E Heath

Development and validation of a BEAMnrc component module for accurate Monte Carlo modelling of the Varian dynamic Millennium multileaf collimator

A new component module (CM), designated DYNVMLC, was developed to fully model the geometry of the Varian Millennium 120 leaf collimator using the BEAMnrc Monte Carlo code. The model includes details such as the leaf driving screw hole, support railing groove and leaf tips. Further modifications also allow sampling of leaf sequence files to simulate the movement of the multileaf collimator (MLC) leaves during an intensity modulated radiation therapy (IMRT) delivery. As an initial validation of the code, the individual leaf geometries were visualized by tracing particles through the component module and recording their position each time a leaf boundary was crossed. A model of the Varian CL21EX linear accelerator 6 MV photon beam incorporating the new CM was built with the BEAMnrc user code. The leaf material density and abutting leaf air gap were chosen to match simulated leaf leakage profiles with film measurements in a solid water phantom. Simulated depth dose and off-axis profiles for a variety of MLC defined static fields agreed to within 2% with ion chamber and diode measurements in a water phantom. Simulated dose distributions for IMRT intensity patterns delivered using both static and dynamic techniques were found to agree with film measurements to within 4%. A comparison of interleaf leakage profiles for the new CM and an equivalent leaf model using the existing VARMLC CM demonstrated that the simplified geometry of VARMLC is not able to accurately predict the details of the MLC leakage for the 120 leaf collimator.

A direct voxel tracking method for four-dimensional Monte Carlo dose calculations in deforming anatomy

In this work we present a method of calculating dose in deforming anatomy where the position and shape of each dose voxel is tracked as the anatomy changes. The EGSnrc/DOSXYZnrc Monte Carlo code was modified to calculate dose in voxels that are deformed according to deformation vectors obtained from a nonlinear image registration algorithm. The defDOSXYZ code was validated by consistency checks and by comparing calculations against DOSXYZnrc calculations. Calculations in deforming phantoms were compared with a dose remapping method employing trilinear interpolation. Dose calculations with the deforming voxels agree with DOSXYZnrc calculations within 1%. In simple deforming rectangular phantoms the trilinear dose remapping method was found to underestimate the dose by up to 29% for a 1.0 cm voxel size within the field, with larger discrepancies in regions of steep dose gradients. The agreement between the two calculation methods improved with smaller voxel size and deformation magnitude. A comparison of dose remapping from Inhale to Exhale in an anatomical breathing phantom demonstrated that dose deformations are underestimated by up to 16% in the penumbra and 8% near the surface with trilinear interpolation.

Dosimetric evaluation of the clinical implementation of the first commercial IMRT Monte Carlo treatment planning system at 6 MV

In this work we dosimetrically evaluated the clinical implementation of a commercial Monte Carlo treatment planning software (PEREGRINE, North American Scientific, Cranberry Township, PA) intended for quality assurance (QA) of intensity modulated radiation therapy treatment plans. Dose profiles calculated in homogeneous and heterogeneous phantoms using this system were compared to both measurements and simulations using the EGSnrc Monte Carlo code for the 6 MV beam of a Varian CL21EX linear accelerator. For simple jaw-defined fields, calculations agree within 2% of the dose at d(max) with measurements in homogeneous phantoms with the exception of the buildup region where the calculations overestimate the dose by up to 8%. In heterogeneous lung and bone phantoms the agreement is within 3%, on average, up to 5% for a 1 x 1 cm2 field. We tested two consecutive implementations of the MLC model. After matching the calculated and measured MLC leakage, simulations of static and dynamic MLC-defined fields using the most recent MLC model agreed to within 2% with measurements.

IMRT head and neck treatment planning with a commercially available Monte Carlo based planning system

The PEREGRINE Monte Carlo dose-calculation system (North American Scientific, Cranberry Township, PA) is the first commercially available Monte Carlo dose-calculation code intended specifically for intensity modulated radiotherapy (IMRT) treatment planning and quality assurance. In order to assess the impact of Monte Carlo based dose calculations for IMRT clinical cases, dose distributions for 11 head and neck patients were evaluated using both PEREGRINE and the CORVUS (North American Scientific, Cranberry Township, PA) finite size pencil beam (FSPB) algorithm with equivalent path-length (EPL) inhomogeneity correction. For the target volumes, PEREGRINE calculations predict, on average, a less than 2% difference in the calculated mean and maximum doses to the gross tumour volume (GTV) and clinical target volume (CTV). An average 16% +/- 4% and 12% +/- 2% reduction in the volume covered by the prescription isodose line was observed for the GTV and CTV, respectively. Overall, no significant differences were noted in the doses to the mandible and spinal cord. For the parotid glands, PEREGRINE predicted a 6% +/- 1% increase in the volume of tissue receiving a dose greater than 25 Gy and an increase of 4% +/- 1% in the mean dose. Similar results were noted for the brainstem where PEREGRINE predicted a 6% +/- 2% increase in the mean dose. The observed differences between the PEREGRINE and CORVUS calculated dose distributions are attributed to secondary electron fluence perturbations, which are not modelled by the EPL correction, issues of organ outlining, particularly in the vicinity of air cavities, and differences in dose reporting (dose to water versus dose to tissue type).

Quantification of accuracy of the automated nonlinear image matching and anatomical labeling (ANIMAL) nonlinear registration algorithm for 4D CT images of lung

The performance of the ANIMAL (Automated Nonlinear Image Matching and Anatomical Labeling) nonlinear registration algorithm for registration of thoracic 4D CT images was investigated. The algorithm was modified to minimize the incidence of deformation vector discontinuities that occur during the registration of lung images. Registrations were performed between the inhale and exhale phases for five patients. The registration accuracy was quantified by the cross-correlation of transformed and target images and distance to agreement (DTA) measured based on anatomical landmarks and triangulated surfaces constructed from manual contours. On average, the vector DTA between transformed and target landmarks was 1.6 mm. Comparing transformed and target 3D triangulated surfaces derived from planning contours, the average target volume (GTV) center-of-mass shift was 2.0 mm and the 3D DTA was 1.6 mm. An average DTA of 1.8 mm was obtained for all planning structures. All DTA metrics were comparable to inter observer uncertainties established for landmark identification and manual contouring.

Continuous aperture dose calculation and optimization for volumetric modulated arc therapy

Although VMAT delivery features continuous gantry rotation and leaf motion, dose calculation is often performed under the dual assumption of discrete apertures changing instantaneously from one discrete angle to the next. In this work, the validity of these two approximations is determined, as well as their impact on the quality of optimized plans. Further, an accurate method of fluence calculation is derived which does not use the discrete aperture approximation, but instead calculates the fluence as the multi-leaf collimator leaves sweep from one position to another. This continuous aperture fluence calculation is integrated in the VMAT optimization process using the open-source treatment planning system matRad. The three-step approach of VMAT optimization is used: fluence map optimization followed by leaf sequencing and direct aperture optimization, with variable leaf speed, gantry rotation speed, and MU rate. The benefit of the continuous aperture VMAT method over the discrete aperture method is determined by comparing the plan quality of discrete aperture and continuous aperture optimized plans, when the former is recalculated using the continuous aperture fluence calculation. Discrete aperture VMAT plans calculated at 4° spacing result in significant dose errors (10%-35%, depending on the anatomical site) as compared to the reference dose (continuous aperture fluence calculation at 0.5° spacing). These errors are greatly reduced (to 0.8%-2%) when the continuous aperture fluence calculation method was used at the same 4° spacing, implying that the dose error is primarily due to the discrete aperture approximation. Whereas all dose objectives were met by the discrete aperture VMAT optimized plan, many of them failed when the dose was recalculated with the continuous aperture fluence calculation. All objectives were met once again when the plan was optimized with the new continuous aperture VMAT optimization. Further, using only half of the beam angles, the continuous aperture VMAT optimization can achieve the same degree of accuracy with only 40% of the computing time as compared with the standard discrete aperture VMAT.

TH‐E‐M100J‐01: A Novel Deformable Lung Phantom for 4D Radiotherapy Verification

Purpose: To develop a reproducible, tissue equivalent, deformable lung phantom for verification of 4D‐CT scanning procedures, deformable image registration (DIR) and 4D dose calculation in moving/deformable anatomies.Methods and Materials: The phantom consists of a Lucite cylinder filled with water containing a latex balloon filled with dampened natural sponges. The balloon is attached to a piston that mimics the human diaphragm. Nylon wires and Lucite beads, emulating vascular and bronchial bifurcations, were glued at various locations, uniformly, throughout the sponges. The phantom is capable of simulating programmed irregular breathing patterns with varying periods and amplitudes. A deformable, tissue equivalent tumour holding radiochromic film was embedded in the sponge. Eight 3D‐CT datasets (0.7×0.7×1.25 mm) of the phantom in eight static positions of the piston were acquired. 3D trajectories of 12 anatomical point landmarks as well as the tumour center‐of‐mass were studied. Results: Reproducible lung deformation is achieved by piston‐provoked pressure changes in water surrounding the deforming balloon. The resulting mean density for the artificial lung was 0.243 g/cm3 comparable to 0.252 g/cm3 for a real lung. A truly 3D, non‐isotropic deformation of the balloon similar to a real lung has been obtained. The SI displacement of the landmarks varies between 94% and 3% of the diaphragm excursion for positions closer and farther away from the diaphragm, respectively. Reproducibility in the deformed phantom, established by seven repeat scans at the same phantom compression state, was within image resolution. The accuracy of DIR of the extreme phases was 0.7(±0.7) mm. Conclusions: Our novel phantom is tissue‐equivalent, deformable, and can reproducibly emulate 3D non‐isotropic lung deformations. The presence of vascular and bronchial bifurcations allows verification of DIR of 4D‐CT images of the phantom. Registered phases of the phantom can be used in 4D dose calculations that can be validated by comparison with dose measurements.

