C Jeong

Proton radiography and computed tomography with dynamic water range shifter

We introduce a novel dynamic water range shifter (DWRS) that enables range shifting of a mono-energetic proton beam for the purpose of obtaining a water equivalent path length (WEPL). A scintillation screen and a charge-coupled device (CCD) camera were assembled with a mirror for a detecting system. We acquired two sets of proton radiographs with and without the scanned object, and calculated 2D distribution of the WEPL. We first optimized measurement conditions, i.e., the sampling interval and the nearest sampling point from the surface. Then we evaluated the performance of the proton radiography system by use of the step phantoms and a patient-specific range compensator. The step phantom measurement was designed to evaluate WEPL accuracy and standard deviation, and the patient-specific compensator was utilized to evaluate image quality dependency on the shifting motions of the DWRS. Based on the optimization and evaluation in the radiographic setting, a cylindrical phantom having multiple insert holes was used to obtain proton CT and the image contrasts of the holes were evaluated. This study successfully demonstrates the feasibility of using the DWRS for proton radiography and CT.

SU-F-T-460: Dosimetric Matching Between Trilogy Tx and TrueBeam STx

PURPOSE To compare the commissioned beam data for one flattening filter photon mode (6 MV) and two flattening filter-free (FFF) photon modes (6 and 10 MV-FFF) between Trilogy Tx and TrueBeam STx and evaluate the possibility of dosimetric matching METHODS: Dosimetric characteristics of the new Trilogy Tx including percent depth doses (PDDs), profiles, and output factors were measured for commissioning. Linear diode array detector and ion chambers were used to measure dosimetric data. The depth of dose maximum (dmax) and PDD at 10 cm (PDD10) were evaluated: 3×3 cm2 , 10×10 cm2 , and 40×40 cm2 . The beam profiles were compared and then penumbras were evaluated. As a further test of the dosimetric matching, the same VMAT plans were delivered, measured with film, and compared with TPS calculation. RESULTS All the measured PDDs matched well across the two units. PDD10 showed less than 0.5% variation and dmax were within 1.5 mm at the field sizes evaluated. Within the central 80% of transverse axis, profile data were almost identical. TrueBeam data resulted in a slightly greater penumbra width (up to 1.9 mm). The greatest differences of output factors were found at 40 × 40 cm2 : 2.40%, 2.03%, and 2.22% for 6 MV, 6 MV-FFF, and 10 MV-FFF, respectively. For smaller field sizes, less than 1% differences were observed. The film measurements demonstrated over 97.3% pixels passing-gamma analysis (2%/2mm). The results showed excellent agreement between measurements of two machines. CONCLUSION The differences between Trilogy Tx and TrueBeam STx found could possibly affect small field and also very large field sizes in dosimetric matching considerations. These differences encountered are mostly related with the changes in the head design of the TrueBeam. Although it cannot guarantee full interchangeability of two machines, dosimetric matching by field size of 25 × 25 cm2 might be clinically acceptable.

SU-F-T-563: Delivered Dose Reconstruction of Moving Targets for Gated Volumetric Modulated Arc Therapy (VMAT)

PURPOSE Actual delivered dose of moving tumors treated with gated volumetric arc therapy (VMAT) may significantly differ from the planned dose assuming static target. In this study, we developed a method which reconstructs actual delivered dose distribution of moving target by taking into account both tumor motion and dynamic beam delivery of gated VMAT, and applied to abdominal tumors. METHODS Fifteen dual-arc VMAT plans (Eclipse, Varian Medical Systems) for 5 lung, 5 pancreatic, and 5 liver cancer patients treated with gated VMAT stereotactic body radiotherapy (SBRT) were studied. For reconstruction of the delivered dose distribution, we divided each original arc beam into control-point-wise sub-beams, and applied beam isocenter shifting to each sub-beam to reflect the tumor motion. The tumor positions as a function of beam delivery were estimated by synchronizing the beam delivery with the respiratory signal which acquired during treatment. For this purpose, an in-house program (MATLAB, Mathworks) was developed to convert the original DICOM plan data into motion-involved treatment plan. The motion-involved DICOM plan was imported into Eclipse for dose calculation. The reconstructed delivered dose was compared to the plan dose using the dose coverage of gross tumor volume (GTV) and dose distribution of organs at risk (OAR). RESULTS The mean GTV dose coverage difference between the reconstructed delivered dose and the plan dose was 0.2 % in lung and pancreas cases, and no difference in liver cases. Mean D1000cc of ipsilateral lungs was reduced (0.8 ± 1.4cGy). CONCLUSION We successfully developed a method of delivered dose reconstruction taking into account both respiratory tumor motion and dynamic beam delivery, and applied it to abdominal tumors treated with gated VAMT. No significant deterioration of delivered dose distribution indicates that interplay effect would be minimal even in the case of gated SBRT. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2015038710).

