# Automated Test for Monthly Quality Assurance of Optical Surface Imaging Dynamic Localization Accuracy

## TL;DR

This paper presents an automated method for monthly quality assurance testing of surface-guided radiation therapy systems to ensure accurate dynamic localization.

## Contribution

A novel automated procedure for monthly QA testing of SGRT dynamic localization accuracy is developed and validated.

## Key findings

- All translation and rotation deviations remained within AAPM TG-142 SABR tolerance limits.
- Maximum translation deviations from MV imaging, trajectory logs, and Identify logs were -0.94mm, -0.11mm, and -0.78mm, respectively.
- Rotation deviations from trajectory and Identify logs were 0.01 and -0.2 degrees, respectively.

## Abstract

The American Association of Physicists in Medicine (AAPM) recently published the report of Task Group (TG) 302, which provides recommendations on acceptance, commissioning, and ongoing routine quality assurance (QA) for surface-guided radiation therapy (SGRT) systems. One of the recommended monthly QA tests is a dynamic localization accuracy test. This work aimed to develop an automated procedure for monthly SGRT dynamic localization QA.

An anthropomorphic head phantom was rigidly attached to the 6-dof couch of a TrueBeam linac. TrueBeam Developer Mode was used to take an MV image of the phantom at the starting position, then automatically drive the couch through a series of translations and rotations, taking an MV image after each translation. The Identify SGRT system monitored the motion of the phantom surface from the starting position. Translations assessed on MV images were compared to translations reported in trajectory log files and Identify log files. Rotations were compared between trajectory log files and Identify log files.

Three experiments were conducted. None of the translations or rotations from any experiment exceeded the tolerance values for stereotactic ablative body radiation therapy (SABR) recommended by AAPM TG-142. Maximum deviations from the expected translation values from MV imaging, trajectory log files, and Identify log files were -0.94mm, -0.11mm, and -0.78mm, respectively. Maximum deviations from the expected rotation values from trajectory log files and Identify log files were 0.01 and -0.2 degrees, respectively. The proposed method is a simple automated way to complete monthly dynamic localization QA of SGRT systems.