28 Contribution of transit dosimetry (EPID-based) and retrospective dose calculation in the patient's real anatomy (kV-CBCT-based) for the estimation of the dose delivered to the patient (Preliminary study). (December 2018)

• Record Type: Journal Article
• Authors: Sauvinet, R.; Chevillard, C.; Husson, F.; Pierrat, N.; Dumas, J-L.
• Is Part Of: Physica medica. Volume 56(2018)Supplement 1
• Page Start: 52
• Page End: 53
• Publication Date: 2018-12
• Subjects: Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2018.09.110
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Introduction: During an external photon radiotherapy treatment, transit dosimetry performed with the electronic portal imaging device (EPID) makes it possible to record a footprint of the irradiation beam passing through the patient to detect any deviations between predicted and delivered dose. These deviations are most often due to wrong patient positioning or anatomical changes. The on-board X-ray imaging system provides 3D patient imaging (kV-CBCT). Despite lower image quality than the planning CT (pCT), the kV-CBCT provides information on the positioning and anatomy of the patient at the time of the fraction. It is tempting to use this data for recalculation purposes in the planning system (TPS). The objective of this study is to evaluate the potential of the joint contribution of transit dosimetry and a posteriori dose calculation performed in the anatomy of the kV-CBCT of the day. Methods: The study concerns breast treatment plans in RapidArc technique (RX 6MV) on a TrueBeamTM machine (Varian MS, Palo Alto, CA, USA). For each fraction, a kV-CBCT was performed before beam delivered and the EPID images acquired in transit condition (movie mode). A cumulative image over the length of the arc is reconstructed in terms of dose from the movie sequence, it is then compared with a predicted image calculated at the EPID level, from the RT-plan, in the same dose terms and transmission conditions. Image comparison is performed using γ -index maps (5%/4 mm; threshold 40%; overall) and GAI approval value (%). The 3D a posteriori dose calculation was performed using Eclipse TPS (Varian MS) with the PK-AAA algorithm on the pCT corrected from anatomical information of the kV-CBCT by an elastic registration (ISOgray-DOSIsoft). Results: In this study the comparison of a predicted/acquired dose images with the gamma index criteria mainly leads to GAI approval values greater than 90%. The use of the kV-CBCT for a posteriori dose calculation, confirms the deviations detected with the EPID images comparison and allows a more precise quantification of the deviations in terms of dose, or irradiated volumes. The observed differences in dose are more or less marked depending on the patient, with a trend for larger differences on the target volumes (around 5–10%) than for organs at risk (around 2–5%). Conclusions: This preliminary study of the complementarity between EPID transit dosimetry and a posteriori dose calculation in the kV-CBCT anatomy of the day, shows the relevance and consistency of the two approaches. Deviations detected from planning conditions in terms of dose delivered and volumes to be treated or spared can then be more easily interpreted and their causes better identified in clinical practice.