Cross-validation study between the HRRT and the PET component of the SIGNA PET/MRI system with focus on neuroimaging

Julia G Mannheim; Ju-Chieh Kevin Cheng; Nasim Vafai; Elham Shahinfard; Carolyn English; Jessamyn McKenzie; Jing Zhang; Laura Barlow; Vesna Sossi

doi:10.1186/s40658-020-00349-0

Cross-validation study between the HRRT and the PET component of the SIGNA PET/MRI system with focus on neuroimaging

EJNMMI Phys. 2021 Feb 26;8(1):20. doi: 10.1186/s40658-020-00349-0.

Authors

Julia G Mannheim^#^{1

2

3}, Ju-Chieh Kevin Cheng^#^{4

5}, Nasim Vafai⁵, Elham Shahinfard⁵, Carolyn English⁵, Jessamyn McKenzie⁶, Jing Zhang⁷, Laura Barlow⁸, Vesna Sossi⁴

Affiliations

¹ Department of Physics and Astronomy, University of British Columbia, Vancouver, British Columbia, Canada. julia.mannheim@med.uni-tuebingen.de.
² Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, Eberhard-Karls University Tuebingen, Tuebingen, Germany. julia.mannheim@med.uni-tuebingen.de.
³ Cluster of Excellence iFIT (EXC 2180) "Image Guided and Functionally Instructed Tumor Therapies", University of Tuebingen, Tuebingen, Germany. julia.mannheim@med.uni-tuebingen.de.
⁴ Department of Physics and Astronomy, University of British Columbia, Vancouver, British Columbia, Canada.
⁵ Pacific Parkinson's Research Centre, University of British Columbia, Vancouver, British Columbia, Canada.
⁶ Djavad Mowafaghian Centre for Brain Health, Pacific Parkinson's Research Centre, University of British Columbia & Vancouver Coastal Health, Vancouver, British Columbia, Canada.
⁷ Global MR Applications & Workflow, GE Healthcare Canada, Vancouver, British Columbia, Canada.
⁸ UBC MRI Research Centre, University of British Columbia, Vancouver, British Columbia, Canada.

^# Contributed equally.

Abstract

Background: The Siemens high-resolution research tomograph (HRRT - a dedicated brain PET scanner) is to this day one of the highest resolution PET scanners; thus, it can serve as useful benchmark when evaluating performance of newer scanners. Here, we report results from a cross-validation study between the HRRT and the whole-body GE SIGNA PET/MR focusing on brain imaging. Phantom data were acquired to determine recovery coefficients (RCs), % background variability (%BG), and image voxel noise (%). Cross-validation studies were performed with six healthy volunteers using [¹¹C]DTBZ, [¹¹C]raclopride, and [¹⁸F]FDG. Line profiles, regional time-activity curves, regional non-displaceable binding potentials (BP_ND) for [¹¹C]DTBZ and [¹¹C]raclopride scans, and radioactivity ratios for [¹⁸F]FDG scans were calculated and compared between the HRRT and the SIGNA PET/MR.

Results: Phantom data showed that the PET/MR images reconstructed with an ordered subset expectation maximization (OSEM) algorithm with time-of-flight (TOF) and TOF + point spread function (PSF) + filter revealed similar RCs for the hot spheres compared to those obtained on the HRRT reconstructed with an ordinary Poisson-OSEM algorithm with PSF and PSF + filter. The PET/MR TOF + PSF reconstruction revealed the highest RCs for all hot spheres. Image voxel noise of the PET/MR system was significantly lower. Line profiles revealed excellent spatial agreement between the two systems. BP_ND values revealed variability of less than 10% for the [¹¹C]DTBZ scans and 19% for [¹¹C]raclopride (based on one subject only). Mean [¹⁸F]FDG ratios to pons showed less than 12% differences.

Conclusions: These results demonstrated comparable performances of the two systems in terms of RCs with lower voxel-level noise (%) present in the PET/MR system. Comparison of in vivo human data confirmed the comparability of the two systems. The whole-body GE SIGNA PET/MR system is well suited for high-resolution brain imaging as no significant performance degradation was found compared to that of the reference standard HRRT.

Keywords: Binding potentials; Cross-validation study; HRRT; PET/MR; Recovery coefficients.

Grants and funding

240670-13/Natural Sciences and Engineering Research Council grant