Derivation of input function from FDG-PET studies in small hearts

Area of Science:

• Nuclear Medicine
• Medical Imaging
• Pharmacokinetics

Background:

• Extracting pure arterial time-activity curves (TACs) from dynamic PET images in small animals is challenging due to cardiac chamber size and mixed signals.
• Factor analysis of dynamic structures (FADS) has shown limitations in accurately isolating these curves in small animal models.

Purpose of the Study:

• To evaluate the efficacy of FADS for extracting pure blood-pool TACs from dynamic PET studies in small animals.
• To assess the impact of a single late blood sample constraint on FADS performance.

Main Methods:

• Utilized digital phantoms and dynamic FDG PET studies in small monkeys (n=4) with varying left ventricle sizes.
• Applied FADS with and without a single late blood sample constraint to extract blood-pool TACs.
• Compared FADS-derived TACs and kinetic parameters with arterialized well counter measurements and plasma sample data.

Main Results:

• FADS with a blood sample constraint (FADS(+)) significantly reduced spillover fractions (<3%) compared to FADS without (<91%) in phantom studies.
• FADS(+) blood-pool TACs in monkey studies showed good agreement with arterialized well counter measurements (% differences <9%).
• Kinetic parameters derived from FADS(+) TACs closely matched those obtained from plasma samples in three-compartment model fitting.

Conclusions:

• FADS, when incorporating a single late blood sample, can effectively extract pure blood-pool TACs from dynamic PET images of small animals.
• This method eliminates the need for multiple blood samplings, ROI definition, or spillover correction.
• The enhanced FADS technique offers a promising approach for accurate pharmacokinetic analysis in small animal PET imaging.