[Poster] Interprétation de la Dosimétrie In Vivo de Transit Lors de Modifications Anatomiques du Patient par une Représentation Fantôme Equivalent Homogène (EHP)*

*Interpretation of transit in vivo dosimetry using equivalent homogeneous phantom (EHP) for anatomic changes

Chevillard C^1,2, Michel M¹, Dumas J-L¹, Husson F²
1 Radiotherapy Department, Institut Curie, Paris, France
2 R&D Medical Physics, DOSIsoft, Cachan

Presented at SFPM 2016

ABSTRACT
Introduction: In transit in vivo dosimetry condition, the signal received by the electronic portal imaging device (EPID) depends essentially on the thickness (water equivalent) of the patient along each ray. Kilovoltage Cone Beam Computed Tomography (kV-CBCT) reports every anatomical changes and can be used to represent the patient as an equivalent homogeneous phantom (EHP) [1]. The aim of our study was to use the concept of EHP to highlight any anatomical modification between planning Computed Tomography (pCT) of the patient and subsequent CBCTs of the “patient of the day”, to ease the analysis of the transit in vivo dosimetry.

Material and Methods: For validation of the method, we used a heterogeneous phantom [CIRS IMRT thorax phantom (CIRS, Inc., Norfolk, Virginia, USA)] with different bolus thickness to simulate anatomical changes. We applied the method for patients according to different tumor sites (Head and Neck, lung and prostate cancer). Phantom and patients underwent with a pCT, to build the treatment plan and weekly CBCT. The CBCT for each patient was acquired with the same protocol in order to guarantee  constant Hounsfield units (HU) to electron density (d_e) calibration curve to reconstruct the successive EHP. The calibration curves were realized using the phantom « Cheese » (Gammex RMI, Middelton, WI) for the Toshiba scanner (Aquilion^TM LB, Toshiba Medical Systems) and for the Varian Truebeam^TM with on-board imaging (OBI). We calculated Antero-posterior and right/left lateral EHP based on the external contour.

Results: We validated the method to build an EHP with a heterogeneous phantom thanks to the calibration curves. We observed a significant relationship between the in vivo dose values and the shape of the EHPs.

Conclusion: The EHP representation matched with EPID-based in vivo dosimetry, summarizes anatomical patient information for contributing to the analysis of the dosimetric results.