17. Comparison of the sensitivity to cardiac motions of different cameras used for myocardial perfusion imaging

Abstract

Introduction New solid-state CZT cameras provide high quality images, with a better spatial resolution than conventional Anger cameras [1] , but their robustness and sensitivity to cardiac motions remains to be assessed. The aim of this study was to compare the motion-related artefactual variations in the wall uptake of a cardiac phantom between the CZT cameras Discovery NM-530c (GE) and DSPECT (Spectrum Dynamics) and the Anger camera Symbia T2 (Siemens), which was equipped with an astigmatic (IQSPECT) or parallel-hole (conventional SPECT) collimator. Materials and methods Translations of a cardiac insert, which walls were filled-in with a solution containing 99mTc, were applied in 6 different directions (−X/+X, −Y/+Y et −Z/+Z) and with various amplitudes at half of each SPECT recording. Images were reconstructed with routine parameters [2] . The motion-related artefactual variations in myocardial uptake were assessed according to the difference in segmental uptake between each SPECT recorded with an insert translation and the corresponding normal SPECT-recording and by using a 17 segments division of the left ventricle [3] . Differences >10% were considered as significant. Results and conclusion For translations of only 5 mm, the variations in segmental uptake were not or only poorly significant whatever the camera type. For higher translation amplitudes, the CZT D530c camera exhibited the higher sensitivity to motion and the DSPECT, the lower one, whereas Anger cameras were at an intermediate level irrespective of the collimator type (parallel-hole or astigmatic).