Fast 4D-PET parametric imaging computation at the whole field of view level: Reliability under simulated conditions of PET KinetiX, a dedicated software solution

Abstract

Purpose: The reliability of a new academic software, PET KinetiX, designed for fast parametric 4D-PET imaging computation, is assessed under simulated conditions.

Methods: 4D-PET data were simulated using the XCAT digital phantom and realistic time-activity curves (ground truth). Four hundred analytical simulations were reconstructed using CASToR, an open-source software for tomographic reconstruction, replicating the clinical characteristics of two available PET systems with short and long axial fields of view (SAFOV and LAFOV). A total of 2,800 Patlak and 2TCM kinetic parametric maps of ¹⁸F-FDG were generated using PET KinetiX. The mean biases and standard deviations of the kinetic parametric maps were computed for each tissue label and compared to the biases of unprocessed SUV data. Additionally, the mean absolute ratio of kinetic-to-SUV contrast-to-noise ratio (CNR) was estimated for each tissue structure, along with the corresponding standard deviations.

Results: The K_i and v_b parametric maps produced by PET KinetiX faithfully reproduced the predefined multi-tissue structures of the XCAT digital phantom for both Patlak and 2TCM models. Image definition was influenced by the 4D-PET input data: a higher number of iterations resulted in sharper rendering and higher standard deviations in PET signal characteristics. Biases relative to the ground truth varied across tissue structures and hardware configurations, similarly to unprocessed SUV data. In most tissue structures, Patlak kinetic-to-SUV CNR ratios exceeded 1 for both SAFOV and LAFOV configurations. The highest kinetic-to-SUV CNR ratio was observed in 2TCM k₃ maps within tumor regions.

Conclusion: PET KinetiX currently generates K_i and v_b parametric maps that are qualitatively comparable to unprocessed SUV data, with improved CNR in most cases. The 2TCM k₃ parametric maps for tumor structures exhibited the highest CNR enhancement, warranting further evaluation across different anatomical regions and radiotracer applications.

Keywords: Digital phantom; Kinetic modeling; PET; Parametric imaging; Quantification; Simulation.

Fig. 1

Simulation process. XCAT digital phantom and true TACs were used to simulate realistic 4D-PET data with the Customizable and Advanced Software for Tomographic Reconstruction (CASToR). For this purpose, SAFOV and LAFOV-like PET data were simulated with low noise (OSEM algorithm with 2 iterations) and high spatial resolution (OSEM algorithm with 6 iterations)

Fig. 2

Data analyses. The 400 simulated 4D-PET data were processed individually with PET KinetiX to generate a total of 2 800 parametric maps of kinetic parameters at the whole FOV-level: 800 Patlak maps (400 K_i and v_b maps respectively) and 2 000 2 TCM maps (400 K₁, k₂, k₃, v_b and K_i maps respectively)

Fig. 3

Mean and SD of parametric maps. A, B, C, D: For each configuration, $\overline{X}$ and SD correspond respectively to the mean and standard deviations of the parameters estimated from the 100 simulations with PET KinetiX. A. Patlak Ki; B: Patlak v_b; C: 2 TCM Ki; D: 2 TCM v_b. And E: mean and standard deviations of the SUV variabilities estimated from the 100 simulations

Fig. 4

SUV biases. The bias of SUV within the arterial region of interest we used for image derived input function (AIF) are also provided here

Fig. 5

Quantitative noise-bias trade-off plots. Quantitative noise-bias trade-off plots on the 11 tissue structures over 100 noise replicates, for 2 and 6 iterations (circle and square markers respectively). For each analysis, SAFOV (dotted lines) and LAFOV (continuous lines)-configurations are compared. A. K_i (Patlak); B. K_i (2 TCM); and C. SUV. In all the cases, the dashed curve represents SAFOV, while the continuous curve represents LAFOV

Fig. 6

Kinetic to SUV CNR ratios. The red line corresponds to kinetic to SUV CNR ratio of 1. All the results above this line correspond to CNR_kinetic higher than CNR_SUV, whereas all the results under this line correspond to CNR_kinetic lower than CNR_SUV