Abstract
1679
Objectives: To compare 3 methods of sensitivity calibration for SPECT/CT-based Sm-153 EDTMP uptake quantification for tumor dosimetry.
Methods: SPECT/CT scans of known activities (STDs) were performed with 3 different source geometries: 1 mCi/10 cc vial in air (STDair) and an acrylic abdominal phantom (STDacr); and 12 mCi/6080 cc in a 20cm diameter cylinder (STDcyl). Five sets of scans were acquired on a 5/8" SPECT/CT system (Symbia T6, Siemens Medical Solutions) with clinical parameters (128×128, 21 s/view, 128 views/360°, LEHR collimation). A TEW imaging technique (103 keV/20% plus adjacent upper and lower scatter windows of equal widths) was used to estimate Compton plus bremsstrahlung scatter under the photopeak. Scatter window scale factors were derived from spectral analysis of a source in air and 5-15 cm of acrylic. Reconstructions (clinical parameters) were: 3D OSEM modeling resolution, scatter and CT-based attenuation, and 5mm FWHM Gaussian filter. STD volumes and counts were estimated from SPECT via count thresholding. The optimal threshold was that which yielded a volume within 5% of true. The impact of threshold level on STD volume and sensitivity was studied for all source geometries.
Results: The upper and lower scatter scale factors were 1.0 and 0.5. The optimal thresholds were 10%, 28% and 20% for STDair, STDacr and STDcyl, with 5% variation for each. The mean measured sensitivities were 2329, 2236 and 2048 counts/μCi for STDair, STDacr and STDcyl, with relative errors of 5%. The average sensitivities computed using STDair and STDcyl were +4% and –8% relative to STDacr.
Conclusions: All sensitivities were within 7% of the mean. The 4% difference between STDair and STDacr suggests proper scatter correction factors and system modeling. The -12% STDcyl sensitivity relative to STDair warrants further investigation.
- Society of Nuclear Medicine, Inc.