SU‐FF‐J‐130: Validation of Non‐Linear Image Registration‐Based Correction Method for Motion Artifacts in 4D‐CT

Purpose: Motion artifacts in CT can be reduced by using 4D‐CT acquisition techniques where image slices are retrospectively binned according to breathing phase as determined by a breathing trace. However, motion artifacts may still occur due to breathing irregularities. Such artifacts affect organ delineation and lead to complications when using 4D CT data for dose calculation and accumulation. We propose a method for correcting such artifacts by temporal interpolation using non‐linear image registration.Method and Materials: The ANIMAL non‐linear image registration algorithm was used to determine the transformation between artifact‐free phases adjacent to the phase containing the motion artifacts. The weighting factor which, when applied to the transformation, most closely reconstructs the anatomy at the phase to be corrected was determined. CT values in regions of the image containing the artifacts were then replaced with the corresponding CT values from the reconstructed image. The accuracy of the temporal interpolation method was evaluated by simulating motion artifacts resulting from different breathing amplitudes using the NCAT numerical breathing phantom for which the artifact‐free image is available by definition. The reconstructed image was compared to the artifact‐free image. The temporal interpolation method was applied to correct motion artifacts in patient 4D CT data and the corrected images were compared to physician‐delineated contours. Results: Correlation between the NCAT phantom images with and without artifacts was improved from 0.971 to 0.992 after correction of the artifacts by temporal interpolation. The quality of the patient 4D CT data was improved after temporal interpolation and the reconstructedanatomy was consistent with manual contours. Conclusion: We have developed a method for reconstructinganatomy on 4D CTimages in the presence of motion artifacts. The temporal interpolation method was demonstrated to reduce the appearance of these artifacts and therefore improve the accuracy of organ delineation and dose calculation.

SU‐FF‐J‐27: Novel 4D CT Scanning Protocol Using a Helical Single‐Slice CT Scanner

Purpose: In radiotherapy, respiratory motion poses a significant challenge during tumorsimaging,treatment planning and radiation delivery. A method to alleviate respiratory motion problems is to use 4D radiotherapy. We propose novel 4D CT scanning protocol using a helical single‐slice CTscanner (SSCT). Method and Materials: In our protocol, patients torso is scanned three times using the helical mode of a SSCT scanner.CT slices are acquired simultaneously with real‐time tracking of a marker placed on patients torso. At the end of the three scans, CT data is binned into different respiratory phases according to the externally recorded respiratory signal and the volume of interest is reconstructed for several respiratory phases. Results: The protocol was tested on an anthropomorphic phantom to which a realistic respiratory motion was induced by placing it on an inflatable mattress driven by an air pump controlled by a pulsing power supply. 4D CTimages were compared with images obtained after a conventional scan of the static phantom and with the conventional scan of the “breathing” phantom. 4D CTimages show a net improvement with respect to conventional CTimages. The ANIMAL deformable registration algorithm was used to calculate a 3D vector mesh which maps volumes at any given phase to the exhale volume. This vector map and the external breathing signal are used to reconstruct the anatomy at any breathing phase. Conclusion: Our scanning protocol uses the helical scan mode so it is faster than the axial/cine scanning protocols. The gap between CT slices available for each breathing scan is alleviated by taking three successive scans but the use of low tube current keeps the dose to the patient to an acceptable level. Our protocol is easy to implement in any clinic where a single‐slice CTscanner and a real‐time motion tracking system are available.

SU‐FF‐T‐286: Dosimetric Evaluation of the PEREGRINE IMRT MC Treatment Planning System at 6 MV for Small Fields in Heterogeneous Media

Purpose: In a previous study Heath et al. reported good dosimetric accuracy between PEREGRINE and measurements in homogeneous and heterogeneous media, for 1×1 to 30×30 cm2 fields. However, dosimetric accuracy for large field horns and inside the lung‐equivalent phantom remained subject to further investigation. This work compares the effect of parameters directly related to the phantom with those related to linac beam modeling and discusses their effect on the calculated dose inside the heterogeneity. Method and Materials: To investigate the factors influencing the accuracy of 1×1 cm2 depth dose profiles in lung, several dose calculations were performed and the effect of the following parameters were studied: the mathematical phantoms resolution, slice thickness, composition, density and dose collection voxel shape and size; and parameters influencing the PEREGRINE device file (which is the MC‐derived correlated‐histograms model of the beam) such as the width and shape of the electron beam (incident on the linac target) intensity distribution. Results: Our results show that modeling the lung component of the phantom as GAMMEX‐RMI lung‐equivalent material (ρ = 0.271 g/cm3) or as lungtissue (ρ = 0.26 g/ cm3) results in less than 1% difference in dose to the lung, whereas using a device file with an optimized electron beam set of parameters to match the large field off‐axis dose profiles results in 3% dose to lung difference. The agreement in dose to lung between this version of the device file and corresponding EGSnrc calculations is within 1%. Conclusion: When performing dosimetric verification calculations in heterogeneous media especially for small fields, attention must be paid to the effect of the details of the linac beam MCmodel (particularly the electron beam parameters) on the calculated dose in heterogeneities. Conflict of Interest: This work is supported by North American Scientific (Nomos Radiation Oncology Division).