ESTRO 38 Abstract book

S1120 ESTRO 38

500mAs, 1.7±0.9 [0.5;3.3] for 250mAs and 1.3±2.2 [- 1.5;5.1] for 150mAs. We noted the maximal difference about 10.1HU with the iDose 4 level 3 and 100mAs in one of the two air insert. We obtain the same results also for the CIRS phantom: maximal difference about 6.5 HU and for iDose 4 level 3 -1.0±1.9 [-4.8;1.2], 0.4±2.1 [-2.4;3.8] and 1.2±3.6 [-3.9;6.5] HU mean difference respectively with 500, 250 and 150mAs. About the resolution and the homogeneity in the Catphan no difference is visible if we apply more than 200 mAs. With high iDose 4 level and less than 200mAs we lost 2 low constants inserts, with level 6 we lost one more. The qualitative evaluation on the AP CT didn't show any relevant difference and the same result we had with the test plans calculation. We calculated also the CT dose sparing and for Head region we had -30% respect our standard protocol with the level 2, for Thorax -22% with level 3 and for Pelvis -29% with iDose 4 level 3 Conclusion With the iDA we have the possibility to reduce the dose for the planning CT keeping a good resolution and without significant difference in the HU numbers. For the region in which we don't need a very high resolution, for example for simple breast or prostate Patients, suggest to use iDose 4 level 3 and for the other level 1 or 2 are recommended EP-2041 A comparison between 120kv and virtual mixed images in dual energy CT for RT simulation E. Fernandez-Velilla Cepria 1 , J. Quera Jordana 1 , O. Pera Cegarra 1 , M. Prieto Carballo 1 , N. Anton Comelles 1 , M. Algara Lopez 1 1 Hospital del Mar, Radiotherapy Oncology, Barcelona, Spain Purpose or Objective Dual Energy CT (DECT) is being introduced in RT simulation due to its special features. DECT combines low kV (typically 80kV) and high kV (140kV) in order to enhance iodine contrast detection. So far radiotherapists needed two acquisitions in the simulation process: one with contrast for delineation and one without it for calculation. DECT allows the creation of Virtual Non Contrast Images (VNC) at 120kV (standard kV used for simulation), eliminating the need of the non contrast acquisition. The accuracy of this approach depends on the ability of the algorithm to calculate virtual 120kV images from the real 80 and 140kV acquisitions. The so called 120Mixed images are obtained with a weighted formula which uses HU at both kV and a single weighting parameter. A first step in the verification of this formula is aplying it to non contrast images to prove whether it works properly. The objective of this study is checking the reliability of that calculation in real patients without injected iodine. Material and Methods CT Simulation images were acquired by means of a Siemens Confidence RT Plus CT. For each patient two series were imaged: the standard 120kV (used for simulation) and the DECT (composed of 2 consecutive series at 80 and 140kV). For the 120kV series, the acquisition protocol was the one used in clinical routine. For the DECT acquisition, parameters were modified in order to keep the same CTDI as that of the 120kV. Slice thickness was 2mm in all cases. No iodine contrast was used, in order to avoid changes in Hounsfield Units of soft tissues. For each patient, 6 tissues of interest (TOI) were delineated: lung, fat, blood, muscle, liver and cortical bone. Mean HU and standard deviations were calculated for each TOI in the 120kV, 120 Mixed, 80 kV and 140kV series. Comparisons were done in the Contouring module of a Varian Eclipse 13.6 TPS. Results Mean differences in HU between 120kV and 120Mixed series were less than 3 HU in all tissues except bone (see table). 120kV were the series with less noise, followed by

Conclusion DD allows the acquisition of CT images at different kV settings with only one calibration curve in the planning system, with a minimum impact in dose calculations. However it needs matching with the original FBP images for volume delineation due to the lower image quality. Consequently Direct Density™ is accurate enough to be introduced in clinical routine. EP-2040 IDose4 algorithm for radiotherapy planning process: how reduce the dose without image quality loss A. Clivio 1 , E. Barletta 2 , C. Bonacini 1 , R. Graeter 1 1 zrr Zentrum für Radiotherapie Rüti, Radiotherapy, Rüti ZH, Switzerland ; 2 zrr Zentrum für Radiotherapie Rüti/ KSW Kantonsspital Winterthur, Radiotherapy, Rüti ZH/ Winterthur, Switzerland Purpose or Objective With the software upgrade on the Big Bore CT (Philips), we have the opportunity to use the iDose 4 Algorithm (iDA) to reduce the patient dose during the CT acquisition. The aim of our study is to verify that, applying this Algorithm, the CTs are compatible for the radiotherapy planning process in terms of image quality and physical aspects Material and Methods We used three different phantoms. The Catphan CTP504 and CIRS 062M phantom to measure the parameters linked to the image quality analysis (as Hounsfield Unit (HU), Homogeneity, and Resolution). The anthropomorphic Alderson RANDO phantom (AP) to evaluate the image quality on a ‘real’ patient. Using the iDA we have to select one of the seven available strength levels.With the level one we have to reduce about 20% the nominal mAs and for each higher level the reduction is 10% more. We took into account 4 different iDose 4 level (1 to 3 and 6) and we tested those applying nominal mAs value from 100 to 500, with a step of 50 mAs for the first 2 phantom. For the AP phantom, we acquired CT in three regions (Head, Thorax, and Pelvis) with our standard internal protocol and with the above four iDose 4 levels. We evaluate qualitatively the resolution and the image quality and we calculate some standard plans on the AP scan to compare the dose distributions Results For the HU analysis, we found very small differences in the inserts of the CatPhan Phantom between the values obtained with the standard scan and with the iDA. For the scans with 150 mAs or more, the maximal difference is 5.1 HU (for Teflon). When we applied iDose 4 level 3, we had as mean difference in terms of HU over all the inserts (in the square bracket the range), 1.4±0.8 [0.3;2.3] with