Generation of photon energy deposition kernels using the EGS Monte Carlo code

A dose calculation algorithm for boron neutron capture therapy using convolution/superposition method

The convolution/superposition (C/S) method originally designed for photon dose calculation was first applied for developing a treatment planning system for boron neutron capture therapy. The original concept of TEGMA (total energy generated per unit mass) was proposed to represent distinctive dose components from neutron reactions with the elements in the patient's tissue. First, neutron fluence distributions in a homogeneous brain phantom irradiated with an energy-groupwise pencil beam of 2.5 × 2.5 mm2 were calculated using the MCNP6.2 code. Then, a library of energy-groupwise TEGMA and KERMA were generated and stored in the developed C/S code. As a benchmark, dose distributions in a cuboid phantom and a human head phantom were calculated using the developed C/S and PHITS Monte Carlo codes. A neutron beam having a continuous epithermal spectrum and a square field of 22.5 × 22.5 mm2 or a circle field of 22.5 mm in diameter was assumed to be incident on the phantoms. The human head phantom was created by the pre-processing including the voxelization and transformation of test DICOM CT images. The differences in boron doses between C/S and MC ranged from 2% to 6%. In nitrogen doses, the differences were from 4% to 9%. A large discrepancy observed in hydrogen lateral dose profiles could be explained by the differences in cross-section data and recoil-proton transport algorithms of MCNP6.2 and PHITS. With isodose curves normalized at the center of the tumor in the human head phantom, they were almost identical in the range of 60%-110% for both cases. The C/S have underestimated the backscattering neutron and showed a larger absorbed dose gradient around 40% region. The calculation time of C/S using Intel i7-10700 processor was less than 1 min for both phantoms. The calculation time of PHITS using three Intel Xeon E5-2640 v4 processors was 15.5 min for the cuboid phantom and ∼380 min for the human head phantom. The proposed algorithm has the advantages of high speed while promising fair accuracy in BNCT dose calculations.

A deep-learning-based dose verification tool utilizing fluence maps for a cobalt-60 compensator-based intensity-modulated radiation therapy system

Background and purpose

A novel cobalt-60 compensator-based intensity-modulated radiation therapy (IMRT) system was developed for a resource-limited environment but lacked an efficient dose verification algorithm. The aim of this study was to develop a deep-learning-based dose verification algorithm for accurate and rapid dose predictions.

Materials and methods

A deep-learning network was employed to predict the doses from static fields related to beam commissioning. Inputs were a cube-shaped phantom, a beam binary mask, and an intersecting volume of the phantom and beam binary mask, while output was a 3-dimensional (3D) dose. The same network was extended to predict patient-specific doses for head and neck cancers using two different approaches. A field-based method predicted doses for each field and combined all calculated doses into a plan, while the plan-based method combined all nine fluences into a plan to predict doses. Inputs included patient computed tomography (CT) scans, binary beam masks, and fluence maps truncated to the patient's CT in 3D.

Results

For static fields, predictions agreed well with ground truths with average deviations of less than 0.5% for percent depth doses and profiles. Even though the field-based method showed excellent prediction performance for each field, the plan-based method showed better agreement between clinical and predicted dose distributions. The distributed dose deviations for all planned target volumes and organs at risk were within 1.3 Gy. The calculation speed for each case was within two seconds.

Conclusions

A deep-learning-based dose verification tool can accurately and rapidly predict doses for a novel cobalt-60 compensator-based IMRT system.

Comparison of four commercial dose calculation algorithms in different evaluation tests

BACKGROUND

Accurate and fast dose calculation is crucial in modern radiation therapy. Four dose calculation algorithms (AAA, AXB, CCC, and MC) are available in Varian Eclipse and RaySearch Laboratories RayStation Treatment Planning Systems (TPSs).

OBJECTIVES

This study aims to evaluate and compare dosimetric accuracy of the four dose calculation algorithms applying to homogeneous and heterogeneous media, VMAT plans (based on AAPM TG-119 test cases), and the surface and buildup regions.

METHODS

The four algorithms are assessed in homogeneous (IAEA-TECDOCE 1540) and heterogeneous (IAEA-TECDOC 1583) media. Dosimetric evaluation accuracy for VMAT plans is then analyzed, along with the evaluation of the accuracy of algorithms applying to the surface and buildup regions.

RESULTS

Tests conducted in homogeneous media revealed that all algorithms exhibit dose deviations within 5% for various conditions, with pass rates exceeding 95% based on recommended tolerances. Additionally, the tests conducted in heterogeneous media demonstrate high pass rates for all algorithms, with a 100% pass rate observed for 6 MV and mostly 100% pass rate for 15 MV, except for CCC, which achieves a pass rate of 94%. The results of gamma index pass rate (GIPR) for dose calculation algorithms in IMRT fields show that GIPR (3% /3 mm) for all four algorithms in all evaluated tests based on TG119, are greater than 97%. The results of the algorithm testing for the accuracy of superficial dose reveal variations in dose differences, ranging from -11.9% to 7.03% for 15 MV and -9.5% to 3.3% for 6 MV, respectively. It is noteworthy that the AXB and MC algorithms demonstrate relatively lower discrepancies compared to the other algorithms.

CONCLUSIONS

This study shows that generally, two dose calculation algorithms (AXB and MC) that calculate dose in medium have better accuracy than other two dose calculation algorithms (CCC and AAA) that calculate dose to water.

A novel analytical method for computing dose from kilovoltage beams used in Image-Guided radiation therapy

PURPOSE

A modified convolution/superposition algorithm is proposed to compute dose from the kilovoltage beams used in IGRT. The algorithm uses material-specific energy deposition kernels instead of water-energy deposition kernels.

METHODS

Monte Carlo simulation was used to model the Elekta XVI unit and determine dose deposition characteristics of its kilovoltage beams. The dosimetric results were compared with ion chamber measurements. The dose from the kilovoltage beams was then computed using convolution/superposition along with material-specific energy deposition kernels and compared with Monte Carlo and measurements. The material-specific energy deposition kernels were previously generated using Monte Carlo.

RESULTS

The obtained gamma indices (using 2%/2mm criteria for 95% of points) were lower than 1 in almost all instances which indicates good agreement between simulated and measured depth doses and profiles. The comparisons of the algorithm with measurements in a homogeneous solid water slab phantom, and that with Monte Carlo in a head and neck CT dataset produced acceptable results. The calculated point doses were within 4.2% of measurements in the homogeneous phantom. Gamma analysis of the calculated vs. Monte Carlo simulations in the head and neck phantom resulted in 94% of points passing with a 2%/2mm criteria.

