Low-Dose Imaging in a New Preclinical Total-Body PET/CT Scanner

Abstract

Ionizing radiation constitutes a health risk to imaging scientists and study animals. Both PET and CT produce ionizing radiation. CT doses in pre-clinical in vivo imaging typically range from 50 to 1,000 mGy and biological effects in mice at this dose range have been previously described. [¹⁸F]FDG body doses in mice have been estimated to be in the range of 100 mGy for [¹⁸F]FDG. Yearly, the average whole body doses due to handling of activity by PET technologists are reported to be 3-8 mSv. A preclinical PET/CT system is presented with design features which make it suitable for small animal low-dose imaging. The CT subsystem uses a X-source power that is optimized for small animal imaging. The system design incorporates a spatial beam shaper coupled with a highly sensitive flat-panel detector and very fast acquisition (<10 s) which allows for whole body scans with doses as low as 3 mGy. The mouse total-body PET subsystem uses a detector architecture based on continuous crystals, coupled to SiPM arrays and a readout based in rows and columns. The PET field of view is 150 mm axial and 80 mm transaxial. The high solid-angle coverage of the sample and the use of continuous crystals achieve a sensitivity of 9% (NEMA) that can be leveraged for use of low tracer doses and/or performing rapid scans. The low-dose imaging capabilities of the total-body PET subsystem were tested with NEMA phantoms, in tumor models, a mouse bone metabolism scan and a rat heart dynamic scan. The CT imaging capabilities were tested in mice and in a low contrast phantom. The PET low-dose phantom and animal experiments provide evidence that image quality suitable for preclinical PET studies is achieved. Furthermore, CT image contrast using low dose scan settings was suitable as a reference for PET scans. Total-body mouse PET/CT studies could be completed with total doses of <10 mGy.

Keywords: CT; PET; [18F]FDG; imaging; low dose; oncology; preclinical; total-body.

Figure 1

The new PET/CT “Si78”.

Figure 2

High sensitivity total-body PET detector design. Mouse inside field of view.

Figure 3

Novel X-ray beam shaper for low-dose imaging.

Figure 4

Mouse NECR Phantom [¹⁸F]FDG starting 925 kBq signal quantified over 60-min decay.

Figure 5

IQ Phantom images acquired over 20 min with <1 MBq.

Figure 6

micro-Derenzo phantom imaged for 20 min with 3 MBq of [¹⁸F]-FDG. Capillary sizes in mm are shown in the image.

Figure 7

Mouse 1 of fibrosarcoma tumor model. Top: Tumor is marked by white arrow and heart by red arrow. Contrast and resolution are sufficient to resolve heterogenous tumor uptake, see orange arrow. Bottom: Low-dose total-body [¹⁸F]FDG/PET with clearly defined tumor uptake and contrast. Tumor uptake measurements (13% ID/mL) obtained fall within typical ranges for this model. Results for 10-minute and 60-minute scans are consistent, indicating that low-dose imaging even for 10 minutes provides sufficient counts for reliable data and quantifications.

Figure 8

PET MIP images showing bone tracer Na[¹⁸F]. Arrows point to visible ribs as an indication of good resolution. Tracer uptake time was 1 h and scan time 18 min. Each of the animals had been injected with 2 MBq.

Figure 9

(A–E) Dynamic FDG PET images: Representative images at 20 s, 1, 10, 30, and 56 min post FDG administration. (F) Time activity curves: Representative time activity curves for LV blood pool and myocardium, model fits, and model corrected blood input function (MCIF). The MCIF at 56 min agrees with a blood sample (8.67 kBq/cc) at the same time point.

Figure 10

Contrast phantom to evaluate contrast at low-doses (A) Scan time: 140 s and dose 12.5 mGy (B) 57 s and 9.9 mGy (C) 7 s and 5.8 mGy.

Figure 11

Water tube phantom scanned with 200 um resolution (left) and 100 um (resolution) in low dose setting mode.

Figure 12

(A–D) Above show the 4 sagittal views with current and acquisition time settings such that approximately the dose is halved starting at 25.8 and finishing at 2.6 mGy.

Figure 13

Mouse scanned at 39 kV and aluminum filtering of 1 mm. (a) On the left with typical imaging settings and (b) on the right, current and scan duration optimized to minimize dose.

Figure 14

Mouse imaged with 54 kVp, 900 μA, and an acquisition time of 7 s and dose <10 mGy. The image voxel is 200 um.