Related Experiment Video
Updated: Sep 10, 2025
Expedited Radiation Biodosimetry by Automated Dicentric Chromosome Identification ADCI and Dose Estimation
Published on: September 4, 2017
Investigation of an EPID-based dose verification system for patient-specific quality assurance in a harmony device
Dohyeon Yoo1, Hyunggun Lee1,2, Sangmin Lee1
1Department of Radiation Oncology, Yonsei Cancer Center, Yonsei University, College of Medicine & Heavy Ion Therapy Research Institute, Yonsei University College of Medicine, Seoul, South Korea.
Background:
Precise patient-specific quality assurance (PSQA) is critically important in advanced treatments.
Purpose:
This study aimed to evaluate the performance of the RadCalc software's electronic portal imaging device (EPID)-based patient specific QA function, focusing on its ability to accurately reconstruct doses from EPID images acquired on the Elekta Harmony Pro linear accelerators (LINACs).
Methods:
Beam modeling was conducted for 6-MV and 6-MV flattening filter-free (FFF) photon beams on the recently released Harmony Pro LINAC system in RadCalc. Volumetric modulated arc therapy (VMAT) plans for 6-MV (19 patients) and 6-MV FFF (10 patients) were generated using the ArcCHECK phantom to replicate clinical treatments. Dose calculations from RadCalc, using portal images, were compared to those generated by the treatment planning system (TPS) through gamma analysis at 1%/1, 2%/2, and 3%/3 mm, with a 90% minimum passing rate.
Results:
For 6-MV and 6-MV FFF beams, gamma analysis comparing EPID data with RadCalc showed a 98.83% passing rate at 2%/2 mm. For VMAT plans, 6-MV beams achieved the highest passing rate (99.88%) for the breast case at 3%/3 mm and the lowest (82.33%) for the lymphatic node case. Similarly, 6-MV FFF beams yielded the highest passing rate (99.87%) for the abdominal case at 3%/3 mm and the lowest (94.07%) for the bone case.
Conclusions:
RadCalc showed high accuracy for dose evaluation on the Elekta Harmony Pro LINAC. Beam modeling achieved a gamma passing rate of at least 98% (2%/2 mm), demonstrating reliable performance even under minor setup variations.
