Feasibility study of the quantitative corrections for the brain input function imaging from the carotid artery images by an ultra-high resolution dedicated brain PET

Quantitative PET imaging usually requires the arterial blood sampling, which is an invasive measure and may introduce risks or other complications to patients. People are trying several non-invasive methods to obtain the quantitative tracer concentrations by measuring the reconstructed intensity of the artery in the PET imaging. However, all these methods have certain limitations for brain study due to difficulties such as the partial-volume-effect (PVE), no artery big enough in the FOV for obtaining the required data as the cardiology does, etc. Here we carried a simulation study on the feasibility of the quantitative corrections by carotid artery with an ultra-high resolution, large axial FOV dedicated brain PET system. This brain PET has a detector ring diameter of 48 cm and the axial length of 25 cm. The large AFOV ensures that the camera could cover both the brain and the carotid artery region at the same time for dynamic study. The detectors are the 1.4×1.4×11 mm
³
LYSO crystals. The conservative estimation of the resolution is 1.7 to 2.0 mm, which is about 1/3 of the human carotid artery inner diameter. To evaluate the PVE on the quantitative results, a head-and-neck phantom with different-sized sources (5 to 20 mm) embedded that has a 6:1 concentration ratio between source and background is studied using Monte Carlo simulations. As the comparison, a whole-body PET - Siemens TruePoint scanner is also studied. From the reconstructed source intensities we find that with this brain PET, the recovery coefficient could reach 76% to 86% for a typical human carotid artery size source with the diameter between 5 to 7 mm; with the TruePoint scanner the recovery coefficient is only 34% to 54%. The simulation shows that with the help of an ultra-high resolution large axial FOV brain PET camera, the goal of non-invasive quantitative corrections by carotid artery for the brain dynamic study is feasible, which is not possible with other commercial whole-body scanners current available.