Abstract
242546
Introduction: PET and single photon (SP) imaging are used together to plan and evaluate treatment response for radionuclide therapies. Following a diagnostic PET study, effectiveness of radionuclide (beta or alpha emitter) therapy over several cycles is evaluated with SP imaging. Using planar imaging only (and not SPECT) is due to imaging large portions of the body and time constraints. Some therapeutic radionuclides provide a simultaneous imaging opportunity due to x-ray/gamma emissions in addition to alpha/beta particles. A comparison of tumor detectability between modalities and different radionuclides provides a baseline for visually comparing images in nuclear medicine studies. A whole-body phantom was designed to investigate the use of three radionuclides: 18F (PET), 99mTc (SP), and 211At (SP imaging of an alpha emitter).
Methods: Three imaging systems were used: GE Discovery NM/CT 670, GE Discovery 690 PET/CT, and GE Discovery MI PET/CT (5-ring).
A 25 L phantom simulating a medium/large-sized patient (WB) with a 36 cm x 21 cm oval cross section and 40 cm length with a cross sectional plate securing 20 spheres on rods was used. Each side of the plate secured 10 spheres of three diameters at varying depths since SP image quality varies greatly with distance from the collimator (resolution) and depth (attenuation): 3x 3 cm, 3x 2 cm, and 4x 1 cm. The plate separated sphere to background concentrations: 10:1 and 5:1.
Three different radionuclides were used: 18F, 99mTc, and 211At. Due to 211At decay resulting in low-yield 500-900 keV gamma rays in addition to highly abundant 77-92 keV x-rays used for imaging, it has very challenging imaging properties. Considering the phantom volume and some allowance for radioactive decay during patient uptake, the activity used in the phantom for each radionuclide was approximately 25% of the activity typically administered in patients: 4.3, 5.8, and 2.5 mCi.
The 18F PET scan was 3 min. Images were reconstructed with TOF information, AC from CT, and other corrections. Low-energy high resolution (LEHR) collimators were used to image 99mTc, and medium-energy general purpose (MEGP) collimators were used to image 211At, based on previous work that evaluated the impact of high-energy emissions that penetrate the collimator. For both 99mTc and 211At, SP planar images (anterior and posterior) of the phantom were taken for 75, 150, 300, and 600 seconds when the activity in the phantom was 25% of the respective clinical dose.
Image analysis was performed using ImageJ. A circular ROI was placed over visible spheres for each image set to measure mean pixel value. Several large circular ROIs were used to determine average background value. A ratio of the sphere to background was calculated and compared to the expected sphere to background ratio.
Results: All 20 spheres of both concentrations were identifiable in the phantom when imaged with PET and 18F. A total of 11 spheres were identifiable in the 10 min SP planar image using 99mTc, while only 5 spheres were identifiable in the 10 min SP planar image using 211At.
The difference between the sphere to background concentration ratio of the two sets of spheres is the most prominent when imaging with PET and 18F. The sphere to background ratio of the 3 cm spheres when imaging with 211At for both 10:1 and 5:1 spheres are comparable to the 3 and 2 cm spheres when imaging with 99mTc with a 5:1 concentration.
Conclusions: When comparing PET and SP, 2 and 1 cm resolvable objects in PET were a challenge in SP. Given the same concentration, sphere to background ratio is much smaller in planar SP scans which visually affects the contrast of the image. As expected, 211At small lesion image quality is greatly reduced compared to 99mTc, to the degree that only 3 cm lesions can be discerned.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.