Program Information

The Impact of CT Image Used for Attenuation Correction in 4D-PET

Y Cui

Y Cui*, J Bowsher , S Yan , J Cai , S Das , F Yin , Duke University Medical Center, Durham, NC

Presentations

SU-C-9A-6 Sunday 1:00PM - 1:55PM Room: 9A

Purpose: To evaluate the appropriateness of using 3D non-gated CT image for attenuation correction (AC) in a 4D-PET (gated PET) imaging protocol used in radiotherapy treatment planning simulation.

Methods: The 4D-PET imaging protocol in a Siemens PET/CT simulator (Biograph mCT, Siemens Medical Solutions, Hoffman Estates, IL) was evaluated. CIRS Dynamic Thorax Phantom (CIRS Inc., Norfolk, VA) with a moving glass sphere (8 mL) in the middle of its thorax portion was used in the experiments. The glass was filled with ¹⁸F-FDG and was in a longitudinal motion derived from a real patient breathing pattern. Varian RPM system (Varian Medical Systems, Palo Alto, CA) was used for respiratory gating. Both phase-gating and amplitude-gating methods were tested. The clinical imaging protocol was modified to use three different CT images for AC in 4D-PET reconstruction: first is to use a single-phase CT image to mimic actual clinical protocol (single-CT-PET); second is to use the average intensity projection CT (AveIP-CT) derived from 4D-CT scanning (AveIP-CT-PET); third is to use 4D-CT image to do the phase-matched AC (phase-matching-PET). Maximum SUV (SUVmax) and volume of the moving target (glass sphere) with threshold of 40% SUVmax were calculated for comparison between 4D-PET images derived with different AC methods.

Results: The SUVmax varied 7.3%±6.9% over the breathing cycle in single-CT-PET, compared to 2.5%±2.8% in AveIP-CT-PET and 1.3%±1.2% in phase-matching PET. The SUVmax in single-CT-PET differed by up to 15% from those in phase-matching-PET. The target volumes measured from single-CT-PET images also presented variations up to 10% among different phases of 4D PET in both phase-gating and amplitude-gating experiments.

Conclusion: Attenuation correction using non-gated CT in 4D-PET imaging is not optimal process for quantitative analysis. Clinical 4D-PET imaging protocols should consider phase-matched 4D-CT image if available to achieve better accuracy.



Contact Email: