Click here to


Are you sure ?

Yes, do it No, cancel

Establishing the Sensitivity of Halcyon's Automatic Transit Image Acquisition for Treatment Error Detection

X Fave*, KL Moore, UC San Diego Moores Cancer Center, La Jolla, CA


(Sunday, 7/29/2018) 2:05 PM - 3:00 PM

Room: Karl Dean Ballroom A2

Purpose: The Varian Halcyon™ portal imager automatically acquires exit dosimetry images for every patient with full field coverage. This could facilitate “every patient, every monitor unit� QA and adaptive radiotherapy. This study evaluated the system’s sensitivity to potential clinical errors (beam output, patient weight, and misalignment changes) and examined day-to-day variations from clinical exit images.

Methods: Open 10x10 and modulated IMRT fields were delivered for each test. To evaluate output changes, monitor units were scaled by 2-10% and delivered to a solid water stack and CIRS anthropomorphic phantom. To mimic weight changes, 0.5-5.0cm of buildup were added to the solid water. To evaluate misalignment, the oval-shaped CIRS phantom was shifted laterally 2-10cm. For each test, the mean relative differences (MRD) and standard deviation of pixel-difference histograms (SD) between resulting and baseline images were calculated. To assess actual patient variations, MRD and SD were calculated from a clinical prostate patient’s exit images.

Results: The MRD responded linearly to output and buildup changes with shot-to-shot MRD variations of 0.4%, implying 1% output and 0.2cm changes in buildup could be detected with 2.5σ confidence if these were the only source of error. Shifting the oval-shaped CIRS phantom laterally resulted in detectable MRD changes, but SD increased at a higher rate suggesting it would be more sensitive to positioning offsets than MRD. Day-to-day variations in the MRD for a prostate patient ranged from -1.6% to 2.4%, meaning these variations are much larger than the sensitivity in any one type of error.

Conclusion: Rapid analyses of automatically-acquired Halcyon exit images could detect mid-treatment changes with high sensitivity, though appropriate thresholds for typical error types would necessitate more detailed pixel difference image analysis and/or multiple field observations. This study presents the first steps towards developing effortless image evaluation for all aspects of every patient’s treatment.

Funding Support, Disclosures, and Conflict of Interest: Conflicts of interest: Varian research grant, travel, and honoraria.


Electronic Portal Imaging, Transit Dosimetry, Quality Assurance


TH- External beam- photons: portal dosimetry

Contact Email