Nuklearmedizin 2020; 59(02): 129-130
DOI: 10.1055/s-0040-1708248
Wissenschaftliche Vorträge
Medizinische Physik
© Georg Thieme Verlag KG Stuttgart · New York

Design and Simulation of a high-resolution and high-sensitivity BrainPET insert for 7T MRI

C Lerche
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
M Lenz
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
W Bi
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
J Scheins
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
L Tellmann
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
CH Choi
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
ER Kops
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
,
J Felder
2   Affinity Imaging, Jülich, Germany
,
D Arutinov
3   Central Institute of Engineering, Electronics and Analytics, ZEA-2, Electronic Systems, Forschungszentrum, Jülich, Germany
,
S Völkel
3   Central Institute of Engineering, Electronics and Analytics, ZEA-2, Electronic Systems, Forschungszentrum, Jülich, Germany
,
R Heil
3   Central Institute of Engineering, Electronics and Analytics, ZEA-2, Electronic Systems, Forschungszentrum, Jülich, Germany
,
J Collienne
4   Central Institute of Engineering, Electronics and Analytics, ZEA-1 Engineering and New Technologies, Forschungszentrum, Jülich, Germany
,
D Grunwald
4   Central Institute of Engineering, Electronics and Analytics, ZEA-1 Engineering and New Technologies, Forschungszentrum, Jülich, Germany
,
B Weissler
5   Physics of Molecular Imaging System/Institute of Experimental Molecular Imaging, RWTH, Aachen, Germany
11   Hyperion Hybrid Imaging Systems GmbH, Aachen, Germany
,
F Müller
5   Physics of Molecular Imaging System/Institute of Experimental Molecular Imaging, RWTH, Aachen, Germany
,
D Schug
5   Physics of Molecular Imaging System/Institute of Experimental Molecular Imaging, RWTH, Aachen, Germany
11   Hyperion Hybrid Imaging Systems GmbH, Aachen, Germany
,
V Schulz
5   Physics of Molecular Imaging System/Institute of Experimental Molecular Imaging, RWTH, Aachen, Germany
11   Hyperion Hybrid Imaging Systems GmbH, Aachen, Germany
12   III. Physikalisches Institut B, RWTH Aachen, Aachen, Germany
,
JL Lefaucheur
6   Inviscan Imaging Systems SA, Strasbourg, France
,
Z Chen
7   Monash Biomedical Imaging, Monash University, Clayton, Australia
,
G Egan
7   Monash Biomedical Imaging, Monash University, Clayton, Australia
,
NJ Shah
1   Institute of Neuroscience and Medicine 4, INM-4, Forschungszentrum, Jülich, Germany
8   Institute of Neuroscience and Medicine 11, INM-11, JARA, Forschungszentrum, Jülich, Germany
9   JARA - BRAIN - Translational Medicine, Aachen, Germany
10   Department of Neurology, RWTH Aachen University, Aachen, Germany
› Author Affiliations
Further Information

Publication History

Publication Date:
08 April 2020 (online)

 

Ziel/Aim Within the Helmholtz Validation Fund Project “Next generation BrainPET scanner for 7T MRI”, we aim at building a UHF-MRI compatible, BrainPET insert prototype for dedicated human neuroimaging. The BrainPET 7T insert was designed to achieve a sensitivity at the isocentre of » 12 %, a homogeneous spatial image resolution between 1.5 and 2.0 mm over the whole human brain, a coincidence timing resolution £ 500 ps, and an axial field of view of » 24 cm.

Methodik/Methods For achieving homogeneous spatial image a staggered layer pixelated scintillation detector design was chosen. Three layers of LSO scintillator with 24 × 24, 23 × 24, and 22 × 23 pixels, respectively, all with pixel pitch of 2.0 mm in both directions, will be used. A digital readout electronic using highly integrated digital SiPMs with 12 × 12 channels per tile have been developed, which are shielded using a carbon fiber reinforced plastic to avoid interferences with MR. SiPMs and PET module electronics will be thermally stabilized with an MR compatible liquid cooling system. An ML positioning algorithm will be used for crystal pixels identification and g-ray energies estimation. Image reconstruction will be carried with PRESTO which allows for OP-OSEM and median root prior for effective noise suppression.

Ergebnisse/Results MC simulations with GATE of sensitivity and spatial resolution have been done, including intrinsic detector energy resolution of 12 %, inter-crystal Compton scatter, CoG of energy deposition, positron range, g attenuation and non-collinearity, and a typical energy window of 388-634 keV. A sensitivity of » 12 % at the isocentre could be confirmed by simulating a low activity source. Spatial image resolution below 2.0 mm over the whole human brain was confirmed with a simulation of a Derenzo phantom with 20 cm diameter where the sector with a 1.6 mm grid could be resolved.

Schlussfolgerungen/Conclusions The simulation study confirmed, that the envisioned design allows to reach the target PET image quality specifications.