The Bowen Lab is working in collaboration with Siemens Medical Solutions Inc. to assess the performance of a new PET scatter correction method that jointly estimates scatter with radiotracer uptake.
Inaccurate correction for scattered coincidences can have a significant negative impact on radiotracer quantification. Current commercial scatter estimation methods used clinically have notable performance limitations. A common source of reduced image quantification with conventional scatter corrections is attributed to erroneous scaling of the initial scatter estimate to match acquired scattered events in the sinogram. Siemens has developed a new scatter correction method, termed maximum likelihood background scaling (MLBS). MLBS is a joint reconstruction algorithm that alternates between radiotracer uptake and scatter scaling estimation steps. MLBS may have performance advantages over conventional methods by using all available data intersecting the subject.
Our results indicate that for 18F-FDG imaging MLBS is at least a valid substitute to a conventional method (TFBS), but may have performance advantages in cases where TFBS is typically prone to inaccuracies (i.e. patient motion in image above). This analysis is a critical first step in justifying replacing conventional scatter corrections with MLBS for clinical care, and we continue evaluating this method in different PET-CT exam types.