%0 Journal Article %A Adam Kesner %A Jonathan Chung %A Phillip Koo %A David Lynch %T Signal optimization as strategy for non-specific 4D data utilization - An analysis of impact in respiratory gated human PET scans %D 2014 %J Journal of Nuclear Medicine %P 151-151 %V 55 %N supplement 1 %X 151 Objectives Respiratory gating is an enhanced imaging technique supported by hardware/software available from most commercial vendors. However, the clinical utility of 4D imaging has yet to be determined. In this work we explore the use of adaptive filtering techniques for optimization of the 4D signal, potentially applicable for non-specific populations, through analysis of respiratory-gated FDG PET scans. Methods Recently, we have shown that the noise and useful signal introduced to traditional (non-gated) data set through application of gating can be characterized and used for deriving adaptive region-specific filters (Kesner et. al., SNM 2012, EJNMMI research June 2013). Such filters provide access to the statistically supported information gain stemming from the motion characterization, yet avoid unnecessary information loss from superfluous segregation of data. In this work, 10 clinical respiratory-gated FDG-PET scans were signal-optimized and then analyzed. Results Qualitative analysis of the optimized images showed enhanced resolution and motion information compared to gated images (when case-specific motion supported this); however, image quality in regard to noise/quantum mottle was similar to that of ungated images. On average, the mean/SD in background ROIs was 2.0 times greater in optimized images than in the gated images. Motion of identifiable moving structures, defined as maximum displacement of the center of mass, was quantified on average to be preserved at 85% in the optimized images. Conclusions Adaptive filtering signal optimization algorithms, like the one we are presenting, utilize information that is not currently being used, to improve image quality in motion sensitive areas and offer an alternative strategy for 4D data optimization. Processing is fast, characterizable, and robust. We are presently endeavoring to collect a larger population of clinical scans and develop clinically relevant analysis. %U