CONCLUSIONS

The proposed method offers sufficient accuracy in kilovoltage beams dose calculations and has the potential to supplement the conventional megavoltage convolution/superposition algorithms for dose calculations in low energy range.

Impact of geometric distortion on dose deviation for photon and proton treatment plans

We investigated the dose deviation related to geometric distortion and dose gradient on magnetic resonance‐only treatment planning for intensity‐modulated radiation therapy and proton therapy. The residual geometric distortion of two different magnetic resonance imaging (MRI) sequences (A) and (B) was applied in the computed tomography image and the structure set of each patient through a polynomial MRI geometric distortion model to simulate MRI‐based treatment planning. A 3D histogram was generated to specify the relationship of dose deviation to geometric distortion and dose gradient. When the dose gradient (G_d) approached zero, the maximum dose deviation reached 1.64% and 2.71% for photon plans of sequences A and B, respectively. For proton plans, the maximum dose deviation reached 3.15% and 4.89% for sequences A and B, respectively. When the geometric distortion (d) was close to zero, the maximum dose deviation was less than 0.8% for photon and proton plans of both sequences. Under extreme conditions (d = 2 mm and G_d = 4.5%/mm), the median value of dose deviation reached 3% and 3.49% for photon and proton plans, respectively for sequence A, and 2.93% and 4.55% for photon and proton plans, respectively, for sequence B. We demonstrate that the dose deviation is specific to MRI hardware parameters. Compared to the photon plan, the proton plan is more sensitive to the changes in geometric distortion. For typical clinical MRI geometric distortion (d ≤2 mm), the median dose deviation is expected to be within 3% and 5% for photon and proton plans, respectively.

Introducing matrix sparsity with kernel truncation into dose calculations for fluence optimization

Deep learning algorithms for radiation therapy treatment planning automation require large patient datasets and complex architectures that often take hundreds of hours to train. Some of these algorithms require constant dose updating (such as with reinforcement learning) and may take days. When these algorithms rely on commerical treatment planning systems to perform dose calculations, the data pipeline becomes the bottleneck of the entire algorithm's efficiency. Further, uniformly accurate distributions are not always needed for the training and approximations can be introduced to speed up the process without affecting the outcome. These approximations not only speed up the calculation process, but allow for custom algorithms to be written specifically for the purposes of use in AI/ML applications where the dose and fluence must be calculated a multitude of times for a multitude of different situations. Here we present and investigate the effect of introducing matrix sparsity through kernel truncation on the dose calculation for the purposes of fluence optimzation within these AI/ML algorithms. The basis for this algorithm relies on voxel discrimination in which numerous voxels are pruned from the computationally expensive part of the calculation. This results in a significant reduction in computation time and storage. Comparing our dose calculation against calculations in both a water phantom and patient anatomy in Eclipse without heterogenity corrections produced gamma index passing rates around 99% for individual and composite beams with uniform fluence and around 98% for beams with a modulated fluence. The resulting sparsity introduces a reduction in computational time and space proportional to the square of the sparsity tolerance with a potential decrease in cost greater than 10 times that of a dense calculation allowing not only for faster caluclations but for calculations that a dense algorithm could not perform on the same system.

Heart-centered positioning and tailored beam-shaping filtration for reduced radiation dose in coronary artery calcium imaging: A Multi-Ethnic Study of Atherosclerosis (MESA) Study

PURPOSE

Cardiac computed tomography has a clear clinical role in the evaluation of coronary artery disease and assessment of coronary artery calcium (CAC) but the use of ionizing radiation limits the clinical use. Beam-shaping "bow-tie" filters determine the radiation dose and the effective scan field-of-view diameter (SFOV) by delivering higher X-ray fluence to a region centered at the isocenter. A method for positioning the heart near the isocenter could enable reduced SFOV imaging and reduce dose in cardiac scans. However, a predictive approach to center the heart, the extent to which heart centering can reduce the SFOV, and the associated dose reductions have not been assessed. The purpose of this study is to build a heart-centered patient positioning model, to test whether it reduces the SFOV required for accurate CAC scoring, and to quantify the associated reduction in radiation dose.

METHODS

The location of 38,184 calcium lesions (3151 studies) in the Multi-Ethnic Study of Atherosclerosis was utilized to build a predictive heart-centered positioning model and compare the impact of SFOV on CAC scoring accuracy in heart-centered and conventional body-centered scanning. Then, the positioning model was applied retrospectively to an independent, contemporary cohort of 118 individuals (81 with CAC > 0) at our institution to validate the model's ability to maintain CAC accuracy while reducing the SFOV. In these patients, the reduction in dose associated with a reduced SFOV beam-shaping filter was quantified.

RESULTS

Heart centering reduced the SFOV diameter 25.7% relative to body centering while maintaining high CAC scoring accuracy (0.82% risk reclassification rate). In our validation cohort, imaging at this reduced SFOV with heart-centered positioning and tailored beam-shaping filtration led to a 26.9% median dose reduction (25-75th percentile: 21.6%-29.8%) without any calcium risk reclassification.

CONCLUSIONS

Heart-centered patient positioning enables a significant radiation dose reduction while maintaining CAC accuracy.

Dose Calculation Algorithms for External Radiation Therapy: An Overview for Practitioners

Radiation therapy (RT) is a constantly evolving therapeutic technique; improvements are continuously being introduced for both methodological and practical aspects. Among the features that have undergone a huge evolution in recent decades, dose calculation algorithms are still rapidly changing. This process is propelled by the awareness that the agreement between the delivered and calculated doses is of paramount relevance in RT, since it could largely affect clinical outcomes. The aim of this work is to provide an overall picture of the main dose calculation algorithms currently used in RT, summarizing their underlying physical models and mathematical bases, and highlighting their strengths and weaknesses, referring to the most recent studies on algorithm comparisons. This handy guide is meant to provide a clear and concise overview of the topic, which will prove useful in helping clinical medical physicists to perform their responsibilities more effectively and efficiently, increasing patient benefits and improving the overall quality of the management of radiation treatment.

Generation of material-specific energy deposition kernels for kilovoltage x-ray dose calculations

PURPOSE

Dose calculation of kilovoltage x rays used in Image-Guided Radiotherapy has been investigated in recent years using various methods. Among these methods are model-based ones that suffer from inaccuracies in high-density materials and at interfaces when used in the kilovoltage energy range. The main reason for this is the use of water energy deposition kernels and simplifications employed such as density scaling in heterogeneous media. The purpose of this study was to produce and characterize material-specific energy deposition kernels, which could be used for dose calculations in this energy range. These kernels will also have utility in dose calculations in superficial radiation therapy and orthovoltage beams utilized in small animal irradiators.

METHODS

Water energy deposition kernels with various resolutions; and high-resolution, material-specific energy deposition kernels were generated in the energy range of 10-150 kVp, using the EGSnrc Monte Carlo toolkit. The generated energy deposition kernels were further characterized by calculating the effective depth of penetration, the effective radial distance, and the effective lateral distance. A simple benchmarking of the kernels against Monte Caro calculations has also been performed.

RESULTS

There was good agreement with previously reported water kernels, as well as between kernels with different resolution. The evaluation of effective depth of penetration, and radial and laterals distances, defines the relationship between energy, material density, and the shape of the material-specific kernels. The shape of these kernels becomes more forwardly scattered as the energy and material density are increased. The comparison of the dose calculated using the kernels with Monte Carlo provides acceptable results.

CONCLUSIONS

Water and material-specific energy deposition kernels in the kilovoltage energy range have been generated, characterized, and compared to previous work. These kernels will have utility in dose calculations in this energy range once algorithms capable of employing them are fully developed.

Investigation of the Scaling Formula Accuracy for Poly-energetic Kernel Calculation in 6 MV Photon Beam

Background:

Dose distribution can be obtained from total energy released per unit mass (TERMA) and inhomogeneous energy deposition kernel (EDK) convolution. Since inhomogeneous EDK data is location-dependent, it is calculated by employing the density scaling method rather than Monte Carlo based user code EDKnrc.

Objective:

The present study aimed at investigating EDK scaling formula accuracy in the presence of lung and bone inhomogeneities.

Material and Methods:

In this theoretical-practical study, six EDKs datasets with lung and bone inhomogeneity in different radii were generated using EDKnrc user code and density scaling formula. Then the scaling method data and corresponding EDKnrc-generated ones were compared to enhance the calculations, and some correction factors for error reduction were also derived to create more consistency between these data.

Results:

The study has shown that the errors in the theoretical method for calculating inhomogeneous EDKs were significantly reduced based on the attenuation coefficient and ρ^α_rel parameter, with α equal to 1.2 and 0.8 for bone and lung voxels, respectively.

Conclusion:

Although the density scaling method has acceptable accuracy, the error values are significant at the location of lung or bone voxels. By using the mentioned correction factors, the calculation inaccuracy of heterogeneous EDKs can be reduced down to 5%. However, the lung heterogeneity results corrected by the method are not as good as the bone cases.

Parallel beamlet dose calculation via beamlet contexts in a distributed multi-GPU framework

PURPOSE

Dose calculation is one of the most computationally intensive, yet essential tasks in the treatment planning process. With the recent interest in automatic beam orientation and arc trajectory optimization techniques, there is a great need for more efficient model-based dose calculation algorithms that can accommodate hundreds to thousands of beam candidates at once. Foundational work has shown the translation of dose calculation algorithms to graphical processing units (GPUs), lending to remarkable gains in processing efficiency. But these methods provide parallelization of dose for only a single beamlet, serializing the calculation of multiple beamlets and under-utilizing the potential of modern GPUs. In this paper, the authors propose a framework enabling parallel computation of many beamlet doses using a novel beamlet context transformation and further embed this approach in a scalable network of multi-GPU computational nodes.

METHODS

The proposed context-based transformation separates beamlet-local density and TERMA into distinct beamlet contexts that independently provide sufficient data for beamlet dose calculation. Beamlet contexts are arranged in a composite context array with dosimetric isolation, and the context array is subjected to a GPU collapsed-cone convolution superposition procedure, producing the set of beamlet-specific dose distributions in a single pass. Dose from each context is converted to a sparse representation for efficient storage and retrieval during treatment plan optimization. The context radius is a new parameter permitting flexibility between the speed and fidelity of the dose calculation process. A distributed manager-worker architecture is constructed around the context-based GPU dose calculation approach supporting an arbitrary number of worker nodes and resident GPUs. Phantom experiments were executed to verify the accuracy of the context-based approach compared to Monte Carlo and a reference CPU-CCCS implementation for single beamlets and broad beams composed by addition of beamlets. Dose for representative 4π beam sets was calculated in lung and prostate cases to compare its efficiency with that of an existing beamlet-sequential GPU-CCCS implementation. Code profiling was also performed to evaluate the scalability of the framework across many networked GPUs.

RESULTS

The dosimetric accuracy of the context-based method displays <1.35% and 2.35% average error from the existing serialized CPU-CCCS algorithm and Monte Carlo simulation for beamlet-specific PDDs in water and slab phantoms, respectively. The context-based method demonstrates substantial speedup of up to two orders of magnitude over the beamlet-sequential GPU-CCCS method in the tested configurations. The context-based framework demonstrates near linear scaling in the number of distributed compute nodes and GPUs employed, indicating that it is flexible enough to meet the performance requirements of most users by simply increasing the hardware utilization.

CONCLUSIONS

The context-based approach demonstrates a new expectation of performance for beamlet-based dose calculation methods. This approach has been successful in accelerating the dose calculation process for very large-scale treatment planning problems - such as automatic 4π IMRT beam orientation and VMAT arc trajectory selection, with hundreds of thousands of beamlets - in clinically feasible timeframes. The flexibility of this framework makes it as a strong candidate for use in a variety of other very large-scale treatment planning tasks and clinical workflows.

Validation of post-treatment PET-based dosimetry software for hepatic radioembolization of Yttrium-90 microspheres

PURPOSE

Yttrium-90 (⁹⁰ Y) microsphere radioembolization enables selective internal radiotherapy for hepatic malignancies. Currently, there is no standard postdelivery imaging and dosimetry of the microsphere distribution to verify treatment. Recent studies have reported utilizing the small positron yield of ⁹⁰ Y (32 ppm) with positron emission tomography (PET) to perform treatment verification and dosimetry analysis. In this study, we validated a commercial dosimetry software, MIM SurePlan™ LiverY90 (MIM Software Inc., Cleveland, OH), for clinical use.

METHODS

A MATLAB-based algorithm for ⁹⁰ Y PET-based dosimetry was developed in-house and validated for the purpose of commissioning the commercial software. The algorithm is based on voxel S values and dosimetry formalism reported in MIRD Pamphlet 17. We validated the in-house algorithm to establish it as the ground truth by comparing results from a digital point phantom and a digital uniform cylinder to manual calculations. Once we validated our in-house MATLAB-based algorithm, we used it to perform acceptance testing and commissioning of the commercial dosimetry software, MIM SurePlan, which uses the same dosimetry formalism. A 0.4 cm/5% gamma test was performed on PET-derived dose maps from each algorithm of uniform digital and nonuniform physical phantoms filled with ⁹⁰ Y chloride solution. Average dose (D_avg ) and minimum dose to 70% (D₇₀ ) of a given volume of interest (VOI) were compared for the digital phantom, the physical phantom, and five patient cases (27 tumor VOIs), representing different clinical scenarios.

RESULTS

The gamma-pass rates were 97.26% and 97.66% for the digital and physical phantoms, respectively. The differences between D_avg and D₇₀ were 0.076% and 0.10% for the digital phantom, respectively, and <5.2% for various VOIs in the physical phantom. In the clinical cases, 96.3% of the VOIs had a difference <5% for D_avg , and 88.9% of the VOIs had a difference <5% for D₇₀ .

CONCLUSIONS

Dose calculation results from MIM SurePlan were found to be in good agreement with our in-house algorithm. This indicates that MIM SurePlan performs as it should and, hence, can be deemed accepted and commissioned for clinical use for post-implant PET-based dosimetry of ⁹⁰ Y radioembolization.

Spatial frequency performance limitations of radiation dose optimization and beam positioning

The flexibility and sophistication of modern radiotherapy treatment planning and delivery methods have advanced techniques to improve the therapeutic ratio. Contemporary dose optimization and calculation algorithms facilitate radiotherapy plans which closely conform the three-dimensional dose distribution to the target, with beam shaping devices and image guided field targeting ensuring the fidelity and accuracy of treatment delivery. Ultimately, dose distribution conformity is limited by the maximum deliverable dose gradient; shallow dose gradients challenge techniques to deliver a tumoricidal radiation dose while minimizing dose to surrounding tissue. In this work, this 'dose delivery resolution' observation is rigorously formalized for a general dose delivery model based on the superposition of dose kernel primitives. It is proven that the spatial resolution of a delivered dose is bounded by the spatial frequency content of the underlying dose kernel, which in turn defines a lower bound in the minimization of a dose optimization objective function. In addition, it is shown that this optimization is penalized by a dose deposition strategy which enforces a constant relative phase (or constant spacing) between individual radiation beams. These results are further refined to provide a direct, analytic method to estimate the dose distribution arising from the minimization of such an optimization function. The efficacy of the overall framework is demonstrated on an image guided small animal microirradiator for a set of two-dimensional hypoxia guided dose prescriptions.

3D treatment planning system-Pinnacle system

The treatment planning system is key for the success of external beam radiotherapy, directly impacting the quality of treatment plans and accuracy of dose calculation in the plans. In this article, we provided an overview of the Pinnacle treatment planning system for external beam planning, including 3-dimensional (3D) conformal plans, step-shoot intensity modulated radiotherapy (IMRT) plans, and volumetric modulated arc therapy (VMAT) plans. We discussed dose calculation algorithm and other utilities, including image fusion, plan documentation, and adaptive planning. Based on our many years of clinical experience with the system, the aim of this article is to provide readers with a summary of this particular planning system.

Performance evaluation of a collapsed cone dose calculation algorithm for HDR Ir-192 of APBI treatments

PURPOSE

Most dose calculations for HDR brachytherapy treatments are based on the AAPM-TG43 formalism. Because patient's anatomy, heterogeneities, and applicator shielding are not considered, the dose calculation based on this formalism is inaccurate in some cases. Alternatively, collapsed cone (CC) methods as well as Monte Carlo (MC) algorithms belong to the model-based dose calculation algorithms, which are expected to improve the accuracy of calculated dose distributions. In this work, the performance of a CC algorithm, ACE in Oncentra Brachy 4.5 (ACE 4.5), has been investigated by comparing the calculated dose distributions to the AAPM-TG43 and MC calculations for 10 HDR brachytherapy accelerated partial breast irradiation treatments (APBI). Comparisons were also performed with a corrected version of ACE 4.5 (ACE 4.5/corr).

METHODS

The brachytherapy source microSelectron mHDR-v2 (Elekta Brachytherapy) has been implemented in a MC environment and validated by comparing MC dose distributions simulated in a water phantom of 80 cm in diameter with dose distributions calculated with the AAPM-TG43 algorithm. Dose distributions calculated with ACE 4.5, ACE 4.5/corr, AAPM-TG43 formalism, and MC for 10 APBI patients plans have then been computed and compared using HU scaled densities. In addition, individual dose components have been computed using ACE 4.5, ACE 4.5/corr, and MC, and compared individually.

RESULTS

Local differences between MC and AAPM-TG43 calculated dose distributions in a large water phantom are < 1%. When using HUs scaled densities for the breast cancer patients, both accuracy levels of ACE 4.5 overestimate the MC calculated dose distributions for all analyzed dosimetric parameters. In the planning target volume (PTV), ACE 4.5 (ACE 4.5/corr) overestimates on average V_100%,PTV by 3% ± 1% (1% ± 1%) and D_50,PTV by 3% ± 1% (1% ± 1%) and in the organs at risk D_{1cc, skin} by 4% ± 2% (1% ± 1%), D_{0.5cc, ribs} by 4% ± 2% (0% ± 1%), and D_{1cc, heart} by 8% ± 2% (3% ± 1%) compared to MC. Comparisons of the individual dose components reveals an agreement for the primary component of < 2% local differences for both ACE 4.5 and ACE 4.5/corr. Local differences of about 40% (20%) for the first and residual scatter components where observed when using ACE 4.5 (ACE 4.5/corr). Using uniform densities for one case shows a better agreement between ACE 4.5 and MC for all dosimetric parameters considered in this work.

CONCLUSIONS

In general, on the 10 APBI patients the ACE 4.5/corr algorithm results in similar dose distributions as the commonly used AAPM-TG43 within the PTV. However, the accuracy of the ACE 4.5/corr calculated dose distribution is closer to MC than to AAPM-TG43. The differences between commercial version ACE 4.5 and MC dose distributions are mainly located in the first and residual scatter components. In ACE 4.5/corr, the changes done in the algorithm for the scatter components substantially reduce these differences.

Validation of a GPU-Based 3D dose calculator for modulated beams

A superposition/convolution GPU-accelerated dose computation algorithm (the Calculator) has been recently incorporated into commercial software. The algorithm requires validation prior to clinical use. Three photon energies were examined: conventional 6 MV and 15 MV, and 10 MV flattening filter free (10 MVFFF). For a set of IMRT and VMAT plans based on four of the five AAPM Practice Guideline 5a downloadable datasets, ion chamber (IC) measurements were performed on the water-equivalent phantoms. The average difference between the Calculator and IC was -0.3 ± 0.8% (1SD). The same plans were projected on a phantom containing a biplanar diode array. We used the forthcoming criteria for routine gamma analysis, 3% dose-error (global (G) normalization, 2 mm distance to agreement, and 10% low dose cutoff). The γ (3%G/2 mm) average passing rate was 98.9 ± 2.1%. Measurement-guided three-dimensional dose reconstruction on the patient CT dataset (excluding the Lung) resulted in a similar average agreement rate with the Calculator: 98.2 ± 2.0%. The mean γ (3%G/2 mm) passing rate comparing the Calculator to the TPS (again excluding the Lung) was 99.0 ± 1.0%. Because of the significant inhomogeneity, the Lung case was investigated separately. The calculator has an alternate heterogeneity correction mode that can change the results in the thorax for higher-energy beams (15 MV). As this correction is nonphysical and was optimized for simple slab geometries, its application leads to mixed results when compared to the TPS and independent Monte Carlo calculations, depending on the CT dataset and the plan. The Calculator vs. TPS 15 MV Guideline 5a IMRT and VMAT plans demonstrate 96.3% and 93.4% γ (3%G/2 mm) passing rates respectively. For the lower energies, which should be predominantly used in the thoracic region, the passing rates for the same plans and criteria range from 98.6 to 100%. Overall, the Calculator accuracy is sufficient for the intended use.

A review of uncertainties in radiotherapy dose reconstruction and their impacts on dose-response relationships

Proper understanding of the risk of radiation-induced late effects for patients receiving external photon beam radiotherapy requires the determination of reliable dose-response relationships. Although significant efforts have been devoted to improving dose estimates for the study of late effects, the most often questioned explanatory variable is still the dose. In this work, based on a literature review, we provide an in-depth description of the radiotherapy dose reconstruction process for the study of late effects. In particular, we focus on the identification of the main sources of dose uncertainty involved in this process and summarise their impacts on the dose-response relationship for radiotherapy late effects. We provide a number of recommendations for making progress in estimating the uncertainties in current studies of radiotherapy late effects and reducing these uncertainties in future studies.

Helical tomotherapy of spinal chordomas: French Multicentric, retrospective study of a cohort of 30 cases

PURPOSE

To evaluate the efficacy and toxicity of helical tomotherapy (HT) in the management of spine chordomas when proton therapy is unavailable or non-feasible.

METHODS AND MATERIALS

Between 2007 and 2013, 30 patients with biopsy-proven chordomas were treated by HT in five French institutions. Information regarding local control (LC), overall survival (OS), progression-free survival (PFS) and metastasis-free survival (MFS) was collected. Clinical efficacy, toxicity and treatment quality were evaluated.

RESULTS

Two-year actuarial LC, OS, PFS and MFS were 69.9%, 96.7%, 61.2% and 76.4%, respectively. HT treatments were well tolerated and no Grade 4-5 toxicities were observed. HT permitted the delivery of a mean dose of 68 Gy while respecting organ at risk (OAR) dose constraints, in particular in the spinal cord and cauda equina.

CONCLUSIONS

This multicentric, retrospective study demonstrated the feasibility of HT in the treatment of spine chordomas, in the absence of hadron therapy.

Experimental Validation of Monte Carlo Simulations Based on a Virtual Source Model for TomoTherapy in a RANDO Phantom

A virtual source model for Monte Carlo simulations of helical TomoTherapy has been developed previously by the authors. The purpose of this work is to perform experiments in an anthropomorphic (RANDO) phantom with the same order of complexity as in clinical treatments to validate the virtual source model to be used for quality assurance secondary check on TomoTherapy patient planning dose. Helical TomoTherapy involves complex delivery pattern with irregular beam apertures and couch movement during irradiation. Monte Carlo simulation, as the most accurate dose algorithm, is desirable in radiation dosimetry. Current Monte Carlo simulations for helical TomoTherapy adopt the full Monte Carlo model, which includes detailed modeling of individual machine component, and thus, large phase space files are required at different scoring planes. As an alternative approach, we developed a virtual source model without using the large phase space files for the patient dose calculations previously. In this work, we apply the simulation system to recompute the patient doses, which were generated by the treatment planning system in an anthropomorphic phantom to mimic the real patient treatments. We performed thermoluminescence dosimeter point dose and film measurements to compare with Monte Carlo results. Thermoluminescence dosimeter measurements show that the relative difference in both Monte Carlo and treatment planning system is within 3%, with the largest difference less than 5% for both the test plans. The film measurements demonstrated 85.7% and 98.4% passing rate using the 3 mm/3% acceptance criterion for the head and neck and lung cases, respectively. Over 95% passing rate is achieved if 4 mm/4% criterion is applied. For the dose–volume histograms, very good agreement is obtained between the Monte Carlo and treatment planning system method for both cases. The experimental results demonstrate that the virtual source model Monte Carlo system can be a viable option for the accurate dose calculation of helical TomoTherapy.

Performance Evaluation of Algorithms in Lung IMRT: A comparison of Monte Carlo, Pencil Beam, Superposition, Fast Superposition and Convolution Algorithms

BACKGROUND

Inclusion of inhomogeneity corrections in intensity modulated small fields always makes conformal irradiation of lung tumor very complicated in accurate dose delivery.

OBJECTIVE

In the present study, the performance of five algorithms via Monte Carlo, Pencil Beam, Convolution, Fast Superposition and Superposition were evaluated in lung cancer Intensity Modulated Radiotherapy planning.

MATERIALS AND METHODS

Treatment plans for ten lung cancer patients previously planned on Monte Carlo algorithm were re-planned using same treatment planning indices (gantry angel, rank, power etc.) in other four algorithms.

RESULTS

The values of radiotherapy planning parameters such as Mean dose, volume of 95% isodose line, Conformity Index, Homogeneity Index for target, Maximum dose, Mean dose; %Volume receiving 20Gy or more by contralateral lung; % volume receiving 30 Gy or more; % volume receiving 25 Gy or more, Mean dose received by heart; %volume receiving 35Gy or more; %volume receiving 50Gy or more, Mean dose to Easophagous; % Volume receiving 45Gy or more, Maximum dose received by Spinal cord and Total monitor unit, Volume of 50 % isodose lines were recorded for all ten patients. Performance of different algorithms was also evaluated statistically.

