Improvement of likelihood estimation in Logan graphical analysis using maximum a posteriori for neuroreceptor PET imaging

ObjectiveTo reduce variance of the total volume of distribution (VT) image, we improved likelihood estimation in graphical analysis (LEGA) for dynamic positron emission tomography (PET) images using maximum a posteriori (MAP).MethodsIn our proposed MAP estimation in graphical analysis (MEGA), a set of time-activity curves (TACs) was formed with VT varying in physiological range as a template, and then the most similar TAC was sought out for a given measured TAC in a feature space. In simulation, MEGA was compared with other three methods, Logan graphical analysis (GA), multilinear analysis (MA1), and LEGA using 500 noisy TACs, under each of seven physiological conditions (from 9.9 to 61.5 of VT). PET studies of [11C]SA4503 were performed in three healthy volunteers. In clinical studies, the VT images estimated from MEGA were compared with region of interest (ROI) estimates from a nonlinear least square (NLS) fitting over four brain regions.ResultsIn the simulation study, the estimated VT by GA had a large underestimation (y = 0.27x + 8.72, r2 = 0.87). Applying the other methods (MA1, LEGA, and MEGA), these noise-induced biases were improved (y = 0.80x + 4.04, r2 = 0.98; y = 0.85x + 3.05, r2 = 0.99; y = 0.96x + 1.21, r2 = 0.99, respectively). MA1 and LEGA produced increased variance of the estimated VT in clinical studies. However, MEGA improved signal-to-noise ratio (SNR) in VT images with linear correlations between ROI estimates with NLS (y = 0.87x + 5.1, r2 = 0.96).ConclusionsMEGA was validated as an alternative strategy of LEGA to improve estimates of VT in clinical PET imaging.