Evaluation of Varian Identify SGRT System for Localization of Cranial Targets
V Stakhursky1*, W Chang2, M Sewell3, I Stakhurska4, S Kriminski5, (1) USONCOLOGY CMOH, Philadelphia, PA, (2) U Oncology, Philadelphia, PA,(3) US Oncology, Snellville, GA, (4) Karlsruher Institut Fur Technologie Studienkolleg, Karlsruher, DE, (5) Health Quest, Wappingers Falls, NY
Presentations
PO-GePV-P-27 (Sunday, 7/25/2021) [Eastern Time (GMT-4)]
Purpose: Identify (Varian, Palo Alto, CA) surface-guided localization system is a new imaging option offered to monitor inter-fractional and intra-fractional motion of patients. This study evaluates accuracy of target localization for non-coplanar beam treatments using Identify.
Methods: An anthropomorphic head phantom was scanned using GE Discovery RT, 0.625mm slice thickness, and a multi-arc non-coplanar plan was created in Eclipse TPS (Varian). The DICOM structures of head contour and plan isocenter were sent to Identify server for surface image processing . The head phantom was placed on 6D TrueBeam (Varian) treatment couch in 10 different positions varying 3 translational (<1.5cm) and 3 rotational (<3deg) degrees of freedom, emulating inter -fractional variations. For each phantom placement CBCT imaging was performed using 1mm slice thickness. Imaging followed by a 6-DOF 3D automatic registration to reference image. Registration accuracy was manually reviewed by checking positions of three fiducials.The calculated shift parameters were compared to those generated by Identify system based on comparison of optically rendered head surface and radiographic external contour. After applying CBCT shifts the couch angle was shifted to +-90deg position and Identify registration parameter were recorded.
Results: For 0-degree couch position the maximum magnitude of translational offset recorded was 0.8mm(avg 0.7mm), with largest component in vertical direction(0.5mm). Rotational errors were 0.4deg(yaw), 0.1deg(roll) and 0.2deg(pitch). When Identify was generating shifts based on comparison of optical surface to optical surface captured during the first treatment session, translational match improved to 0.4mm magnitude, 0.2deg(yaw), 0.1deg(roll) and 0.1deg(pitch). When Identify tracked positional shifts during couch angle changes (+-90, +-45deg), the translational offset magnitude was 0.6mm and rotational errors were under 0.6deg.
Conclusion: Identify surface guided positioning system is capable of localizing cranial targets with a better than 1 mm precision for non-coplanar treatments.
ePosters
Keywords
Validation, Image Guidance, Radiation Therapy
Taxonomy
TH- RT Interfraction Motion Management: General (most aspects)
Contact Email