Shunsuke Moriya

SU-F-J-57: Effectiveness of Daily CT-Based Three-Dimensional Image Guided and Adaptive Proton Therapy

PURPOSE Daily CT-based three-dimensional image-guided and adaptive (CTIGRT-ART) proton therapy system was designed and developed. We also evaluated the effectiveness of the CTIGRT-ART. METHODS Retrospective analysis was performed in three lung cancer patients: Proton treatment planning was performed using CT image datasets acquired by Toshiba Aquilion ONE. Planning target volume and surrounding organs were contoured by a well-trained radiation oncologist. Dose distribution was optimized using 180-deg. and 270-deg. two fields in passive scattering proton therapy. Well commissioned Simplified Monte Carlo algorithm was used as dose calculation engine. Daily consecutive CT image datasets was acquired by an in-room CT (Toshiba Aquilion LB). In our in-house program, two image registrations for bone and tumor were performed to shift the isocenter using treatment CT image dataset. Subsequently, dose recalculation was performed after the shift of the isocenter. When the dose distribution after the tumor registration exhibits change of dosimetric parameter of CTV D90% compared to the initial plan, an additional process of was performed that the range shifter thickness was optimized. Dose distribution with CTV D90% for the bone registration, the tumor registration only and adaptive plan with the tumor registration was compared to the initial plan. RESULTS In the bone registration, tumor dose coverage was decreased by 16% on average (Maximum: 56%). The tumor registration shows better coverage than the bone registration, however the coverage was also decreased by 9% (Maximum: 22%) The adaptive plan shows similar dose coverage of the tumor (Average: 2%, Maximum: 7%). CONCLUSION There is a high possibility that only image registration for bone and tumor may reduce tumor coverage. Thus, our proposed methodology of image guidance and adaptive planning using the range adaptation after tumor registration would be effective for proton therapy. This research is partially supported by Japan Agency for Medical Research and Development (AMED).

Dosimetric comparison between proton beam therapy and photon radiation therapy for locally advanced esophageal squamous cell carcinoma

BackgroundThe purpose of this study was to perform a dosimetric comparison between proton beam therapy (PBT) and photon radiation therapy in patients with locally advanced esophageal squamous cell carcinoma (ESCC) who were treated with PBT in our institution. In addition, we evaluated the correlation between toxicities and dosimetric parameters, especially the doses to normal lung or heart tissue, to clarify the clinical advantage of PBT over photon radiation therapy.MethodsA total of 37 consecutive patients with Stage III thoracic ESCC who had received PBT with or without concurrent chemotherapy between October 2012 and December 2015 were evaluated in this study. The dose distributions of PBT were compared with those of dummy 3-dimensional conformal radiation therapy (3DCRT) and Intensity Modulated Radiation Therapy (IMRT), focusing especially on the doses to organs at risk, such as normal lung and heart tissue.ResultsOf the 37 patients, the data from 27 patients were analyzed. Among these 27 patients, four patients (15%) developed grade 2 pericardial effusion as a late toxicity. None of the patients developed grade 3 or worse acute or late pulmonary and cardiac toxicities. When the dosimetric parameters between PBT and planned 3DCRT were compared, all the PBT domestic variables for the lung dose except for lung V10 GyE and V15 GyE were significantly lower than those for the dummy 3DCRT plans, and the PBT domestic variables for the heart dose were also significantly lower than those for the dummy 3DCRT plans. When the PBT and IMRT plans were compared, all the PBT domestic variables for the doses to the lung and heart were significantly lower than those for the dummy IMRT plans. Regarding the correlation between the grades of toxicities and the dosimetric parameters, no significant correlation was seen between the occurrence of grade 2 pericardial effusion and the dose to the heart.ConclusionsWhen the dosimetric parameters of the dose distributions for the treatment of patients with locally advanced stage III ESCC were compared between PBT and 3DCRT or IMRT, PBT enabled a significant reduction in the dose to the lung and heart, compared with 3DCRT or IMRT.

Impact of shoulder deformation on volumetric modulated arc therapy doses for head and neck cancer

