Evaluation of techniques for the elimination of "hot" bladder artifacts in SPECT of the pelvis.

The purpose of this study was to validate the usefulness of two digital filtering techniques used to eliminate the artifacts caused by rapid bladder filling during SPECT of the pelvis. A dynamic phantom model was used containing two hips and a bladder. The phantom was studied under three conditions--bladder empty, filling, and full. The ability of the pixel truncation and interpolative background replacement techniques to eliminate bladder artifacts was assessed. Both techniques gave similar results and resulted in significant (but not complete) recovery of activity in the hips. Quantitative analysis of pixel counts over each hip shows that the measured activity was variable and approximately 20%-30% less than that seen in the empty bladder study. The use of left/right ratio to quantitate differences in hip activity was highly inaccurate despite the use of these filtering techniques. In summary, while these techniques significantly improve image quality, caution should be exercised, particularly in the evaluation of the medial aspects of the hips.