PT - JOURNAL ARTICLE AU - Timothy Turkington AU - Stephen Lokitz AU - James Colsher TI - Use of tube current modulation to minimize radiation dose in CT-based PET attenuation correction DP - 2011 May 01 TA - Journal of Nuclear Medicine PG - 1984--1984 VI - 52 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/52/supplement_1/1984.short 4100 - http://jnm.snmjournals.org/content/52/supplement_1/1984.full SO - J Nucl Med2011 May 01; 52 AB - 1984 Objectives In some PET/CT studies the CT scans are used exclusively for PET attenuation correction (AC). This allows the use of a non-diagnostic (and lower dose) CT images. However, inadequate x-ray flux can lead to artificially low CT pixel values, resulting in quantitative inaccuracies in the corrected PET images. We have evaluated the use of an existing tube current modulation (TCM) algorithm that adjusts mA according to attenuation and body size as measured in the planar scout image to achieve a minimally acceptable technique for PET AC. Methods A solid Acrylic tapering oval phantom ranging in cross section from 22x12 cm2 to 49x39 cm2 was scanned on a 64-slice CT associated with a GE Discovery 690 PET/CT. Imaging was done at 120 kVp and 140 kVp, over a wide range of mAs values (4 - 500). Scans were were also performed with TCM (AutomA, noise index = 30, 50, and 70). Images were reconstructed with 70 cm FOV. A 10 cm circular ROI was applied in each slice to assess the mean CT number. The result was a mean CT value for all phantom slices for all scans. Results Using the fixed-mAs scans, it was found that from 4 mAs (small end) to 76 mAs (large end) was required to produce non-biased CT images(at 140 keV). All TCM scans eliminated the quantitative bias over the length of the phantom indicating sufficient flux. In addition, the acquisition with noise index of 70 closely matched the minimal mAs requirements throughout the phantom as determined by the fixed-mAs scans for both 120 kVp and 140 kVp. Conclusions The fixed-mAs CT protocols typically used for low-dose AC are not optimal since the actual mAs requirements ranges widely with body dimensions. A fixed protocol will either over-expose small regions, or yield quantitative inaccuracy in large areas. A TCM technique originally designed to provide uniform image noise has been found to work well at delivering the minimally required x-ray flux over a wide range of body sizes