Program Information
Comparison of 4D-Measurement-Guided Dose Reconstructions (MGDR) with COMPASS and OCTAVIUS 4D System
R Leung1*, M Wong2 , O Blanck3 , V Lee4 , G Law5 , K Lee6 , S Tung7 , M Chan8 , (1) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R), (2) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R), (3) University Clinic Schleswig-Holstein, Kiel, Kiel, (4) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R), (5) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R), (6) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R), (7) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R), (8) Tuen Mun Hospital, Hong Kong, Hong Kong (S.A.R)
Presentations
SU-E-T-202 (Sunday, July 12, 2015) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose:
To cross-validate the MGDR of COMPASS (IBA dosimetry, GmbH, Germany) and OCTAVIUS 4D system (PTW, Freiburg, Germany).
Methods:
Volumetric-modulated arc plans (5 head-and-neck and 3 prostate) collapsed to 40° gantry on the OCTAVIUS 4D phantom in QA mode on Monaco v5.0 (Elekta, CMS, Maryland Heights, MO) were delivered on a Elekta Agility linac. This study was divided into two parts: (1) error-free measurements by gantry-mounted EvolutionXX 2D array were reconstructed in COMPASS (IBA dosimetry, GmbH, Germany), and by OCTAVIUS 1500 array in Versoft v6.1 (PTW, Freiburg, Germany) to obtain the 3D doses (COM4D and OCTA4D). COM4D and OCTA4D were compared to the raw measurement (OCTA3D) at the same detector plane for which OCTAVIUS 1500 was perpendicular to 0° gantry axis while the plans were delivered at gantry 40°; (2) beam steering errors of energy (Hump=-2%) and symmetry (2T=+2%) were introduced during the delivery of 5 plans to compare the MGDR doses COM4D_Hump (COM4D_2T), OCTA4D_Hump (OCTA4D_2T), with raw doses OCTA3D_Hump (OCTA3D_2T) and with OCTA3D to assess the error reconstruction and detection ability of MGDR tools. All comparisons used Υ-criteria of 2%(local dose)/2mm and 3%/3mm.
Results:
Averaged Υ passing rates were 85% and 96% for COM4D ,and 94% and 99% for OCTA4D at 2%/2mm and 3%/3mm criteria respectively. For error reconstruction, COM4D_Hump (COM4D_2T) showed 81% (93%) at 2%/2mm and 94% (98%) at 3%/3mm, while OCTA4D_Hump (OCTA4D_2T) showed 96% (96%) at 2%/2mm and 99% (99%) at 3%/3mm. For error detection, OCTA3D doses were compared to COM4D_Hump (COM4D_2T) showing Υ passing rates of 93% (93%) at 2%/2mm and 98% (98%), and to OCTA4D_Hump (OCTA4D_2T) showing 94% (99%) at 2%/2mm and 81% (96%) at 3%/3mm, respectively.
Conclusion:
OCTAVIUS MGDR showed better agreement to raw measurements in both error- and error-free comparisons. COMPASS MGDR deviated from the raw measurements possibly owing to beam modeling uncertainty.
Contact Email: