technique is the accuracy required for the probe positioning and the dependence to the beam angle. These effects can lead to uncer- tainties which cannot be differentiated from treatment errors. The aim of this study is to show the interest of transit dosimetry with EPID, to limit these error elements. Material&methods: In vivo portal dosimetry is performed with an Elekta IviewGTÒ EPID mounted on an Elekta PreciseÒ linac. EPIgrayÒ (Dosisoft) software based on the RTMFs method is used to reconstruct the delivered dose to prescription point in the patient. The in vivo reconstructed dose can then be compared to the calculated dose by XioÒ Elekta treatment planning system. The IBA in vivo dosimetry system (EDP15 diodes, DPD3 elec- trometer, and InViDosÒ software) is used to measure the entrance dose during the two first treatment sessions. The portal dosimetry is carried out during the three following sessions. The patients included in this study are treated for H&N and breast cancer. These two methods for in vivo dosimetry will be compared. Results: For the H&N localization, the first results with diode and EPID show differences within the accepted tolerance, except for the larynx localization, where a small positioning error can cause a depth variation on the beam axis and an important error on the reconstructed dose using the portal images. The study on breast is beginning and first results will be available soon. Conclusion: Transit dosimetry can really improve in vivo dosimetry method. Indeed, it requires no additional time for detector posi- tioning, data acquisition and avoid an «operator-depending» detector positioning. Breast irradiation is complex to control with diode in vivo system because of the dependency of diode’s response with beam angle. The portal dosimetry allows to check in the same time the position of the patient and the delivered dose during the treatment and so the interpretation of results is facilitated. Keywords: in-vivo dosimetry, EPID, transit dosimetry 48 The routine use of in vivo portal dosimetry at the Institut Curie P. Boissard, P. Franc ¸ois and A. Mazal Institut Curie / Service de Physique Me´dicale - France Purpose: Transit dosimetry performed with Electronic Portal Imaging Devices (EPIDs) appears as an interesting alternative for in vivo dose verification. This work presents the results of in vivo portal dosimetry in clinical routine conditions during three year at the Institut Curie radiotherapy department. Materials and Methods: In vivo portal dosimetry was performed for 432 patients (304 prostate cancers, 42 breast cancers, 88 others) using aSi EPID mounted on 4 linear accelerators Varian MSä. 90% of the prostate cancers were treated by using Intensity Modulated Radiation Therapy (IMRT). The dose at the point of prescription in the patient is estimated from the transmitted signal acquired with the portal imager using an algorithm developed at the Institut Curie (implemented in EPIGray, Dosisoftä) and compared with calculated data (Eclipse software using pencil beam algorithm, Varian MSä). In vivo dose was checked for each beam during the three first sessions. Results: Ratios between stated and measured doses for the entire treatment plan are reported. They are within the accepted toler- ance of classical in vivo dosimetry (mean¼ 0.994; 1SD¼ 0.023; 95% of the results included in the interval [0.95;1.05]). Conclusions: Transit dosimetry is easy to use systematically in clinical routine (eg: no additional time for data acquisition). The accuracy of this method was more than acceptable for in vivo purpose. The use of the “image of the day” can be used to differ- entiate the origin of specific discrepancies in the results. Keywords: In Vivo dosimetry, transit dose, EPID 49 Comparison of three measurement systems for RapidArc treatment verification J. Molinier, L. Vieillevigne, T. Brun, M. Vidal, L. Broussillou, C. Sarrieu and R. Ferrand Institut Claudius Regaud (ICR) e De´partementd’inge´nierieet de physique me´dicale, Toulouse, France Introduction: Intensity Modulated radiation therapy treatments need accurate quality assurance due to their complexity. Different kinds of systems are used to validate the comparison between calculated and measured doses of a treatment plan delivery. This comparison is performed thanks to dose profile, dose distribution and gamma index. The aim of this study is to evaluate three measurement systems to define their characteristics, sensitivity, complementarities and uncertainties. Materials and Methods: Three independent measurement systems were studied: - ArcCheck (diode cylinder of SunNuclear with SNC 6.1 software) - Ionisation chambers array with Octavius phantom (PTW, Veri- soft 4.0 software) - Portal imager (Varian with EPIQA (EPIDOS) 2.0.10 software) We first evaluated reproducibility, repeatability, dose rate depen- dence, linearity, field size, and different beam incidences for conventional field. Furthermore, 47 gamma results of RapidArc plans (29 head and neck cases, 18 prostate cases) were analysed. Next step will be to introduce errors in treatment plans (position, gantry and MLC speed, couch and collimator position) to evaluate sensitivity of such detectors. Results: Repeatability and reproducibility for all measurement systems were under 1%, which shows they are good dosimeters. The average and standard deviation corresponding to gamma index (3%/3mm) were: - for the 29 ORL plans: 99.36%/0.67% (ArcCheck) vs 97.54%/ 1.71% (matrice PTW) vs 97.61%/1.19% (EPID) - for the 18 prostate plans: 99.39%/0.52% (ArcCheck) vs 98.65%/ 1.45% (matrice PTW) vs 99.53%/0.57% (EPID). Results regarding the introduction of error are in progress. Conclusion: The three measurements systems give similar gamma results, wherever the complexity of the localisation. Intrinsic characteristics of detectors and results due to the intro- duction of errors should help us to better understand their limits and thus to optimize patient quality assurance. Keywords: RapidArc, quality assurance, gamma index 50 In-vivo dosimetry: The patient can be used as a detector J.-M. Denis, D. Prieels, G. Janssens, J. Orban de Xivry, E. Sterpin and S. Vynckier UCL e Universite´Catholique de Louvain, MIRO - Poˆle d’imagerie mole´culaire, radiothe´rapie et oncologie, 1200 Bruxelles, Belgium Introduction: The purpose of this presentation is to introduce and discuss a novel method for in-vivo dosimetry. This method is based on the electrical current induced in the body of the patient when this one is subject to high energy photon irradiation in external beam radiation therapy. Material and methods: Typical X-rays in radiotherapy (6MV, 18MV, 25MV) generate secondary electrons in the patient, mainly due to the Compton interaction. Dose deposition is mainly due to these secondary electrons. Some of them escape the body, mainly in the region where the photon beam comes out of the patient. As the body is conductive, if it is electrically isolated from the ground, it is SFPM Annual Meeting 2012 S13