CONCLUSION

MC and PB algorithms found better as for tumor coverage, dose distribution homogeneity in Planning Target Volume and minimal dose to organ at risks are concerned. Superposition algorithms found to be better than convolution and fast superposition. In the case of tumors located centrally, it is recommended to use Monte Carlo algorithms for the optimal use of radiotherapy.

Technical Note: Dose gradients and prescription isodose in orthovoltage stereotactic radiosurgery

PURPOSE

The purpose of this work is to examine the trade-off between prescription isodose and dose gradients in orthovoltage stereotactic radiosurgery.

METHODS

Point energy deposition kernels (EDKs) describing photon and electron transport were calculated using Monte Carlo methods. EDKs were generated from 10  to 250 keV, in 10 keV increments. The EDKs were converted to pencil beam kernels and used to calculate dose profiles through isocenter from a 4π isotropic delivery from all angles of circularly collimated beams. Monoenergetic beams and an orthovoltage polyenergetic spectrum were analyzed. The dose gradient index (DGI) is the ratio of the 50% prescription isodose volume to the 100% prescription isodose volume and represents a metric by which dose gradients in stereotactic radiosurgery (SRS) may be evaluated.

RESULTS

Using the 4π dose profiles calculated using pencil beam kernels, the relationship between DGI and prescription isodose was examined for circular cones ranging from 4 to 18 mm in diameter and monoenergetic photon beams with energies ranging from 20 to 250 keV. Values were found to exist for prescription isodose that optimize DGI.

CONCLUSIONS

The relationship between DGI and prescription isodose was found to be dependent on both field size and energy. Examining this trade-off is an important consideration for designing optimal SRS systems.

Pinnacle3 modeling and end-to-end dosimetric testing of a Versa HD linear accelerator with the Agility head and flattening filter-free modes

The Elekta Versa HD incorporates a variety of upgrades to the line of Elekta linear accelerators, primarily including the Agility head and flattening filter‐free (FFF) photon beam delivery. The completely distinct dosimetric output of the head from its predecessors, combined with the FFF beams, requires a new investigation of modeling in treatment planning systems. A model was created in Pinnacle³ v9.8 with the commissioned beam data. A phantom consisting of several plastic water and Styrofoam slabs was scanned and imported into Pinnacle³, where beams of different field sizes, source‐to‐surface distances (SSDs), wedges, and gantry angles were devised. Beams included all of the available photon energies (6, 10, 18, 6 FFF, and 10 FFF MV), as well as the four electron energies commissioned for clinical use (6, 9, 12, and 15 MeV). The plans were verified at calculation points by measurement with a calibrated ionization chamber. Homogeneous and heterogeneous point‐dose measurements agreed within 2% relative to maximum dose for all photon and electron beams. AP photon open field measurements along the central axis at 100 cm SSD passed within 1%. In addition, IMRT testing was also performed with three standard plans (step and shoot IMRT, as well as a small‐ and large‐field VMAT plan). The IMRT plans were delivered on the Delta⁴ IMRT QA phantom, for which a gamma passing rate was >99.5% for all plans with a 3% dose deviation, 3 mm distance‐to‐agreement, and 10% dose threshold. The IMRT QA results for the first 23 patients yielded gamma passing rates of 97.4%±2.3%. Such testing ensures confidence in the ability of Pinnacle³ to model photon and electron beams with the Agility head.

PACS numbers: 87.55.D, 87.56.bd

A nonvoxel-based dose convolution/superposition algorithm optimized for scalable GPU architectures

PURPOSE

Real-time adaptive planning and treatment has been infeasible due in part to its high computational complexity. There have been many recent efforts to utilize graphics processing units (GPUs) to accelerate the computational performance and dose accuracy in radiation therapy. Data structure and memory access patterns are the key GPU factors that determine the computational performance and accuracy. In this paper, the authors present a nonvoxel-based (NVB) approach to maximize computational and memory access efficiency and throughput on the GPU.

METHODS

The proposed algorithm employs a ray-tracing mechanism to restructure the 3D data sets computed from the CT anatomy into a nonvoxel-based framework. In a process that takes only a few milliseconds of computing time, the algorithm restructured the data sets by ray-tracing through precalculated CT volumes to realign the coordinate system along the convolution direction, as defined by zenithal and azimuthal angles. During the ray-tracing step, the data were resampled according to radial sampling and parallel ray-spacing parameters making the algorithm independent of the original CT resolution. The nonvoxel-based algorithm presented in this paper also demonstrated a trade-off in computational performance and dose accuracy for different coordinate system configurations. In order to find the best balance between the computed speedup and the accuracy, the authors employed an exhaustive parameter search on all sampling parameters that defined the coordinate system configuration: zenithal, azimuthal, and radial sampling of the convolution algorithm, as well as the parallel ray spacing during ray tracing. The angular sampling parameters were varied between 4 and 48 discrete angles, while both radial sampling and parallel ray spacing were varied from 0.5 to 10 mm. The gamma distribution analysis method (γ) was used to compare the dose distributions using 2% and 2 mm dose difference and distance-to-agreement criteria, respectively. Accuracy was investigated using three distinct phantoms with varied geometries and heterogeneities and on a series of 14 segmented lung CT data sets. Performance gains were calculated using three 256 mm cube homogenous water phantoms, with isotropic voxel dimensions of 1, 2, and 4 mm.

RESULTS

The nonvoxel-based GPU algorithm was independent of the data size and provided significant computational gains over the CPU algorithm for large CT data sizes. The parameter search analysis also showed that the ray combination of 8 zenithal and 8 azimuthal angles along with 1 mm radial sampling and 2 mm parallel ray spacing maintained dose accuracy with greater than 99% of voxels passing the γ test. Combining the acceleration obtained from GPU parallelization with the sampling optimization, the authors achieved a total performance improvement factor of >175 000 when compared to our voxel-based ground truth CPU benchmark and a factor of 20 compared with a voxel-based GPU dose convolution method.

CONCLUSIONS

The nonvoxel-based convolution method yielded substantial performance improvements over a generic GPU implementation, while maintaining accuracy as compared to a CPU computed ground truth dose distribution. Such an algorithm can be a key contribution toward developing tools for adaptive radiation therapy systems.

