Validation of cardiac image-derived input functions for functional PET quantification

• DOI: 10.1007/s00259-024-06716-8
• Publication Date: 2024-04-27T04:02:20Z
• Authors: Murray Bruce Reed; Patricia Anna Han...; Clemens Schmidt; Matej Murgaš; David Gomola; Christian Milz; Sebastian Klug; Benjamin Eggersto...; Lisa Aichinger; Godber Mathis God...; Lukas Nics; Tatjana Traub-Wei...; Marcus Hacker; Rupert Lanzenberger; Andreas Hahn
• Keywords: Gold standard (test); Blood sampling

Abstract Purpose Functional PET (fPET) is a novel technique for studying dynamic changes in brain metabolism and neurotransmitter signaling. Accurate quantification of fPET relies on measuring the arterial input function (AIF), traditionally achieved through invasive blood sampling. While non-invasive image-derived functions (IDIF) offer an alternative, they suffer from limited spatial resolution field view. To overcome these issues, we developed validated scan protocol utilizing cardiac IDIF, aiming to mitigate known IDIF limitations. Methods Twenty healthy individuals underwent fPET/MR scans using [ 18 F]FDG or 6-[ F]FDOPA, bed motion shuttling capture task-induced changes. Arterial venous sampling was used validate IDIFs. Participants performed monetary incentive delay task. IDIFs various pools composites estimated linear fit over all (3VOI) further supplemented with samples (3VOIVB) were compared AIF. Quantitative task-specific images both tracers assess performance each gold standard. Results For radiotracer cohorts, moderate high agreement (r: 0.60–0.89) between AIF cohorts observed, improvement 0.87–0.93) composite (3VOI 3VOIVB). Both methods showed equivalent quantitative values 0.975–0.998) AIF-derived measurements. Conclusion Our proposed enables accurate estimation full changes, addressing limitations imaging by larger thorax. These advancements increase applicability any scanner clinical research setting reducing experimental complexity increasing patient comfort.