PURPOSE When using volumetric modulated arc therapy (VMAT) for head and neck cancer, setup errors regarding the shoulders can create loss of target coverage or increased organ-at-risk doses. This study created variations of realistic shoulder deformations to understand the associated VMAT dosimetric effects and investigated water-equivalent thickness (WET) differences using in-house software. METHODS Ten patients with head and neck cancer with lower neck involvement were retrospectively and randomly enrolled. Their retrospective analysis comprised treatment planning using RayStation 5.0 (RaySearch Laboratories, Stockholm, Sweden), shoulder deformation of 5-15 mm in three-dimensional axes using the ImSimQA package (Oncology Systems Limited, Shrewsbury, Shropshire, UK), and evaluation of the clinical impact of the dose distribution after recalculating the dose distribution using computed tomography images of deformed shoulders and deforming the dose distribution. Additionally, our in-house software program was used to measure WET differences for shoulder deformation. RESULTS WET differences were greater in the superoinferior (SI) direction than in the other directions (the WET difference was >20 mm for 15-mm SI deformation). D99%, D98%, and D95% for all clinical target volumes were within 3%. Local dose differences of more than ±10% were found for normal tissues at the level of the shoulder for 15-mm movement in the SI direction. CONCLUSIONS Shoulder deformation of >6 mm could cause large dose variations delivered to the targeted tissue at the level of the shoulder. Thus, to ensure delivery of appropriate treatment coverage to the targeted tissue, shoulder deformation should be taken into consideration during the planning stage.

Evaluation of deformation parameters for deformable image registration-based ventilation imaging using an air-ventilating non-rigid phantom

PURPOSE This study aimed to evaluate different deformable image registration (DIR) parameters for the open-source NiftyReg package in its application to DIR-based ventilation imaging. METHODS Two three-dimensional (3D)-computed tomography (CT) scans of a non-rigid air-ventilating phantom were acquired at peak exhalation and peak inhalation, with xenon (Xe) gas being used as an air-based contrast agent. We compared four different sets of DIR parameters, including one set with two-step deformation and three sets with four-step deformation. For spatial accuracy, the target registration error (TRE) was calculated for 16 landmarks. For ventilation imaging accuracy, DIR-based ventilation images were generated using Jacobian determinant (JD) metrics, and changes in Hounsfield unit (HU) values between the two exhalation and inhalation CT images were subsequently measured. The correlation coefficients between the JD metrics and changes in HU values were calculated. RESULTS The mean TRE values were 4.5 ± 4.7 mm (maximum, 12.3 mm), 1.47 ± 0.71 mm (maximum, 2.6 mm), 1.56 ± 0.70 mm (maximum, 2.8 mm), and 1.53 ± 0.66 mm (maximum, 2.5 mm) for the two-step deformation and three four-step deformations, respectively. The four-step deformations (R =  - 0.71, -0.65, and -0.61) showed stronger correlation coefficients than the two-step deformation (R =  -0.40). CONCLUSIONS The accuracy of DIR-based ventilation imaging may vary with different DIR parameter settings, even though spatial accuracy may be tolerable and within guidelines. We found adequate parameter settings for four-step NiftyReg DIR for visualization of simulated pulmonary ventilation function.

Design and development of a nonrigid phantom for the quantitative evaluation of DIR‐based mapping of simulated pulmonary ventilation

PURPOSE The validation of deformable image registration (DIR)-based pulmonary ventilation mapping is time consuming and prone to inaccuracies and is also affected by deformation parameters. In this study, we developed a nonrigid phantom as a quality assurance (QA) tool that simulates ventilation to evaluate DIR-based images quantitatively. METHODS The phantom consists of an acrylic cylinder filled with polyurethane foam designed to simulate pulmonic alveoli. A polyurethane membrane is attached to the inferior end of the phantom to simulate the diaphragm. In addition, tracheobronchial-tree-shaped polyurethane tubes are inserted through the foam and converge outside the phantom to simulate the trachea. Solid polyurethane is also used to model arteries, which closely follow the model airways. Two three-dimensional (3D) CT scans were performed during exhalation and inhalation phases using xenon (Xe) gas as the inhaled contrast agent. The exhalation 3D-CT image is deformed to an inhalation 3D-CT image using our in-house program based on the NiftyReg open-source package. The target registration error (TRE) between the two images was calculated for 16 landmarks located in the simulated lung volume. The DIR-based ventilation image was generated using Jacobian determinant (JD) metrics. Subsequently, differences in the Hounsfield unit (HU) values between the two images were measured. The correlation coefficient between the JD and HU differences was calculated. In addition, three 4D-CT scans are performed to evaluate the reproducibility of the phantom motion and Xe gas distribution. RESULTS The phantom exhibited a variety of displacements for each landmark (range: 1-20 mm). The reproducibility analysis indicated that the location differences were <1 mm for all landmarks, and the HU variation in the Xe gas distribution was close to zero. The mean TRE in the evaluation of spatial accuracy according to the DIR software was 1.47 ± 0.71 mm (maximum: 2.6 mm). The relationship between the JD and HU differences had a large correlation (R = -0.71) for the DIR software. CONCLUSION The phantom implemented new features, namely, deformation and simulated ventilation. To assess the accuracy of the DIR-based mapping of the simulated pulmonary ventilation, the phantom allows for simulation of Xe gas wash-in and wash-out. The phantom may be an effective QA tool, because the DIR algorithm can be quickly changed and its accuracy evaluated with a high degree of precision.

