Program Information
Proton Stopping Power Ratio Estimation for Different Phantom Sizes Using Dual-Energy CT Scanners
V. Taasti1,2*, G. Michalak2, A. Deisher2, J. Kruse2, L. Muren1, J. Petersen1, D. Hansen1, C. McCollough2, (1) Aarhus University Hospital, Aarhus, Denmark, (2) Mayo Clinic, Rochester, MN
Presentations
SU-I-GPD-J-70 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: The accuracy of proton treatment planning (TP) is limited by several factors, including conversion from CT-number to stopping power ratio (SPR). Dual-energy CT (DECT) has been shown to increase accuracy compared to single-energy CT (SECT). However, these studies have been performed on dual-source scanners, which are not well-suited for proton therapy due to their limited field-of-view (FOV). This study investigated the accuracy of SPR estimates across patient sizes and DECT scanners, including a full-FOV scanner suitable for TP.
Methods: Tissue equivalent inserts from an electron density phantom (CIRS) were placed in torso-shaped water tanks of different sizes (15-45 cm lateral dimension). These were scanned with different DECT scanners: single-source AS-64 (two-scan technique) and dual-source FLASH and FORCE (Siemens Healthcare). SECT scans were also acquired and used in stoichiometric conversion. For DECT, SPR-values for the inserts were estimated using the Rho-Z software (Siemens Healthcare). Reference SPR-values were measured in a proton beam. To test for size dependence of the accuracy, the root-mean-square (RMS) of the relative errors for the inserts was calculated for each tank size.
Results: No clear trends in size dependencies were found for SPR deviations using the dual-source scanners, all RMS below 2.7%. In contrast, a small error increase was seen as size increased using the two-scan technique, RMS increased from 2.5% to 3.6%. However, in most cases, the errors were comparable to the dual-source scanners. Larger size dependence was seen for the SECT estimations, which generally produced the largest errors. The largest errors were found for lung and dense bone inserts; excluding these, dual-source scans produced lower errors than single source.
Conclusion: SPR estimation based on the Rho-Z software gave comparable results for single- and dual-source DECT scanners. Since single-source scanners are not limited by a small FOV, they may be useful for DECT-based proton TP.
Funding Support, Disclosures, and Conflict of Interest: Dr. McCollough received funding from Siemens Healthcare.
Contact Email: