Evaluation of respiratory motion effects in comparison with other parameters affecting PET image quality

Respiratory motion during emission imaging reduces both quality and quantitative accuracy. Proposed correction methodologies concentrate on the use of respiration gated acquisitions. However, the individual frame images are of reduced resolution as well as overall quality since they contain only a fraction of the counts available throughout a normal respiration average PET acquisition. The objectives of our study were to investigate (a) the effect of reduced image statistics present in respiratory gated images and (b) the relative significance of respiratory motion effects in comparison to other common parameters affecting image quality and quantitation in PET. The NCAT phantom, including different size lesions (7 mm to 22 mm) placed at various locations in the lungs and liver, was used in combination with a MC simulation of the GE Advance PET scanner. Ten individual time frames throughout the respiratory cycle were simulated. Five different levels of image statistical quality were considered. Images of individual frames and corresponding respiration average frames were reconstructed with and without scatter or attenuation corrections. Our results demonstrate that >6-8 million total coincidences are necessary per frame in order to realise any advantages of gated acquisitions in terms of reducing the effects of respiratory motion. Considering this level of statistical image quality, the magnitude of improvement is on average equivalent to performing a scatter correction.