Initial End-to-End testing of the ExacTrac Dynamic Deep Inspiration Breath Hold workflow using a breath hold breast phantom

DOI: https://doi.org/10.21203/rs.3.rs-2519456/v1

Abstract

The ExacTrac Dynamic (ETD) system provides a Deep Inspiration Breath Hold (DIBH) workflow for treatment of left sided breast patients. Combined stereoscopic x-ray imaging with optical and thermal mapping allows localisation against simulation imaging, alongside surface guided breath hold monitoring. This work aimed to determine appropriate x-ray breast imaging parameters, optimal Hounsfield Unit (HU) values for patient contour generation and workflow evaluation via end-to-end (E2E) positioning testing. After phantom localisation via existing Image Guidance (IG), stereoscopic imaging was performed with a range of parameters to determine minimal residual error. Similarly, residual errors in prepositioning were minimised using a range of HU value external contours. E2E positioning was completed for clinical workflows and residual positioning error was measured using treatment field deviations and existing IG. Parameters of 60 kV and 25mAs were determined appropriate for patient imaging and HU values between -600 HU and 200 HU enabled adequate prepositioning. The average residual position error measured using treatment fields was 1.0 ± 0.9 mm, 0.4 ± 1.0 mm and 0.1 ± 0.5 mm in the lateral, longitudinal and vertical directions, respectively. Errors measured using existing IG were -0.6 ± 1.1 mm, 0.5 ± 0.7 mm and 0.2 ± 0.4 mm in the lateral, longitudinal and vertical directions, and rotational errors averaged of 0.0 ± 1.0 o, 0.5 ± 1.7 o and -0.8 ± 1.8o for pitch roll and yaw, respectively. The use of bone weighted matching increased residual error. A reduction of phantom DIBH volume maintained target positioning accuracy.