Purpose: To measure interfraction setup uncertainty of patients undergoing IMRT treatment who are immobilized by a prototype reinforced thermoplastic mask. Methods and Materials: Two mask designs are utilized depending on the anatomical location of the lesion. To validate patient setup before each treatment fraction, the Novalis Body/ExacTrac system is used. That system uses optically imaged infrared (IR) retroreflectors as well as kilovoltage x-ray imaging to verify patient setup. Patients were immobilized with the Orfit HP IMRT immobilization system, and IR markers were placed anteriorly on the prototype mask. Patients were scanned with a Philips ACQsim CT, and the image set was transferred to BrainScan TPS where the IR markers are identified and a plan is created. The plan is then tranferred to ExacTrac. The patient is initially setup using the IR markers. Two kV x-rays images are taken and fused to the corresponding DRRs generated from the reference dataset used for planning. The patient is moved into position and treated based on the shifts generated from the fusion transformation. The on-going study currently has seven patients with a total of 99 treatment fractions. Results: The average x-ray shift and maximum deviation for each patient will be profiled. The current results show an average vertical, longitudinal, and lateral magnitude of 0.98 mm, 0.96 mm, and 1.02 mm, respectively with a standard deviation of 0.49 mm, 0.65 mm, and 0.64 mm. The maximum observed shift was 4.92 mm and is accredited to a cephalad lesion where there was poor image contrast which affected the fusion quality. Conclusion: This system adequately immobilizes the patient and constrains interfraction setup uncertainty to within two millimeters for most patients. At our institution, this immobilization system is used to treat lesions proximal to critical structures. Conflict of Interest: The masks were provided by Orfit Industries, Belgium.
Purpose: The accurate localization of lung tumors when performing stereotactic body radiotherapy (SBRT) is challenging because the breathing motion can cause imaging artifacts. To address this motion, an internal target volume (ITV) is created based on maximum intensity projections (MIP) reconstructed from all phases of the respiratory cycle by finding the maximum electron density at every voxel within the 4DCT. To assess whether a target volume created from the average intensity projections (AIP) of electron densities can also aid localization, the visualized tumor in cone-beam CT (CBCT) is compared to the ITV as delineated on the 4DCT using the MIP and the AIP datasets. Methods: The tumor was delineated on sixty- five kilovoltage CBCT datasets (CBCT ITV). The distances in each dimension were measured and the mean differences between the CBCT and both the AIP and MIP ITV were analyzed. Results: The mean differences between the CBCT ITV and AIP ITV are 0.06 ± 0.20 cm, −0.10 ± 0.27 cm, and 0.14 ± 0.26 cm in the axial, sagittal, and coronal dimensions respectively. The mean differences between CBCT ITV and the MIP ITV are −0.13 ± 0.19 cm, −0.24 ± 0.31 cm, and −0.06 ± 0.25 cm. In corresponding dimensions, the mean difference between CBCT ITV and AIP ITV is statistically greater than the mean difference between CBCT ITV and MIP ITV (p-value < 0.02) Conclusions: Accurately aligning the target in the lung is difficult without a method to visualize the intended area of treatment. The CBCT ITV is larger than the AIP ITV in the axial and coronal dimensions and is smaller than the MIP ITV in all dimensions. This indicates that CBCT ITV can be better evaluated using AIP ITV and MIP ITV as lower and higher limits for localization, respectively.
Purpose: To compare the calculation times in Raystation treatment planning system (TPS) for different graphical processor units (GPU). Also, to determine the amount of time saved by using a GPU based calculation. Methods: Four different hardware configurations were utilized. (DC1) – Intel Xeon E5‐2650 dual CPU with 64 GB RAM and nVidia Quadro K4000 GPU. (DC1 Titan Black) – Same configuration as (DC1) but with nVidia Titan Black GPU. (RVW1) – Intel Xeon E5‐2650 CPU with 32 GB RAM and nVidia Quadro K4000 GPU. (RVW2) – Intel Xeon E5‐2650 CPU with 16 GB of RAM and Quadro K2000 GPU. Dose calculation was also performed without the GPU on DC1 and RVW2. Seven clinically delivered treatment plans were elected. The plans were picked with varying IMRT complexities. Each plan was calculated consecutively eleven times, and each plan had a uniform dose grid voxel size of 2 mm 3 . Results: When compared to non‐GPU calculation, the average time on DC1 with GPU improved by 21.8% (range: 9.4%–38.9%). Similarly, the average time on RVW2 with GPU improved by 42.3% (range: 35.3%–50.5%). Also, the DC1 Titan Black average calculation time was 48.6% faster (range: 13.1%–73.4%) than the nVidia Quadro GPU models. Overall, the DC1 Titan Black calculation times outperformed DC1 non‐GPU by an average of 58.3% (range: 32.9%–77.5%) For the prostate and nodes study, the DC1 Titan Black reduced times from 724±2.5 seconds 205±1.0 seconds. Conclusion: Dose calculation time is reduced with the addition of a GPU. Also, the nVidia Titan Black GPU provided better time savings than the nVidia Quadro GPU models. Available RAM also plays a role in calculation time due to system lag when the memory reaches capacity. The clinical user must determine the best hardware specifications for his or her institution based on speed but also on monetary cost.
