Impact of respiratory motion compensation in cardiac SPECT when attenuation maps are not motion-compensated

During normal breathing, heart motion is about 15 mm along the body axis in humans. In a previous work, we proposed a method to track and to correct this motion after a list-mode acquisition which involves the recording of a signal proportional to respiratory volume. The goal of this work is to investigate the impact of motion compensation on cardiac SPECT emission when transmission data used for attenuation correction is not compensated for such motion. We use the MCAT chest phantom to simulate 24 temporal frames regularly spaced during the respiratory cycle, for 60 projection angles over 360 degrees. A 15 mm respiratory translation motion is simulated for the heart, liver and spleen, in both emission projections and attenuation maps. We also acquire list-mode emission and transmission data of the DataSpectrum anthropomorphic torso phantom with its cardiac insert with our Irix gamma-camera during mechanical respiratory motions. Motion is tracked using a pressure sensor, and a method for motion compensation is applied. Simulations with the MCAT phantom indicate that attenuation maps blurred by respiratory motion yield slight artifacts in the inferior wall. Also, minimal artifact is obtained with nonmotion compensated attenuation maps when the frame at midrespiratory cycle is used as a reference for motion compensation of the projections. The results obtained with the anthropomorphic phantom show that the 1.5 and 3.0 cm axial motions simulated during emission acquisitions induce apparent low uptake in the inferior and anterior wall of the myocardium. When the liver contains as much activity as the myocardium, the apparent low uptake is less visible in the inferior part of the myocardium.