Nuklearmedizin 2022; 61(02): 187
DOI: 10.1055/s-0042-1746087
Abstracts | NuklearMedizin 2022
WIS-Poster
Medizinische Physik/Radiochemie

Determining Partial Volume Effect in Patient-Specific 3D Printed Tc-99m-SPECT/CT Brain Phantom Scans

A. Grings
1   University Hospital Erlangen, Clinic of Nuclear Medicine, Erlangen
,
M. Reymann
2   Siemens Healthineers GmbH, Forchheim
,
A.H. Vija
3   Siemens Medical Solution USA, Molecular Imaging, Hoffman Estates
,
T. Kuwert
1   University Hospital Erlangen, Clinic of Nuclear Medicine, Erlangen
,
P. Ritt
1   University Hospital Erlangen, Clinic of Nuclear Medicine, Erlangen
› Institutsangaben
 

Ziel/Aim Uptake distributions of radioactive tracer in SPECT brain perfusion imaging deviates from the actual distribution due to the partial volume effect (PVE). To determine the PVE, we used two high-resolution, patient-specific, fillable, 3D printed brain phantoms.

Methodik/Methods The two phantoms were created based on data from two patients undergoing brain perfusion scans with Tc-99m-HMPAO-SPECT/CT with additional MRI imaging. Gray- and white matter were segmented using the SPM12 algorithm on the MRI images (t1 mprage sequence) and used to create fillable models for each patient. The models were printed with an Ultimaker 3 using custom profiles. PVE was determined based on a SPECT/CT acquisition of the phantoms filled with an uptake ratio of 2:1 (gray to white matter). Mean measured uptake was 115.6 kBq/mL and 60.25 kBq/mL for gray matter and white matter, respectively. We choose an approximately 5-fold increased activity level compared to patient admissions, for better evaluation of PVE. A fully quantitative SPECT reconstruction (xSPECT Quant) was performed using standard parameters. Reconstructions were evaluated by an expert reader (TK) for visual plausibility, the appearance of artifacts and their similarity to actual patient images. In addition, the Recovery Coefficient (RC) was determined for the two compartments.

Ergebnisse/Results Visual analysis revealed reasonable distribution of tracer in the brain phantoms compared to the actual patient images. The RC in the two brain phantoms was determined as 59.8/49.9% for gray matter and 94.9/62.3% for white matter. Lower RC in white matter of the second phantom could be explained by more fine structures of the gray matter and the resulting decrease wash out in this area.

Schlussfolgerungen/Conclusions We demonstrated that 3D printing can be used to create realistic and patient-specific brain phantoms, showing expected characteristics like PVE. Further research will determine suitability of the phantoms for partial volume correction.



Publikationsverlauf

Artikel online veröffentlicht:
14. April 2022

© 2022. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany