Impact of Patient Size on Cone-Beam CT Quantitative Image Quality Using Ethos Hypersight Imaging
Abstract
Purpose
To evaluate the impact of patient size on cone-beam CT (CBCT) image quality using iterative reconstruction approaches on the Varian Ethos HyperSight imaging system and to assess the adequacy of attenuation and scatter correction across patient sizes.
Methods
Abdominal CBCT images acquired with the HyperSight system using consistent kVp and mAs acquisition values for 30 distinct patients were retrospectively analyzed and grouped by patient cross-sectional area, estimated as π × (lateral × posterior dimension). Patients were categorized as small (3000 cm²). Regions of interest (ROIs) were placed in liver, vessels, and fat. Image quality metrics including signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) for liver–vessel and liver–fat contrast were calculated. Qualitative assessment of noise, artifacts, and soft-tissue visibility was also performed.
Results
Image quality quantitative metrics degraded systematically with increasing patient size. Small patients demonstrated low noise, stable HU values, and clear differentiation between liver, vessels, and fat, with SNR values of approximately 3–4. Medium patients showed moderate noise increase and reduced soft-tissue contrast, with SNR values of 1.5–2.5. Large patients exhibited marked noise, streak artifacts, unstable HU values, and poor soft-tissue visibility, with SNR frequently below 1 and occasionally negative, rendering SNR unreliable. Liver–vessel CNR showed a clear downward trend with increasing patient size, while liver–fat CNR was less sensitive but displayed increased variability in larger patients. These findings suggest noise escalation dominates image degradation in large patients despite iterative reconstruction.
Conclusion
Patient size is a dominant factor influencing CBCT image quality on the Ethos HyperSight system. Increasing patient size leads to increased noise, reduced CNR, and compromised vessel and organ visibility. The results indicate a need for patient-size–adaptive CBCT reconstruction and acquisition protocols. Future work will assess CBCT artifacts, artifact-reduction strategies, and noise power spectrum analysis to characterize size-dependent degradation.