Quantitative Comparison of Hypersight CBCT Image Quality with Conventional CT Simulators for Therapy Planning
Abstract
Purpose
To quantitatively compare image quality from HyperSight CBCT to dedicated CT Simulation scanners, using a standard image quality phantom and conventional imaging metrics to assess performance differences and their statistical significance.
Methods
Catphan 604 scans were acquired for head, thorax, and pelvis protocols across each imaging system, which include a HyperSight CBCT on Ethos, a Siemens Biograph PET/CT scanner, and a Siemens diagnostic CT scanner using treatment planning protocols. Metrics analyzed included geometric distortion, resolution (lp/mm), uniformity (HU), contrast, noise, slice thickness, and HU accuracy for reference inserts (air, bone, Teflon, plastics) using SunCHECK QA software. Results were compared across scanners using one-way ANOVA followed by Tukey’s HSD post-hoc testing to identify significant pairwise differences.
Results
ANOVA demonstrated significant differences across three systems for geometric distortion, uniformity, HU calibration, and contrast (p 0.1). Ethos was slightly lower in spatial resolution (0.53 lp/mm) compared with PET/CT and diagnostic CT (0.64–0.66 lp/mm).
Conclusion
Conventional CT Simulators provided the most consistent HU accuracy and uniformity, serving as a benchmark. PET/CT exhibited higher noise but maintained stable HU values. HyperSight CBCT offered reduced noise in the thorax/pelvis protocol, but demonstrated systematic HU bias and degraded uniformity, particularly affecting bone and low-density materials.