Standardization and optimization of Siemens Biograph TruePoint PET/CT acquisition and reconstruction parameters: Simultaneous qualitative and quantitative assessments

Introduction: Data acquisition and image reconstruction protocols affect image quality and quantification accuracy in PET imaging. We aimed to standardize and optimize image acquisition, and reconstruction parameter sets using a simultaneous quantitative and qualitative assessment framework for a lutetium oxyorthosilicate (LSO)-based PET/CT scanner.Methods: The NEMA IEC Body Phantom acquisition was performed in list mode for ۱۰ minutes with four spheres to background ratios (SBRs). Raw PET data were reconstructed using ۶۰ different protocols. Image quality was evaluated for standardization using contrast, CNR, and noise. Recovery coefficient (RC) measurements were performed for different common VOI definitions.Results: No significant differences were observed between RCs for various acquisition durations. The contrast to noise ratio (CNR) increased at all SBRs by expanding the acquisition duration from ۶۰ to ۶۰۰ seconds. PET scan time was reduced to ۹۰ seconds per bed position while preserving image quality. Up to ۵۰% improvement in CNR for the highest sub-iteration with a high level of smoothing was observed. PSF-based reconstruction produced a positive bias of RCmax in high SBRs (۸ and ۱۰) using higher sub-iterations (۳۰ to ۶۰) with Gaussian filters less than ۶ mm FWHM. Moreover, a Sub-iteration of more than ۳۲ with a ۴-۶ mm FWHM Gaussian filter provides optimized reconstruction sets. Conclusion: Our study demonstrates it would be feasible for PET image acquisition and reconstruction settings to simultaneously allow optimal lesion detection with high image quality while providing accurate quantification.

Keywords:

PET/CT , Standardization , Optimization , Image quality