Nuklearmedizin 2023; 62(02): 107-108
DOI: 10.1055/s-0043-1766212
Abstracts | NuklearMedizin 2023
WIS-Vortrag
Medizinische Physik

Simulation Studies and Experimental Model Validation of the Biograph Vision Quadra

C. M. Pommranz
1   Eberhard Karls University Tuebingen, Department of Preclinical Imaging and Radiopharmacy, Tuebingen
,
E. Elmoujarkach
2   University of Lübeck, Institute of Medical Engineering, Lübeck
,
J. Cabello
3   Siemens Medical Solutions USA, Inc., Knoxville, TN
,
M. Rafecas
2   University of Lübeck, Institute of Medical Engineering, Lübeck
,
J. G. Mannheim
1   Eberhard Karls University Tuebingen, Department of Preclinical Imaging and Radiopharmacy, Tuebingen
,
A. Santangelo
4   Eberhard Karls University Tuebingen, Institute for Astronomy and Astrophysics, Tuebingen
,
C. la Fougère
5   University Hospital Tuebingen, Department of Nuclear Medicine and Clinical Molecular Imaging, Tuebingen
,
B. J. Pichler
1   Eberhard Karls University Tuebingen, Department of Preclinical Imaging and Radiopharmacy, Tuebingen
,
F. P. Schmidt
5   University Hospital Tuebingen, Department of Nuclear Medicine and Clinical Molecular Imaging, Tuebingen
› Author Affiliations
 

Ziel/Aim The new generation of total body PET scanners enables a broad field of innovative applications, such as low dose, late or simultaneous multi-organ imaging. The ground truth information and easy repeatability of simulations are highly suitable to address the need for the establishment of scan protocols or validation of corrections in this field with limited experience. The purpose of this study was to develop and validate a Monte Carlo simulation framework of the Biograph Vision Quadra.

Methodik/Methods The Geant4 Application for Emission Tomography (GATE) and a proprietary coincidence sorter were used to determine list-mode event data, which is reconstructed using the same software, e7tools (Siemens), that is used for the scanner. Simulated phantom studies following the NEMA guidelines including image quality (IQ), scatter and sensitivity assessment were performed and compared to real experimental measurements. Further patient-like simulations were performed using the XCAT phantom. Agreement between simulated and measured data sets was assessed by visual image comparison and with quantitative metrics, e.g. the contrast recovery coefficients (CRC) of a clinical image quality phantom (ratio 8:1, F-18, 3MBq/ml).

Ergebnisse/Results Preliminary qualitative comparison of the IQ for the phantom studies attested a good match between images and attenuation maps based on the simulation framework and their experimental counterparts. First quantitative evaluation confirmed this agreement, e.g. a CRC of 89.6% and 85.4% (37mm sphere), 85.5% and 81.8% (22mm sphere) and 73.7% and 60.4% (10mm sphere) for the simulation and the experiments, respectively.

Schlussfolgerungen/Conclusions The proposed simulation framework will be further tuned to match the scanner based on the NEMA validation. The final framework then can be used to foster the understanding of the impact of dose, patient position and scan time as well as provide ground truth information for machine learning training data sets or validation of new kinetic modeling methods.



Publication History

Article published online:
30 March 2023

© 2023. Thieme. All rights reserved.

Georg Thieme Verlag
Rüdigerstraße 14, 70469 Stuttgart, Germany