Dosimetry and adequacy of attenuation correction with low-dose CT for pediatric SPECT/CT

337 Objectives: To evaluate the dosimetry and the adequacy of attenuation correction using low-dose CT in pediatric SPECT/CT. Methods: Measurements were made using a Philips Precedence 6-slice SPECT/CT scanner, 5 anthropomorphic abdominal phantoms (newborn, 1, 5, 10 year old and medium-sized adult), and an ion chamber. We evaluated the dose received from CT for a variety of acquisition parameters including tube current of 200 mA, tube voltages of 90, 120, 140 kVp, gantry rotation times of 0.75, 1.0 and 1.5 s per rotation and pitches of 0.656:1, 0.90:1 and 1.41:1. A phantom study was used to evaluate the CT noise and its impact on the reconstructed SPECT data. CT-based attenuation correction (CTAC) was applied to a SPECT scan of a 20-cm cylindrical phantom using scans acquired with a variety of acquisition parameters (tube currents of 15, 30, 50, 100, 150 mAs; tube voltages of 90, 120 and 140 kVp). Results: The anthropomorphic volume CT dose index (CTADIvol) varied by a factor of 35 for each phantom over the range of acquisition parameters (e.g. 0.86 and 29.35 mGy for 90 kVp/15 mAs and 140 kVp/160 mAs, respectively for a 10 year old). CTADIvol of a newborn was about twice that for the adult for the same acquisition parameters. Although the noise in the CT scan varied significantly over the range of acquisition parameters, its contribution to PET noise was very minimal (> 1%) and data reconstructed with a CT of 90 kVp/15 mAs was practically identical to that acquired with 140 kVp/150 mAs. Conclusions: For pediatric patients, adequate CT-based attenuation correction can be obtained with very low dose CT (90 kVp, 15 mAs). Such a CT scan leads to a radiation dose, as much as a factor of 35 less than that of a diagnostic quality CT.