Related Experiment Video
Updated: Sep 27, 2026
Irradiator Commissioning and Dosimetry for Assessment of LQ α and β Parameters, Radiation Dosing Schema, and in vivo Dose Deposition
Published on: March 11, 2021
Development and evaluation of a water-profile-based calibration method for 2D ionization chamber arrays in medical
Kihong Pak1, Sung-Woo Kim1, Byungchul Cho2
1Department of Radiation Oncology, Asan Medical Center, Songpa-gu, Seoul, Republic of Korea.
Purpose:
To develop and evaluate a water-profile-based (WP) calibration method for two-dimensional (2D) ionization chamber arrays used in medical linear accelerator (linac) quality assurance (QA), and to compare its accuracy, efficiency, and cross-energy applicability with the manufacturer-provided wide-field (WF) calibration method.
Methods:
Beam profiles were measured using an IC Profiler (ICP) and a 3D water scanner on three TrueBeam linacs (TB1-TB3) at five photon energies (4, 6, 10 MV flattened; 6, 10 MV flattening-filter-free [FFF]) with a 20 × 20 cm2 field size at a depth of 10 cm. A total of 240 profiles were acquired across eight independent measurement sessions. The WP method derived adjusted calibration factors (ACFs) by directly matching ICP detector responses to water-scanner reference profiles, and was evaluated under three strategies: (i) machine-specific (single linac reference), (ii) machine-integrated (reference pooled across three linacs), and (iii) cross-energy applicability across 20 inter-energy combinations. Performance metrics included per-detector relative differences (mean absolute error [MAE] and standard deviation [SD]) and beam profile characterization (symmetry and flatness). Paired t-tests (n = 30) compared the machine-integrated WP method with the WF method.
Results:
The WF method yielded a per-detector MAE of 0.31%, with symmetry and flatness MAEs of 0.49% and 0.44%, respectively. The WP method significantly reduced these uncertainties to 0.15-0.16% (per-detector), 0.21-0.27% (symmetry), and 0.27-0.28% (flatness), with medium-to-large effect sizes (Cohen's d = 0.55-0.99; all p < 0.01). Cross-energy application of WP-derived ACFs yielded a per-detector MAE of 0.25%, higher than that of energy-specific WP calibration but lower than that of the WF method in this dataset. Total calibration time decreased from approximately 75 min (WF) to 10-15 min (WP).
Conclusion:
For the tested 20 × 20 cm2 field on TrueBeam linacs with a CC13/Blue Phantom2 water-scanner reference, the proposed WP calibration method provided closer agreement with the water-scanner reference and a substantially shorter calibration time than the manufacturerprovided WF method. Machine-integrated calibration is recommended for routine multi-linac implementation, while energy-specific ACFs should be used whenever feasible to maximize accuracy.