Dosimetric validation and clinical implementation of two 3D dose verification systems for quality assurance in volumetric-modulated arc therapy techniques

A pretreatment quality assurance program for volumetric techniques should include redundant calculations and measurement-based verifications. The patient-specific quality assurance process must be based in clinically relevant metrics. The aim of this study was to show the commission, clinical implementation, and comparison of two systems that allow performing a 3D redundant dose calculation. In addition, one of them is capable of reconstructing the dose on patient anatomy from measurements taken with a 2D ion chamber array. Both systems were compared in terms of reference calibration data (absolute dose, output factors, percentage depth-dose curves, and profiles). Results were in good agreement for absolute dose values (discrepancies were below 0.5%) and output factors (mean differences were below 1%). Maximum mean discrepancies were located between 10 and 20 cm of depth for PDDs (-2.7%) and in the penumbra region for profiles (mean DTA of 1.5 mm). Validation of the systems was performed by comparing point-dose measurements with values obtained by the two systems for static, dynamic fields from AAPM TG-119 report, and 12 real VMAT plans for different anatomical sites (differences better than 1.2%). Comparisons between measurements taken with a 2D ion chamber array and results obtained by both systems for real VMAT plans were also performed (mean global gamma passing rates better than 87.0% and 97.9% for the 2%/2 mm and 3%/3 mm criteria). Clinical implementation of the systems was evaluated by comparing dose-volume parameters for all TG-119 tests and real VMAT plans with TPS values (mean differences were below 1%). In addition, comparisons between dose distributions calculated by TPS and those extracted by the two systems for real VMAT plans were also performed (mean global gamma passing rates better than 86.0% and 93.0% for the 2%/2 mm and 3%/ 3 mm criteria). The clinical use of both systems was successfully evaluated.

Investigation of various energy deposition kernel refinements for the convolution∕superposition method

PURPOSE

Several simplifications used in clinical implementations of the convolution∕superposition (C∕S) method, specifically, density scaling of water kernels for heterogeneous media and use of a single polyenergetic kernel, lead to dose calculation inaccuracies. Although these weaknesses of the C∕S method are known, it is not well known which of these simplifications has the largest effect on dose calculation accuracy in clinical situations. The purpose of this study was to generate and characterize high-resolution, polyenergetic, and material-specific energy deposition kernels (EDKs), as well as to investigate the dosimetric impact of implementing spatially variant polyenergetic and material-specific kernels in a collapsed cone C∕S algorithm.

METHODS

High-resolution, monoenergetic water EDKs and various material-specific EDKs were simulated using the EGSnrc Monte Carlo code. Polyenergetic kernels, reflecting the primary spectrum of a clinical 6 MV photon beam at different locations in a water phantom, were calculated for different depths, field sizes, and off-axis distances. To investigate the dosimetric impact of implementing spatially variant polyenergetic kernels, depth dose curves in water were calculated using two different implementations of the collapsed cone C∕S method. The first method uses a single polyenergetic kernel, while the second method fully takes into account spectral changes in the convolution calculation. To investigate the dosimetric impact of implementing material-specific kernels, depth dose curves were calculated for a simplified titanium implant geometry using both a traditional C∕S implementation that performs density scaling of water kernels and a novel implementation using material-specific kernels.

RESULTS

For our high-resolution kernels, we found good agreement with the Mackie et al. kernels, with some differences near the interaction site for low photon energies (<500 keV). For our spatially variant polyenergetic kernels, we found that depth was the most dominant factor affecting the pattern of energy deposition; however, the effects of field size and off-axis distance were not negligible. For the material-specific kernels, we found that as the density of the material increased, more energy was deposited laterally by charged particles, as opposed to in the forward direction. Thus, density scaling of water kernels becomes a worse approximation as the density and the effective atomic number of the material differ more from water. Implementation of spatially variant, polyenergetic kernels increased the percent depth dose value at 25 cm depth by 2.1%-5.8% depending on the field size, while implementation of titanium kernels gave 4.9% higher dose upstream of the metal cavity (i.e., higher backscatter dose) and 8.2% lower dose downstream of the cavity.

CONCLUSIONS

Of the various kernel refinements investigated, inclusion of depth-dependent and metal-specific kernels into the C∕S method has the greatest potential to improve dose calculation accuracy. Implementation of spatially variant polyenergetic kernels resulted in a harder depth dose curve and thus has the potential to affect beam modeling parameters obtained in the commissioning process. For metal implants, the C∕S algorithms generally underestimate the dose upstream and overestimate the dose downstream of the implant. Implementation of a metal-specific kernel mitigated both of these errors.

Optical dosimetry of radiotherapy beams using Cherenkov radiation: the relationship between light emission and dose

Recent studies have proposed that light emitted by the Cherenkov effect may be used for a number of radiation therapy dosimetry applications. There is a correlation between the captured light and expected dose under certain conditions, yet discrepancies have also been observed and a complete examination of the theoretical differences has not been done. In this study, a fundamental comparison between the Cherenkov emission and absorbed dose was explored for x-ray photons, electrons, and protons using both a theoretical and Monte Carlo-based analysis. Based on the findings of where dose correlates with Cherenkov emission, it was concluded that for x-ray photons the light emission would be optimally suited for narrow beam stereotactic radiation therapy and surgery validation studies, for verification of dynamic intensity-modulated and volumetric modulated arc therapy treatment plans in water tanks, near monoenergetic sources (e.g., Co-60 and brachy therapy sources) and also for entrance and exit surface imaging dosimetry of both narrow and broad beams. For electron use, Cherenkov emission was found to be only suitable for surface dosimetry applications. Finally, for proton dosimetry, there exists a fundamental lack of Cherenkov emission at the Bragg peak, making the technique of little use, although post-irradiation detection of light emission from radioisotopes could prove to be useful.

Independent absorbed-dose calculation using the Monte Carlo algorithm in volumetric modulated arc therapy

Purpose

To report the result of independent absorbed-dose calculations based on a Monte Carlo (MC) algorithm in volumetric modulated arc therapy (VMAT) for various treatment sites.

Methods and materials

All treatment plans were created by the superposition/convolution (SC) algorithm of SmartArc (Pinnacle V9.2, Philips). The beam information was converted into the format of the Monaco V3.3 (Elekta), which uses the X-ray voxel-based MC (XVMC) algorithm. The dose distribution was independently recalculated in the Monaco. The dose for the planning target volume (PTV) and the organ at risk (OAR) were analyzed via comparisons with those of the treatment plan.

