# Classification performances of two diode arrays for patient‐specific quality assurance of stereotactic body radiation therapy treatments based on absolute dose measurements in phantom

## TL;DR

This paper compares two diode arrays for quality assurance in radiation therapy, finding optimal criteria for dose measurement accuracy.

## Contribution

The study identifies device-specific gamma analysis criteria for accurate patient-specific QA in SBRT.

## Key findings

- ArcCHECK AUCs are more significant than SRS MapCHECK AUCs in QA assessments.
- Field-by-field measurements show higher significance than per-plan comparisons.
- SRS MapCHECK requires tighter DTA criteria due to higher spatial resolution.

## Abstract

Despite the rapid growth in the clinical practice, no specific recommendations on pre‐treatment patient‐specific quality assurance of volumetric modulated arc therapy‐based stereotactic body radiation therapy plans have been established.

In this contest, the study aims to identify optimal gamma analysis criteria and thresholds for the Sun Nuclear ArcCHECK and SRS MapCHECK arrays.

Twenty SBRT plans were delivered on both devices per plan and field‐by‐field. The measurements were compared with calculations and Gamma Passing Rates (GPRs), obtained using global normalization in absolute dose and 10% threshold, with six different Dose Difference (DD) / Distance To Agreement (DTA) criteria stricter than those universally suggested by the Report of the AAPM Task Group No. 218, were recorded. Receiver Operating Characteristics analysis was performed on GPRs while varying the threshold from 0% to 100%, the agreement between calculations and absolute dose measurements, obtained with a IBA Razor chamber at the isocenter in phantom at different levels (i.e., 1%, 2%, 3%, 4%, and 5%), being the Ground Truth. Significance of the resulting Areas Under Curve (AUCs) against the random guess was tested.

AUCs obtained with ArcCHECK are generally more significant than with SRS MapCHECK, while those measured field‐by‐field are more significant than per plan. Within the considered DD/DTA criteria, the most discriminative ones are device‐specific, that is, 2%/2 mm or 1%/2 mm for ArcCHECK and 2%/1 mm or 1%/1 mm for SRS MapCHECK.

Our results on ArcCHECK confirm the indication of the AAPM Task Group No. 218, while for SRS MapCHECK, acceptable discriminating capabilities are possible with DTA = 1 mm, suggesting that devices with native higher spatial resolution, preferred in SBRT for the better sampling of the dose distribution, require tighter DTA.