Purpose : Ultrasound, CBCT, and Electro‐Magnetic image‐guided radiotherapy (IGRT) techniques are widely used in prostate localization. Using a single phantom, our aim is to investigate the accuracy of these technologies. Method and Materials : A CT scan (1 mm slice) of an ultrasound phantom with three implanted transponders was acquired for this study. Ten treatment plans with 10 different isocenters within a range of 5 cm from the origin of the phantom were created. The contoured cavity and coordinates of the isocenters were exported to the three systems. The origin of the phantom was aligned to the lasers and the isocenters were localized by the three techniques. During the 30 localization experiments, the phantom was not moved. The measured shifts from each localization method were compared with the known shifts from the treatment plans. The differences between these shifts were analyzed. Results : The mean vector differences are 2.7 ± 0.5 mm, 0.9 ± 0.3 mm and 0.3 ± 0.1 mm for Clarity, CBCT, and Calypso system, respectively. Specifically, the mean differences in the lateral direction are 2.1 ± 0.4 mm, 0.3 ± 0.3 mm, and −0.17 ± 0.09 mm, respectively. In the longitudinal direction, the mean differences are −1.6 ± 0.2 mm, 0.6 ± 0.2 mm, and −0.1 ±0.1 mm, respectively inthe 0.4 ± 0.2 mm, 0.6 ± 0.2 mm, and −0.1 ±0.1 mm, respectively in the vertical direction. The maximum deviation from the known shifts is 3.1 mm for Clarity, followed by 0.9 mm for CBCT, and 0.3 mm for Calypso. These shifts are independent of the distances from each isocenter to the origin of the phantom. Conclusion : The three IGRT systems achieve different accuracy and it is important to understand the limitation of each system. Further studies for the impact of rotation on localization are warranted.
Purpose: To measure and compare the changes in dose distributions produced by the presence of a carbon fiber extension board (used to provide necessary patient extension beyond the gun end of the treatment couch). The tested carbon fiber extension boards are designed to be “radiotranslucent” so that the transmitted radiation beam is not significantly altered. We measured the attenuation due to the boards being placed in the beam's path, compared the relative dose distributions obtained using different boards, and also examined the effect of “build‐up”. Method and Materials: The three boards evaluated were the Med‐Tec Type‐S system, the Brainscan H&N Tx system, and the Orfit HP long baseplate for IMRT. Dosimetric data were obtained for 6 MV photons with an Exradin A12, 0.65 cm 3 Farmer ionization chamber, with Kodak EDR film, and with an Exradin P11, 0.62 cm 3 plane parallel chamber (for measuring dose buildup in a Gammex RMI solid water phantom). Results: The Orfit board's maximum attenuation was 1.2%. The Med‐Tec Type‐S board's maximum attenuation was 2.3%. The Brainscan board's maximum attenuation was 13.7%. The isodose distribution and profiles for each board will be presented. Conclusion: We determined that the Orfit board did not significantly alter the radiation beam and resulting dose distribution. Its adaptability to various tables was favorable compared to the Med‐Tec system, which is well suited for the Varian Exact couch. The Med‐Tec board is also well suited for therapeutic radiation, but had a higher change in attenuation than the Orfit system. The Brainscan board was sub‐optimal due to non‐air equivalent material used as structural support in the board's interior, which caused significant beam attenuation. Conflict of Interest: Research was supported by equipment loan from Orfit Industries and Med‐Tec, Inc.