SU-E-T-178: Clinical Feasibility of Multi-Leaf Collimator Based Dynamic Wedge

Purpose: A multi-leaf collimator (MLC) based dynamic wedge (MDW), which provide similar dose profile of physical wedge (PW) along x-jaw direction while significant monitor unit (MU) reduction, was developed and investigated for clinical use. Methods: A novel technique was used to create the wedge profile using MLC. A modification was applied to the DICOM-RT format file of the plan made with the PW to replace PW with MDW. The Varian enhanced dynamic wedge profile was used to produce MLC sequence, while the MU of the wedged field was recalculated using PW factor and fluence map. The profiles for all possible MDWs to substitute PWs were verified in 6/15 MV x-ray irradiations. New plans with MDWs were compared with the original plans in 5 rectal, 5 RT breast and 5 liver cases. Results: The wedge profile of the MDW fields were well matched with those of PWs inside the fields while less scatter than PW out of the fields. For plan comparisons of the clinical cases no significant dose discrepancy was observed between MDW plan and PW’s with the dose volume histograms. The maximum and mean doses in PTVs are agreed within 1.0%. The Result of OARs of MDW plans are slightly improved in the maximum doses (3.22 ∼ 150.4 cGy) and the mean doses (17.18 ∼ 85.52 cGy) on average for all cases while the prescribed doses are 45 Gy for rectal cases, 40 or 45 Gy for liver cases and 50 Gy for breast cases. The MUs of the fields which replace PW with MDW are reduced to 68% of those of PW. Conclusion: We developed a novel dynamic wedge technique with MLC that shows clinical advantage compared to PW.

SU-E-T-484: A New Method of Aligning Patient Setup Lasers in Radiation Therapy

PURPOSE To develop a new method to align the patient setup lasers in radiation therapy and examine its validity and effectiveness. METHODS The new laser alignment method was realized by a device composed of both a metallic base plate and few acrylic transparent plates with a cross hair line on each of them. The holders of radiochromic films were prepared in the device to find a radiation isocenter. The right laser positions could be found optically by matching the shadows of all the cross hairs in the gantry head and the device. The repeatability and reproducibility (R&R) of laser alignments and the dependency of the alignment on the position error of the light source were evaluated by comparing the standard deviations and the means of the measured laser positions. After aligning the lasers optically, a radiation isocenter was found by a collimator spoke shot and the gantry spoke shot, and then the lasers were parallely translated to the isocenter. RESULTS In the R&R test, the standard deviation was 1.14 mm for the new method whereas it was 1.49 mm or 2.76 mm for the conventional method with either high- or low-precision levels. In the test of the dependency on the position error of the light source, the mean laser position was shifted by 5.3 mm corresponding to the shift of the light source, 4.8 mm for the new method, but for the conventional method the laser position was shifted more than 7 times than that. The positional shift could be corrected by a parallel translation to the isocenter in the new method. CONCLUSION A new laser alignment method was devised for radiation therapy and tested successfully. The method enabled us to align the lasers easily and accurately without repetition, and all lasers could be finally aligned to the radiation isocenter.

