We propose to develop and evaluate two new scatter compensation algorithms to minimize the degrading impact of Compton scatter in quantitative Single Photon Emission Computed Tomography (SPECT). One, a 'deconvolution' scatter compensation algorithm in which the non-scattered data is deconvolved from the acquired data in Fourier transform space, has been preliminarily developed and will be evaluated as a fast, clinically implementable algorithm. The other, a unified reconstruction algorithm using inverse Monte Carlo techniques that simultaneously includes the effects of attenuation, scatter and depth-dependent collimator resolution, will be developed to include compensation for non-uniform attenuation and scattering. Two data sets will be used in the evaluation of the compensation algorithms: one, experimentally acquired phantom data in which the actual contrast ratios are known and two, Monte Carlo simulations in which the actual scatter contributions are known. Evaluation will consist of comparing concentration as well as contrast ratios from compensated images with the known values from the phantom and simulation data. Both of the new algorithms will be compared with standard filtered backprojection reconstruction.
| Floyd Jr, C E; Jaszczak, R J; Greer, K L et al. (1987) Brain phantom: high-resolution imaging with SPECT and I-123. Radiology 164:279-81 |
| Floyd Jr, C E; Jaszczak, R J; Coleman, R E (1987) Convergence of the maximum likelihood reconstruction algorithm for emission computed tomography. Phys Med Biol 32:463-76 |
| Floyd Jr, C E; Jaszczak, R J; Greer, K L et al. (1986) Inverse Monte Carlo as a unified reconstruction algorithm for ECT. J Nucl Med 27:1577-85 |
