Position dependent attenuation artifacts with a multi-pinhole dedicated cardiac camera

Dedicated cardiac SPECT cameras are an exciting new technology. Some of the new cameras use multi-pinhole collimation. With pinholes, the effects of attenuation can vary with source position. Our purpose was to determine the magnitude of the variation in attenuation effects as a function of position within the field-of-view (FOV) of a multi-pinhole dedicated cardiac camera (Discovery NMS30c, GE Healthcare). A cardiac-torso phantom with a small defect was measured at multiple axial positions from -5cm to +5cm through the center of the FOV of a cardiac multi-pinhole camera. Images were reconstructed using standard clinical parameters both with and without CT-based attenuation correction (AC). Images were analyzed using the 4DM SPECT clinical software package. The summed perfusion score (SPS) of the normalized uptake was compared. Without AC, the SPS varied from 5 to 12.5. This variation is significantly higher than the statistical uncertainty measured in the central position (SPS= 8.0 +/-0.7; F-test, p<;0.01). With AC, the variation was significantly reduced (F-test, p_i0.001) to between 0.4 and 2, supporting the hypothesis that the changes are primarily due to differences in attenuation. A shift of 1cm from the central position causes a change in SPS of up to 2. At 5cm, there is a visible increase in the size and severity of the lesion. In cardiac SPECT, images from rest and stress are compared and fixed attenuation with normal wall motion is used as a sign that the abnormality is caused by attenuation. Therefore, these changes are large enough to be clinically relevant and care must be taken in positioning the patient consistently and in the center of the FOV. In conclusion, variation in attenuation due to changes in source position can lead to significant changes in attenuation artifacts with multi-pinhole cardiac SPECT.