A multi-institutional study of independent calculation verification in inhomogeneous media using a simple and effective method of heterogeneity correction integrated with the Clarkson method

Abstract In inhomogeneous media, there is often a large systematic difference in the dose between the conventional Clarkson algorithm (C-Clarkson) for independent calculation verification and the superposition-based algorithms of treatment planning systems (TPSs). These treatment site–dependent differences increase the complexity of the radiotherapy planning secondary check. We developed a simple and effective method of heterogeneity correction integrated with the Clarkson algorithm (L-Clarkson) to account for the effects of heterogeneity in the lateral dimension, and performed a multi-institutional study to evaluate the effectiveness of the method. In the method, a 2D image reconstructed from computed tomography (CT) images is divided according to lines extending from the reference point to the edge of the multileaf collimator (MLC) or jaw collimator for each pie sector, and the radiological path length (RPL) of each line is calculated on the 2D image to obtain a tissue maximum ratio and phantom scatter factor, allowing the dose to be calculated. A total of 261 plans (1237 beams) for conventional breast and lung treatments and lung stereotactic body radiotherapy were collected from four institutions. Disagreements in dose between the on-site TPSs and a verification program using the C-Clarkson and L-Clarkson algorithms were compared. Systematic differences with the L-Clarkson method were within 1% for all sites, while the C-Clarkson method resulted in systematic differences of 1–5%. The L-Clarkson method showed smaller variations. This heterogeneity correction integrated with the Clarkson algorithm would provide a simple evaluation within the range of −5% to +5% for a radiotherapy plan secondary check.

Feasibility of dynamic adaptive passive scattering proton therapy with computed tomography image guidance in the lung

Purpose Hypo‐fractionated proton beam therapy (PBT) is an approach that has been increasingly explored over the past decade. It requires high geometric accuracy for targeting of the PBT beams. However, image‐guided PBT is currently commonly performed with kV X‐ray images of bony anatomy. A dynamic adaptive passive scattering PBT system using computed tomography‐based three‐dimensional image guidance was developed, and its effectiveness was then evaluated retrospectively in patients with nonsmall cell lung cancer (NSCLC). Methods The dynamic adaptive PBT system consisted of computed tomography‐based image registration and proton dose calculation using a simplified Monte Carlo algorithm, with a range adaptation system that could adjust the range shifter thickness to alter the dose distribution. Three patients were retrospectively analyzed. All plans, which each had a total dose of 60 Gy (relative biological effectiveness; RBE), were generated using two fields (Gantry angles: 270 degree and 180 degree) in a passive scattering method. Three dose distributions were generated for each patient according to the following different registrations: bone registration, tumor registration, and tumor registration with range adaptation. The following dosimetric parameters were compared with the original plan: target dose coverage at D95% for the clinical target volume (CTV), homogeneity of D5% to D95% for the CTV, and dose distributions in normal tissue (Dmax of Spinal cord and V20 Gy of lung). Results For the bone registration method, the average D95% and D5% to D95% for the CTV showed average differences from the original plan of −3.7 ± 4.1 Gy (mean ± 1SD; RBE) and 3.6 ± 3.9 Gy (RBE) respectively. The tumor registration method achieved better coverage than the bone registration method, although the dosimetric parameters for coverage and homogeneity still showed average differences in −2.0 ± 2.3 Gy (RBE) and 1.9 ± 2.2 Gy (RBE) respectively. The range adaptive plan showed comparable coverage and homogeneity [D95%: −1.0 ± 1.3 Gy (RBE) and D5% to D95%: 0.9 ± 1.0 Gy (RBE) on average] to the original plan, as well as demonstrating similar normal tissue sparing. The approach could be completed in less than 10 min, including CT acquisition, image registration, dose recalculation with range optimization, and the operators visual verification. Conclusions The tumor dose coverage in patients with NSCLC may deteriorate as a result of respiratory or body movement if daily proton range adaptation is not performed. Our approach may provide higher geometric accuracy for localization of the tumor, and the dynamic range adaptation enables us to achieve the planned dose distribution for hypo‐fractionated PBT in the lung.