PURPOSE: To evaluate interfraction pancreatic motion as evidence by daily image guided radiotherapy (IGRT) using daily CBCT. MATERIALS AND METHODS: Daily kilovoltage CBCT was used during the definitive treatment of seven patients. Daily CBCT images were registered to planning CT with automatic software tools that put higher importance on boney anatomy as proxy for pancreatic motion. Validation scans were employed when preset constraints were not met. A single treating physician reviewed and approved each fusion. Data regarding interfraction deviation from these daily shifts are analyzed for translation and rotation in the anterior‐posterior (AP), right‐left (RL), and superior‐inferior (SI) directions. RESULTS: One hundred and eighty registration scans were utilized and revealed average shifts of 0.24, 0.15, and .01 cm in the SI, AP, and RL directions, respectively with standard deviations of 0.42, 0.27, and 0.45 cm. Using this data in the Van‐Herk model suggests treatment (PTV) margins of 0.39 cm in RL, 0.91 cm in SI, and 0.58 cm in AP direction to ensure that 90% of all of the clinical target volumes (CTV) receive 95% of the prescribed dose. The average vector displacement was 0.61 cm with a standard deviation of 0.41 cm. The maximum observed vector shift was 1.95 cm in the SI direction. The registration yielded an average rotation of 0.97°, −.06°, and −.20° in the coronal, sagittal, and axial planes respectively with standard deviations of 0.8°, 1.25°, and 1.1°. The maximum rotation noted in any direction was 4.6°. CONCLUSIONS: CBCT is an important tool to monitor interfraction pancreatic motion. Bony fusion as proxy provides some confidence and may translate into smaller margins. However, further assessment with soft tissue fusion and the use of internal fiducials are needed to help provide further confidence.
Purpose: Non-flat beam is attractive for SBRT and IMRT delivery. This study is to investigate whether non-flat beam can achieve uniform dose distribution for whole breast irradiation. Method and Materials: A 6MV non-flat and flat beams were commissioned in an Oncor Linac (Siemens). Fifteen early-stage breast cancer patients were selected. For each patient, three plans were generated for comparison: (a) clinically approved forward planning (FP); (b) mixed open fields using flat beams with direct aperture openand DAO IMRT fields using non-flat beams (mixed DAO); (c) All open and DAO IMRT fields using non-flat beams (non-flat DAO). All plans were prescribed for >95% of the breast volume receiving the prescription dose. Plan quality was evaluated according to homogeneity index (HI), conformality index (COIN), D95%, V105% and D1cc as well as V20Gy to ipsilateral lung, V1Gy to contralateral breast and V25Gy to heart for left breast. All treatment plans were created using Pinnacle 8.0. Results: No significant difference was observed for COIN among all plans (all p>0.05). Average HIs of FP, mixed DAO, and non-flat DAO were 0.88±0.01, 0.88±0.01 and 0.87±0.01, respectively. The mixed DAO had significantly lower V105% than FP and non-flat DAO. For organs at risk (OAR), no significant difference was observed. For D1cc, non-flat DAO was significantly higher than FP and mixed DAO (p<0.005). The lowest MU required was FP plan without wedges, followed by mixed DAO, FP with wedges, and non-flat DAO plans. Conclusion: Both mixed DAO and non-flat DAO plans can achieve equivalent planning quality as the clinically approved flat-beam FP plans. In addition, the mixed DAO plan has smaller hotspot area than non-flat DAO and FP while dose to OARs is not significantly different. Overall, non-flat beams are applicable to breast RT.
'%3e%3ccircle r='20' fill='%23008'/%3e%3ccircle r='17.5' fill='%23fff'/%3e%3ccircle r='3.5' fill='%23008'/%3e%3cg id='d'%3e%3cg id='c'%3e%3cg id='b'%3e%3cg id='a' fill='%23008'%3e%3ccircle r='.875' transform='rotate(7.5 -8.75 133.5)'/%3e%3cpath d='M0 17.5.6 7 0 2l-.6 5L0 17.5z'/%3e%3c/g%3e%3cuse xlink:href='%23a' transform='rotate(15)'/%3e%3c/g%3e%3cuse xlink:href='%23b' transform='rotate(30)'/%3e%3c/g%3e%3cuse xlink:href='%23c' transform='rotate(60)'/%3e%3c/g%3e%3cuse xlink:href='%23d' transform='rotate(120)'/%3e%3cuse xlink:href='%23d' transform='rotate(-120)'/%3e%3c/g%3e%3c/svg%3e)