The effect of SPECT reconstruction corrections on the absolute and relative quantitative accuracy of myocardial perfusion studies

This paper reports the findings of investigations into the performance of SPECT corrections for photon attenuation, distance-dependent resolution loss and photon scatter on the absolute and relative quantitative accuracy of myocardial perfusion studies. The measurements of myocardial wall thickness and myocardial infarct size were used to estimate the accuracy of the absolute and relative quantitative accuracy, respectively. A series of phantom studies were performed and additional information was gathered from a group of 37 normal patients. Each set of data was reconstructed with (1) filtered-back projection (FBP), (2) ordered subset expectation maximization (OSEM), (3) OSEM plus attenuation correction (AC), (4) OSEM plus detector response compensation (DRC), (5) OSEM plus AC and DRC, and (6) OSEM plus AC, DRC and scatter correction (SC). The image analysis toolbox iQuant was used to perform the analysis. Both patient and phantom data showed SPECT image corrections to have a significant effect on myocardial wall thickness, with reconstructions involving SC providing the most accurate results. Phantom data showed that good estimates of anterior and lateral wall infarct sizes are provided by all reconstruction techniques, whereas good estimates of inferior and septal wall infarct sizes are only provided by reconstructions including AC. Accurate measurement of in fa rets in any location and of any size is only possible with SC. This analysis suggests that in order to achieve perfect quantitative accuracy of myocardial perfusion studies, corrections for photon attenuation, distance-dependent resolution loss and photon scatter should be applied. However, for routine clinical analysis involving visual interpretation and an estimate of infarct size, the application of attenuation correction and detector resolution compensation might be considered sufficient.