Program Information
High-Speed KV-CBCT-Imaging with a Faster Linac Gantry Rotation: 3D Imaging Within One Breath-Hold
A Arns*, J Fleckenstein, F Schneider, V Steil, F Wenz, H Wertz, Universitaetsmedizin Mannheim, Heidelberg University, Mannheim, Baden Wuerttemberg, Germany
Presentations
SU-K-FS4-11 (Sunday, July 30, 2017) 4:00 PM - 6:00 PM Room: Four Seasons 4
Purpose: Gantry speed for CBCT patient positioning imaging was accelerated to 18°/s to limit acquisition time to 10-20s. For proof-of-concept, customer acceptance tests were performed and compared to clinically-established CBCT regarding (1) image quality, (2) dosimetry, and (3) registration accuracy.
Methods: Gantry speed of a research linac equipped with on-board kV-CBCT (VersaHD with XVI, Elekta, Stockholm, Sweden) was accelerated to 18°/s. Comparison tests of faster CBCT (18°/s) with conventional, clinical CBCT (3°/s) were performed. (1) Image quality was analyzed following customer acceptance tests (spatial resolution, uniformity, low contrast visibility, geometry) with a CatPhan-phantom and dedicated QA presets in slow and faster gantry speed. (2) Dosimetry and (3) registration accuracy were determined for different clinical presets at both gantry speeds. The CT dose index (CTDI) was measured. Registration accuracy was analyzed by scanning and automatically registering an off-centered ballbearing-phantom.
Results: Comparison tests for (1) image quality resulted in a spatial resolution of 12 line pairs per centimeter for both gantry speeds and low contrast visibility of 2.7% (slow gantry) versus 8.0% (faster gantry), with specification <3.0%. Uniformity and geometry passed the customer acceptance criteria for both speeds. (2) The CTDI for different clinical presets was reduced by a factor of 2.5-6.1 (59.5-83.7%) with faster gantry rotation; less reduced dose was measured in the rotational start- and endpoint due to gantry braking characteristics. (3) Registration accuracy of different clinical presets was (0.5±0.2)mm for both speeds.
Conclusion: With accelerated CBCT gantry speed, image quality was maintained; only contrast was out of tolerance due to undersampling, however still sufficient. Dose output was considerably reduced. Registration accuracy was maintained below 1mm. Thus, first tests of CBCT-imaging with faster gantry rotation showed promising results. Further improvements on projection acquisition (faster panel readout) would lead to clinical applicability for high-speed imaging, particularly important for breath-hold treatment strategies.
Funding Support, Disclosures, and Conflict of Interest: The department receives grants from Elekta AB, Stockholm, Sweden, during the conduct of this study.
Contact Email: