman bra in phantom. A clinically realistic count level is used. A cold lesion within the brain phantom is created during the SPECT scan but not duri ng the MRI to demonstrate that the estimation procedure can detect ECT structure that is not present anatomically.