Fast generation of 4D PET-MR data from real dynamic MR acquisitions

Abstract

We have implemented and evaluated a framework for simulating simultaneous dynamic PET-MR data using the anatomic and dynamic information from real MR acquisitions. PET radiotracer distribution is simulated by assigning typical FDG uptake values to segmented MR images with manually inserted additional virtual lesions. PET projection data and images are simulated using analytic forward projections (including attenuation and Poisson statistics) implemented within the image reconstruction package STIR. PET image reconstructions are also performed with STIR. The simulation is validated with numerical simulation based on Monte Carlo (GATE) which uses more accurate physical modelling, but has 150× slower computation time compared to the analytic method for ten respiratory positions and is 7000× slower when performing multiple realizations. Results are validated in terms of region of interest mean values and coefficients of variation for 65 million coincidences including scattered events. Although some discrepancy is observed, agreement between the two different simulation methods is good given the statistical noise in the data. In particular, the percentage difference of the mean values is 3.1% for tissue, 17% for the lungs and 18% for a small lesion. The utility of the procedure is demonstrated by simulating realistic PET-MR datasets from multiple volunteers with different breathing patterns. The usefulness of the toolkit will be shown for performance investigations of the reconstruction, motion correction and attenuation correction algorithms for dynamic PET-MR data.

Partial Volume Correction using Structural–Functional Synergistic Resolution Recovery: Comparison with Geometric Transfer Matrix Method

“…A fast analytic PET scanner simulation toolkit was then used to generate realistic dynamic projection data for the ECAT EXACT3D PET tomograph, following the method described in Tsoumpas et al 30,31 This simulation is based on the Software for Tomographic Image Reconstruction Release 2 (STIR) package, 32 derived from the concept first described by Ma et al 33,34 A recent study using this approach has shown that the most important cause of bias in the accuracy of kinetic parameters was PVEs. 35 Attenuation coefficient projection data were derived using the measured transmission data.…”

A 5D computational phantom for pharmacokinetic simulation studies in dynamic emission tomography

“…To simulate different types of realistic motion during a dynamic scan, the fast analytic simulation toolkit (FAST) is used [15]. A dynamic 4D MRI scan, providing uniform temporal sampling over consecutive respiratory cycles during the dynamic scan, is used to generate the motion fields.…”

“…From the difference image between the 2 UTE images, a number of regions of interest were segmented (soft tissue, cortical bones, liver and lungs) and used to construct the anatomical phantom. The myocardium, heart ventricles and large vessels were also segmented using a different ECG triggered balanced B-TFE MRI scan during free breathing (TR/TE 4.7 ms/2.36 ms, TFE factor 26, o ) [15,31]. The scan was subsequently respiratory gated again to the end-exhale position using a virtual navigator.…”

“…The NCAT phantom and the latest generation in this family of phantoms, the extended cardiac-torso phantom (XCAT) [13], were generated from multi-detector respiratory-gated CT data, to model cardiac and respiratory motion. Similarly, dynamic MRI has also been used to derive the motion information, which is then used in a four-dimensional (4-D) simulation framework [14,15]. Although such 4-D phantoms combine accurate anatomical information with models of temporally periodic physiological processes, they do not take into account the variable and temporally non-periodic functional processes occurring during the course of the study, constraining the level of realism and thus their potential application in dynamic studies.…”

“…However, the proposed methodology counterbalances this by benefiting from a fast simulation environment, with semi-automatic segmentation techniques, making it possible to simulate personalized dynamic protocols [39]. Having the MR data available, construction of the anatomical phantom, kinetic modelling simulation, motion field estimation and application, and forward projection to generate the raw emission and attenuation data should take between 20 and 25 hours [15]. However the majotiy of this time is taken by the forward projection step and depends on the imaging system, the number of respiratory gates and the kinetic modelling framing being simulated.…”