ESTRO 2024 - Abstract Book

S3743

Physics - Image acquisition and processing

ESTRO 2024

Table1: Contrast, HU accuracy, and spatial resolution for Kyoto CBCT. Cells show mean and standard deviation (S.D.) over time period, for 2 non-HyperSight Ethos systems, compared to Truebeam data. Cells marked are significantly better than TrueBeam.

Contrast

HU prostate diff

HU Bones diff

Spatial resolution MTF@10% (lp/cm)

Ethos1 iOff Ethos2 iOff

3.5 (0.4) 3.7 (0.2)

-16.1 (1.2) -25.1 (1.5)

-2.6 (2.0) -8.3 (1.4)

4.49 (0.57) 3.04 (0.23)

Ethos1 iOn Ethos2 iOn

4.2 (0.5) 3.0 (0.2)

-2.3 (3.2)

2.6 (2.0)

2.87 (0.08) 5.51 (0.97)

-26.2 (2.1)

-8.8 (1.6)

TrueBeam

3.5 (0.1)

-18.0 (3.9)

50.8 (7.1)

3.50 (0.24)

Figure 1: left-right HU profiles of half beam (pelvis) acquisition mode, for a 20 cm water phantom.

Conclusion:

The study demonstrates that large phantoms, which require half beam acquisition, reveal differences between CBCT systems that smaller phantoms do not detect. Even between Ethos systems, results differ, and so does the added value of using the iterative option. Ethos systems have exactly the same hardware and protocols. A plausible explanation seems variation in calibration during QA, resulting in asymmetric left-right profiles. The effect of this variation on clinically experienced image quality needs to be investigated further.