SU-E-T-241: Monte Carlo Simulation Study About the Prediction of Proton-Induced DNA Strand Breakage On the Double Helix Structure

PURPOSE In particle therapy and radiobiology, the investigation of mechanisms leading to the death of target cancer cells induced by ionising radiation is an active field of research. Recently, several studies based on Monte Carlo simulation codes have been initiated in order to simulate physical interactions of ionising particles at cellular scale and in DNA. Geant4-DNA is the one of them; it is an extension of the general purpose Geant4 Monte Carlo simulation toolkit for the simulation of physical interactions at sub-micrometre scale. In this study, we present Geant4-DNA Monte Carlo simulations for the prediction of DNA strand breakage using a geometrical modelling of DNA structure. METHODS For the simulation of DNA strand breakage, we developed a specific DNA geometrical structure. This structure consists of DNA components, such as the deoxynucleotide pairs, the DNA double helix, the nucleosomes and the chromatin fibre. Each component is made of water because the cross sections models currently available in Geant4-DNA for protons apply to liquid water only. Also, at the macroscopic-scale, protons were generated with various energies available for proton therapy at the National Cancer Center, obtained using validated proton beam simulations developed in previous studies. These multi-scale simulations were combined for the validation of Geant4-DNA in radiobiology. RESULTS In the double helix structure, the deposited energy in a strand allowed to determine direct DNA damage from physical interaction. In other words, the amount of dose and frequency of damage in microscopic geometries was related to direct radiobiological effect. CONCLUSION In this report, we calculated the frequency of DNA strand breakage using Geant4- DNA physics processes for liquid water. This study is now on-going in order to develop geometries which use realistic DNA material, instead of liquid water. This will be tested as soon as cross sections for DNA material become available in Geant4-DNA.

SU‐E‐T‐112: Dose Distribution Verification of Proton Beam Using Light Output On Scintillation Plate

PURPOSE To verify dose distribution of proton beam using CCD camera scintillation screen system. METHODS CCD camera scintillation screen system was developed to verify dose distribution at National Cancer Center in Korea. In this study, we used a high sensitivity scintillation plate to measure visible light acquired by CCD camera. The light output on scintillation plate was measured while changing the irradiation time and beam current in proton beam. Then, the results were analyzed to obtain the depth dose distribution and dose profile. The measured dose distributions in double scattering delivery mode were compared with those in ionization chamber. RESULTS The relationship between the light output and beam current showed good linearity. Also, the light output increased linearly with the increase of irradiation time. For the proton beam for double scattering mode, light output response on scintillation plate was compared with the dose distribution from ionization chamber and showed the good agreement. CONCLUSION We evaluated the dosimetry characteristic of proton beam using CCD camera scintillation screen system. Also, we verified the dos - e distribution of proton beam for double scattering delivery mode. This work was supported by a research grant from the Ministry of Education, Science and Technology in Korea (No. 2010-0026071) and a fund from research project of the Korea National Cancer Center (No. NCC-0910110).

SU‐E‐I‐98: Novel Method for Proton Radiography Using Plastic Scintillation Plate and Beam Energy Modulation Water Phantom

PURPOSE The purpose of this study is to present the experimental evaluation and quantification of proton radiography using plastic scintillation plate and beam energy modulation (BEM) water phantom. METHODS Using the newly designed water phantom, proton beam energy was modulated by controlling water depth horizontally. Modulated proton beam which passed through water phantom was measured by dose-measurement system which consists of a plastic scintillation plate, a mirror and a CCD camera in a dark box. A range compensator (RC) was positioned between BEM water phantom and scintillation plate for proton radiography. While a proton beam which has 15 cm range in water was being irradiated on RC, water depth is controlled from 71 to 170 mm in 1 mm increment and radiographs of RC on a scintillation plate were saved in every 1 mm increment using a dose-measurement system. One hundred images were stacked and the position of Bragg peak was found along the stack order. The same radiograph procedure was performed without RC in order to calculate the depth of RC by subtracting the position index of Bragg peak between radiograph of RC and background. This difference was divided by water equivalent thickness value of RC material and then measured depth of RC was compared to plan data, quantifying the precision with the mean absolute depth difference (MADD) and the standard deviation (SD). RESULTS The MADD and the SD over the entire areas of RC was calculated as 0.62 mm and 0.67 mm respectively. MADD and SD precisions over chosen flat regions were less than 0.54 mm and 0.25 mm. CONCLUSION Good image quality with high precision of depth measurement was observed by our proton radiography system. These system can also be used as a 3D dosimetry tool. This study was supported by funds from research projects of the National Cancer Center of Korea (nos. NCC 1210210) and National Research Foundation of Korea (K1A3A1A21 2010 0026071).