KEY POINTS
- This proof-of-concept technical study evaluated a dual-scan conformal cone-beam computed tomography method using a digital head phantom, simulated nasopharyngeal and prostate cancer datasets, and an anthropomorphic head phantom measured on a Varian Edge system.
- The protocol combined a low-dose full-field scan with a higher-dose conformal scan, using angle-dependent dynamic collimation, target-region-based intensity optimization, and regularized weighted least-squares reconstruction with anisotropic total-variation regularization.
- In the digital phantom, target-region structural similarity increased from 0.702 to 0.855, contrast-to-noise ratio from 0.514 to 1.463, and signal-to-noise ratios from 33.8/32.8 to 113.1/114.6 versus conventional filtered back-projection.
- In the simulated prostate case, contrast-to-noise ratio increased from 0.443 to 1.944, while signal-to-noise ratios increased from 25.6/23.3 to 74.6/82.4.
- In the measured anthropomorphic head phantom, contrast-to-noise ratio increased from 1.036 to 2.379, with signal-to-noise ratios improving from 42.8/50.1 to 89.8/98.0. Organ-specific imaging doses and clinical reconstruction times were not evaluated.
CLINICAL TAKEAWAY
Dual-scan conformal cone-beam computed tomography may improve visualization around selected targets while retaining full-field anatomy needed for setup verification and adaptive workflows. The evidence remains preclinical and technical: clinical translation requires patient validation, organ-dose quantification, automated reconstruction settings, workflow benchmarking, and compatible dynamic-collimation hardware.