Wednesday, April 2, 2025
4/2/2025
Total-body PET Parametric Imaging using Relative Patlak Plot - Project Abstract/Summary Positron emission tomography (PET) with 18F-fluorodeoxyglucose (FDG) is an advanced molecular imaging technique to measure metabolic activities within the human body. Unlike standard static PET, which yields a semi-quantitative standard uptake value (SUV) for quantification, dynamic PET imaging combined with tracer kinetic modeling enables parametric imaging, offering critical physiological information such as metabolic rate. The standard Patlak plot, a simple yet efficient linear model, is widely used to describe FDG kinetics. The resulting slope parameter 𝐾i, which represents the FDG influx rate, has demonstrated significant advantages for cancer diagnosis and therapy assessment compared to SUV alone. Whole-body parametric imaging with the standard Patlak plot has been implemented on conventional PET scanners with an axial field- of-view (AFOV) ranging from 15-30 cm using a multi-bed and multi-pass acquisition strategy. The advent of total- body PET scanners with an AFOV greater than 1 meter, such as UIH EXPLORER and Siemens Quadra, has further simplified and improved the implementation for standard Patlak parametric imaging because of much- improved detection sensitivity and simultaneous coverage of multiple organs. However, despite these advancements, the full-time dynamic scan duration (e.g., 1 hour) has remained the same due to the need for an image-derived blood input function. In this project, we aim to develop an efficient total-body PET parametric imaging approach that can be applied to a short dynamic scan regenerated from the routine clinical static scan (e.g., a 20-minute or shorter scan at 1-hour post-injection). The enabling approach uses a Relative Patlak (RP) plot that does not require the early-time input function. The RP plot has recently been deployed on GE commercial scanners. Our early study and another recent clinical study have demonstrated the equivalence between the RP slope 𝐾i′ and the standard Patlak 𝐾i in clinically relevant tasks, such as lesion detection and tumor volume segmentation. However, one major challenge of applying this approach is the higher noise level in parametric images when the scan duration is shortened. Our preliminary work has indicated the potential of shortening the scan duration to 20 minutes for RP parametric imaging, making it possible to be directly applied to clinical static scans. The specific aims of this proposal are to further develop the enabling technique, push the method for shorter scans (e.g., 10 minutes), and evaluate the benefits of RP parametric images on top of SUV. Specifically, we will (1) develop a deep learning solution for improving RP parametric imaging within a shortened scan duration and (2) evaluate the benefits of RP parametric images on top of SUV using clinical cohorts. Completing these specific aims will improve the diagnosis and comfort of patients by providing an efficient total- body PET parametric imaging approach that is adaptive to clinical static scan protocols without adding any imaging time and scan costs. Our proposed approach will offer a unique solution for pediatric parametric imaging, as the full-time dynamic scan is impractical for pediatric patients.
HHS’ Tracking Accountability in Government Grants System (TAGGS) website provides access to detailed descriptions of grants, loans, aggregated direct payments and other types of financial assistance awarded by the HHS Operating Divisions (OPDIVs) and the Office of the Secretary Staff Divisions (STAFFDIVs).