Generalized population-based input function estimation given incomplete blood sampling in quantitative dynamic FDG PET studies

Objectives To validate and optimize a new population-based plasma input function (PIF) estimation method for quantification of dynamic FDG PET with sparsely sampled blood data.

Methods Eight 60-min monkey dynamic FDG PET studies with arterial blood sampling were performed. Time activity curves (TACs) of 7 cerebral regions of interests (ROIs) were generated from each study. A generalized population-based approach to recover full kinetics of the PIF from sparsely sampled blood data is proposed. The estimated PIF (ePIF) from the incomplete sampling data was determined by interpolation and extrapolation using scale-calibrated population mean of normalized PIFs. The leave-one-out cross validation method was used to generate a mean PIF for each subject. Patlak plot (t* = 10 min) was applied to ROI TACs to estimate the FDG uptake rate constant Ki. The Ki estimated from the measured PIF (mPIF) was used as reference. For a fixed number of total samples, the PIFs were estimated from different blood sampling schemes. The optimal sampling protocol was then determined by statistical analysis of the Ki estimates as obtained from ePIF vs. mPIF.

Results The linear correlations between the Ki estimates from the ePIF (with optimal sampling scheme) and those from the mPIF were: Ki(ePIF; 1 sample at 40 min) = 1.03 Ki(mPIF) -0.00, R² = 0.97; Ki(ePIF; 2 samples at 40 and 50 min) = 1.03Ki(mPIF) - 0.00, R² = 0.96; Ki(ePIF; 3 samples at 12, 40, and 50 min) = 1.05Ki(mPIF) - 0.00, R²= 0.98; and Ki(ePIF; 4 samples at 10, 25, 40, and 50 min) = 1.02Ki(mPIF)-0.00, R²=0.99. As the total sample size became >= 4, the Ki estimates from the ePIF (using the optimal sampling protocol) were almost identical to those from the mPIF.

Conclusions The generalized population-based PIF estimation method with optimal blood sampling scheme is a reliable method to estimate PIFs from incomplete blood sampling data for quantification of dynamic FDG PET using the Patlak plot.

Research Support This research was partially supported by the National Natural Science Foundation of China (NSFC 30870729 and 81071183)