Kinetic Modelling of Dynamic F-18-FDG Datasets from a Long Axial Field-of-view PET Scanner Using Model Selection Criteria with Deep Learning-Based Organ Segmentations

Ziel/Aim Dynamic PET protocols allow for full quantification of tracer kinetics. With the introduction of long axial field-of-view (LAFOV) PET systems, tracer kinetics of multiple organs of interests can be studied in a single-bed position. These advances enable use of non-linear compartmental kinetic models for estimation of kinetic microparameters from different structures of interest. However, the irreversible 2-tissue-compartment (2TC-3k) might cause artefacts in parametric images when applied to whole-body F-18-FDG datasets. Here, we propose a method that utilizes deep learning-based organ segmentations to select an appropriate kinetic model for each tissue of interest.