SU-F-T-479: Estimation of the Accuracy in Respiratory-Gated Radiotherapy

PURPOSE Irregular respiratory patterns affects dose outputs in respiratorygated radiotherapy and there is no commercially available quality assurance (QA) system for it. We designed and developed a patient specific QA system for respiratory-gated radiotherapy to estimate irradiated output. METHODS Our in-house QA system for gating was composed of a personal computer with the USB-FSIO electronic circuit connecting to the linear accelerator (ONCOR-K, Toshiba Medical Systems). The linac implements a respiratory gating system (AZ-733V, Anzai Medical). During the beam was on, 4.2 V square-wave pulses were continually sent to the system. Our system can receive and count the pulses. At first, our system and an oscilloscope were compared to check the performance of our system. Next, basic estimation models were generated when ionization-chamber measurements were performed in gating using regular sinusoidal wave patterns (2.0, 2.5, 4.0, 8.0, 15 sec/cycle). During gated irradiation with the regular patterns, the number of the pulses per one gating window was measured using our system. Correlation between the number of the pulses per one gating and dose per the gating window were assessed to generate the estimation model. Finally, two irregular respiratory patterns were created and the accuracy of the estimation was evaluated. RESULTS Compared to the oscilloscope, our system worked similarly. The basic models were generated with the accuracy within 0.1%. The results of the gated irradiations with two irregular respiratory patterns show good agreement within 0.4% estimation accuracy. CONCLUSION Our developed system shows good estimation for even irregular respiration patterns. The system would be a useful tool to verify the output for respiratory-gated radiotherapy.

SU-C-202-06: Design and Development of a Non-Rigid Phantom Ventilating Air Quantitatively Evaluating CT-Based Pulmonary Ventilation Imaging

PURPOSE A validation study using human or sheep for CT pulmonary ventilation (CT-V) imaging are inefficient and partially unstable. In this study, we designed and developed non-rigid phantom ventilating air to quantitatively evaluate the CT-V image. METHODS The phantom consisted of an acryl cylinder filled with polyurethane foam designed to pulmonic alveoli and a polyurethane membrane was attached to the inferior end of the phantom to simulate a lung diaphragm. Also, trachea-shaped polyurethane tubes were covered by the foam and the tubes passed through the outside in order to exchange a gas. Lung arteries were also modeled with polyurethane and were located adjacent to the trachea model. Two 3D-CT scans were performed at exhale and inhale using an air contrast agent of xenon gas, respectively. The exhale 3D-CT was deformed to the inhale 3D-CT using the in-house program with the NiftyReg. Sixteen landmarks were assigned around trachea and lung arteries in the inhale image and the deformed exhale image. Target Registration Error (TRE) for each landmark between the two images was calculated for commissioning of our program. CT-V image was generated using Jacobian determinant metrics (JD). Subsequently, HU value change between the two images was measured. The correlation coefficient was calculated between the JD and the HU change. RESULTS The deformable and ventilation phantom shows a variety of displacement for each landmark (1 mm - 20 mm). The mean TRE was 4.5 ± 4.7 mm (Maximum: 12.3 mm). The relationship between the JD and the HU change shows high correlation. (R = -0.86). At the several landmarks where TRE values were over 3 mm, the correlation was poor. CONCLUSION The phantom may be a useful QA tool to validate CT-V imaging. We will improve our deformable image registration program for CT-V imaging with the efficient way using the phantom.

SU-E-T-423: Fast Photon Convolution Calculation with a 3D-Ideal Kernel On the GPU

Purpose: The calculation time is a trade-off for improving the accuracy of convolution dose calculation with fine calculation spacing of the KERMA kernel. We investigated to accelerate the convolution calculation using an ideal kernel on the Graphic Processing Units (GPU). Methods: The calculation was performed on the AMD graphics hardware of Dual FirePro D700 and our algorithm was implemented using the Aparapi that convert Java bytecode to OpenCL. The process of dose calculation was separated with the TERMA and KERMA steps. The dose deposited at the coordinate (x, y, z) was determined in the process. In the dose calculation running on the central processing unit (CPU) of Intel Xeon E5, the calculation loops were performed for all calculation points. On the GPU computation, all of the calculation processes for the points were sent to the GPU and the multi-thread computation was done. In this study, the dose calculation was performed in a water equivalent homogeneous phantom with 1503 voxels (2 mm calculation grid) and the calculation speed on the GPU to that on the CPU and the accuracy of PDD were compared. Results: The calculation time for the GPU and the CPU were 3.3 sec and 4.4 hour, respectively. The calculation speed for the GPU was 4800 times faster than that for the CPU. The PDD curve for the GPU was perfectly matched to that for the CPU. Conclusion: The convolution calculation with the ideal kernel on the GPU was clinically acceptable for time and may be more accurate in an inhomogeneous region. Intensity modulated arc therapy needs dose calculations for different gantry angles at many control points. Thus, it would be more practical that the kernel uses a coarse spacing technique if the calculation is faster while keeping the similar accuracy to a current treatment planning system.