Abstract Book

S941

ESTRO 37

Table 1: Measured Mean Leaf Opening Time Percentage Error and Standard Deviation with Different Calibration Files

as not immediately recoverable, the number of in- tolerance tests rose. Four lung and two breast cancer patients presented OTL γ% and γ mean values at the end of treatment course due to both class-1 and class-2 errors. Even though adequate corrections were adopted (for class-1 errors), these tests remained out of tolerance due to limited number of tests following the corrections. Random class-2 errors were not easily recoverable requiring clinic investigation about the morphological changes. In some cases, new CT scan was required for dose recalculation. As small dose variations (for both target coverage and rectal filling) were found, an adaptive plan was not taken into account. As regards the workload, 10 patients per day were subjected to IVD tests at our department. Considering that EPID images processing and loading requires about 5 minutes and the time necessary for the visual inspection of OTL tests was about 4 minutes for patient, and a maximum of 4 patients with OTL tests, the daily workload was observed to be about 20 minutes per day. An extra time could be necessary when the cause of error was not immediately recognized.

Conclusion The overall performance of the exit detector is stable to a high extent. An average of 13 operational hours per day for irradiation did not cause any significant deviation for the LOT, despite using a different calibration file from morning and evening respectively. In conclusion, doing the exit detector calibration once per day is sufficient for dosimetric verification purposes. EP-1755 In-Vivo Dosimetry: A Feasibility Study in Routine Clinical Practice M.D. Falco 1 , S. Giancaterino 1 , A. De Nicola 1 , N. Adorante 1 , R. Gimenez De Lorenzo 1 , M. Di Tommaso 1 , A. Vinciguerra 1 , M. Trignani 1 , A. Allajbej 1 , F. Greco 2 , M. Grusio 2 , A. Piermattei 2 , D. Genovesi 1 1 Ospedale Clinicizzato S.S. Annunziata, of Radiation Oncology “G. D’Annunzio”- University of Chieti, Chieti, Italy 2 Fondazione Policlinico Universitario A. Gemelli- Università Cattolica del Sacro Cuore, Unità Operativa di Fisica Sanitaria, Roma, Italy Purpose or Objective The aim of the In-Vivo dosimetry (IVD) during the fractionated radiation therapy, is the verification of the correct dose delivery to patient. Currently, IVD procedures for photon beams are based on the use of the electronic portal imaging device (EPID) and dedicated software to elaborate EPID images. Material and Methods 8474 IVD tests were carried out -for 386 cancer patients treated with 3DCRT, IMRT and VMAT techniques, using the SOFTDISO software. SOFTDISO uses EPID-images in order to: (i) calculate the R index which is the ratio between daily reconstructed dose and the planned one at isocenter; and (ii) perform a γ-like analysis between the signals, (S), of a reference EPID image and that obtained in a daily fraction to supply two indexes, the percentage of points with γ <1 (γ% ) and the mean γ value, (γ mean ). In γ-like analysis, the pass criteria for the signals agreement ΔS% and distance to agreement Δd, have been selected based on the clinical experience and technology used. The adopted tolerance levels for the three indexes were fixed in: 0.95≤R≤1.05, γ% ≥ 90% and γ mean ≤ 0.5. Results The results of R ratio and γ-like analysis permitted to individuate two classes of errors: (i) class-1, that included errors due to inadequate quality controls and (ii) class-2, due to patient morphological changes. We found that the percentage of tests presenting out-of-levels (OTL) values depended on the technique and anatomical site. Table 1 shows IVD results for 3DCRT technique: in the case of abdomen, 47 out of 262 (18%) tests were in OTL; 2 out of 12 patients presented OTL mean index values (γ% and mean ). If class-2 errors were not included,

Conclusion The IVD procedure with SOFTDISO resulted easily implementable and able to individuate errors with a limited workload. EP-1756 Dose verification of Cyberknife MLC FSPB calculation algorithm in different treatment regions C.W. Kong 1 , Y. Ding 1 , C.W. Cheung 1 , H. Geng 1 , W.W. Lam 1 , T.L. Chiu 1 , K.Y. Cheung 1 , S.K. Yu 1 1 Hong Kong Sanatorium & Hospital, Medical Physics and Research Department, Happy Valley, Hong Kong SAR China Purpose or Objective A new multileaf collimator (MLC) Incise2™ of Cyberknife® M6 version offers a potential reduction in treatment time and total MUs compared to IRIS™. However the initial version of the MLC treatment planning is based on a Finite Size Pencil Beam algorithm (FSPB). The objective of this study is to measure the accuracy of the Finite Size Pencil Beam algorithm on MLC based treatment plan in different treatment regions and compare it with Iris based treatment plan calculated on Monte-Carlo algorithm (MC). Material and Methods An anthropomorphic head phantom and Xsight Lung Treatment (XLT) Phantom were used in the study. A ball cube and mini ball cube in the head phantom can simulate the target in the skull and C-Spine respectively. In the Xsight Lung Treatment Phantom, there is a lung ball cube with 2.5 cm diameter tumor simulating target located at the center. MLC and Iris based treatment plans were created on these two phantoms using 6D skull track, Xsight spine and Xsight Lung respectively. In the final dose calculation of the treatment plans the Iris one was