A phantom investigation into the use of a lower dose 2DkV imaging preset to image a head or head and neck in Proton Beam Therapy and its impact on observer matching.

Gibson, Ellen and Kirby, Mike ORCID: 0000-0001-9765-5641 (2026) A phantom investigation into the use of a lower dose 2DkV imaging preset to image a head or head and neck in Proton Beam Therapy and its impact on observer matching. 12 February 2026, London.

PDF BIR 2026 Presentation Final.pdf - Published version Download (1MB) | Preview

Description

Introduction: Proton Beam Therapy (PBT) offers highly conformal treatment, often requiring different imaging approaches compared to photon radiotherapy. Daily 2DkV imaging is standard across all treatment sites. PBT’s advantage lies in sparing healthy tissue, which is critical for paediatric patients due to their increased radiosensitivity and the need to minimize dose to reduce stochastic effects such as secondary malignancies. A study centre adopted a change in practice to reduce imaging dose during 2DkV acquisition by using the lower-dose ‘Extremity – Normal’ preset instead of the protocol ‘Head – Normal’ preset to image the head and head and neck of paediatric patients. An investigation was performed to determine if observers’ ability to accurately match images was impacted. Methods: A head phantom was imaged while the treatment couch was displaced incrementally from maximum positive to maximum negative positions. This was performed across all parameters. Images were acquired at each increment using both presets, and absolute errors were calculated for comparison. Next, the phantom was randomly displaced in six degrees of freedom, and five operators repeatedly matched images from both presets. The study analysed inter and intra-observer variability to identify any significant differences in matching accuracy between the presets. Results: Clinical insignificance was defined by departmental tolerances of 0.2 cm for translations and 1.0 degree for rotations. Statistically significant differences in error were found for the longitudinal parameter (0.029 cm for Head vs. 0.042 cm for Extremity preset) and pitch (0.017 vs. 0.028 degrees), these were below clinical thresholds. Intra-observer variability was statistically significant for vertical, lateral, and rotational parameters, but errors did not consistently favour either preset. Maximum intra-observer errors were 0.30 cm and 0.352 degrees, both within clinical tolerance. Inter-observer variability showed statistically significant differences in vertical and lateral parameters, yet differences were clinically insignificant. The largest inter-observer errors recorded were 0.029 cm translation and 0.607 degrees rotation, both below tolerance levels. Conclusion: Findings indicate no clinically significant difference in matching accuracy between the Head – Normal and Extremity – Normal 2DkV presets. This supports continued use of the lower dose Extremity - Normal preset at the study centre to reduce imaging dose delivered to paediatric patients without compromising image matching quality.

Item Type:	Presentation Material
Divisions:	Faculty of Health & Life Sciences Faculty of Health & Life Sciences > Inst. Population Health Faculty of Health & Life Sciences > Inst. Population Health > School of AHPs and Nursing
Depositing User:	Symplectic Admin
Date Deposited:	16 Feb 2026 16:56
Last Modified:	16 Feb 2026 16:56
DOI:	10.13140/RG.2.2.35440.11521
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3197047
Disclaimer:	The University of Liverpool is not responsible for content contained on other websites from links within repository metadata. Please contact us if you notice anything that appears incorrect or inappropriate.