Before performing an independent absorbed-dose calculation, the validation was conducted via irradiation from 3 different gantry angles with a 10- × 10-cm² field. For the independent absorbed-dose calculation, 15 patients with cancer (prostate, 5; lung, 5; head and neck, 3; rectal, 1; and esophageal, 1) who were treated with single-arc VMAT were selected. To classify the cause of the dose difference between the Pinnacle and Monaco TPSs, their calculations were also compared with the measurement data.

Result

In validation, the dose in Pinnacle agreed with that in Monaco within 1.5%. The agreement in VMAT calculations between Pinnacle and Monaco using phantoms was exceptional; at the isocenter, the difference was less than 1.5% for all the patients. For independent absorbed-dose calculations, the agreement was also extremely good. For the mean dose for the PTV in particular, the agreement was within 2.0% in all the patients; specifically, no large difference was observed for high-dose regions. Conversely, a significant difference was observed in the mean dose for the OAR. For patients with prostate cancer, the mean rectal dose calculated in Monaco was significantly smaller than that calculated in Pinnacle.

Conclusions

There was no remarkable difference between the SC and XVMC calculations in the high-dose regions. The difference observed in the low-dose regions may have arisen from various causes such as the intrinsic dose deviation in the MC calculation, modeling accuracy, and CT-to-density table used in each planning system It is useful to perform independent absorbed-dose calculations with the MC algorithm in intensity-modulated radiation therapy commissioning.

Validation of the Pinnacle³ photon convolution-superposition algorithm applied to fast neutron beams

We evaluate a photon convolution‐superposition algorithm used to model a fast neutron therapy beam in a commercial treatment planning system (TPS). The neutron beam modeled was the Clinical Neutron Therapy System (CNTS) fast neutron beam produced by 50 MeV protons on a Be target at our facility, and we implemented the Pinnacle3 dose calculation model for computing neutron doses. Measured neutron data were acquired by an IC30 ion chamber flowing 5 cc/min of tissue equivalent gas. Output factors and profile scans for open and wedged fields were measured according to the Pinnacle physics reference guide recommendations for photon beams in a Wellhofer water tank scanning system. Following the construction of a neutron beam model, computed doses were then generated using 100 monitor units (MUs) beams incident on a water‐equivalent phantom for open and wedged square fields, as well as multileaf collimator (MLC)‐shaped irregular fields. We compared Pinnacle dose profiles, central axis doses, and off‐axis doses (in irregular fields) with 1) doses computed using the Prism treatment planning system, and 2) doses measured in a water phantom and having matching geometry to the computation setup. We found that the Pinnacle photon model may be used to model most of the important dosimetric features of the CNTS fast neutron beam. Pinnacle‐calculated dose points among open and wedged square fields exhibit dose differences within 3.9 cGy of both Prism and measured doses along the central axis, and within 5 cGy difference of measurement in the penumbra region. Pinnacle dose point calculations using irregular treatment type fields showed a dose difference up to 9 cGy from measured dose points, although most points of comparison were below 5 cGy. Comparisons of dose points that were chosen from cases planned in both Pinnacle and Prism show an average dose difference less than 0.6%, except in certain fields which incorporate both wedges and heavy blocking of the central axis. All clinical cases planned in both Prism and Pinnacle were found to be comparable in terms of dose‐volume histograms and spatial dose distribution following review by the treating clinicians. Variations were considered minor and within clinically acceptable limits by the treating clinicians. The Pinnacle TPS has sufficient computational modeling ability to adequately produce a viable neutron model for clinical use in treatment planning.

PACS numbers: 87.53 Bn, 28.20.Pr, 87.53.Bn

Improved normal tissue sparing in head and neck radiotherapy using biological cost function based-IMRT

Intensity-modulated radiotherapy (IMRT) has reduced the impact of acute and late toxicities associated with head and neck radiotherapy. Treatment planning system (TPS) advances in biological cost function based optimization (BBO) and improved segmentation techniques have increased organ at risk (OAR) sparing compared to conventional dose-based optimization (DBO). A planning study was undertaken to compare OAR avoidance in DBO and BBO treatment planning. Simultaneous integrated boost treatment plans were produced for 10 head and neck patients using both planning systems. Plans were compared for tar get coverage and OAR avoidance. Comparisons were made using the BBO TPS Monte Carlo dose engine to eliminate differences due to inherent algorithms. Target coverage (V95%) was maintained for both solutions. BBO produced lower OAR doses, with statistically significant improvement to left (12.3%, p = 0.005) and right parotid mean dose (16.9%, p = 0.004), larynx V50 Gy (71.0%, p = 0.005), spinal cord (21.9%, p < 0.001) and brain stem dose maximums (31.5%, p = 0.002). This study observed improved OAR avoidance with BBO planning. Further investigations will be undertaken to review any clinical benefit of this improved planned dosimetry.

Evaluation of heterogeneity dose distributions for Stereotactic Radiotherapy (SRT): comparison of commercially available Monte Carlo dose calculation with other algorithms

Background

The purpose of this study was to compare dose distributions from three different algorithms with the x-ray Voxel Monte Carlo (XVMC) calculations, in actual computed tomography (CT) scans for use in stereotactic radiotherapy (SRT) of small lung cancers.

Methods

Slow CT scan of 20 patients was performed and the internal target volume (ITV) was delineated on Pinnacle³. All plans were first calculated with a scatter homogeneous mode (SHM) which is compatible with Clarkson algorithm using Pinnacle³ treatment planning system (TPS). The planned dose was 48 Gy in 4 fractions. In a second step, the CT images, structures and beam data were exported to other treatment planning systems (TPSs). Collapsed cone convolution (CCC) from Pinnacle³, superposition (SP) from XiO, and XVMC from Monaco were used for recalculating. The dose distributions and the Dose Volume Histograms (DVHs) were compared with each other.

Results

The phantom test revealed that all algorithms could reproduce the measured data within 1% except for the SHM with inhomogeneous phantom. For the patient study, the SHM greatly overestimated the isocenter (IC) doses and the minimal dose received by 95% of the PTV (PTV95) compared to XVMC. The differences in mean doses were 2.96 Gy (6.17%) for IC and 5.02 Gy (11.18%) for PTV95. The DVH's and dose distributions with CCC and SP were in agreement with those obtained by XVMC. The average differences in IC doses between CCC and XVMC, and SP and XVMC were -1.14% (p = 0.17), and -2.67% (p = 0.0036), respectively.

Conclusions

Our work clearly confirms that the actual practice of relying solely on a Clarkson algorithm may be inappropriate for SRT planning. Meanwhile, CCC and SP were close to XVMC simulations and actual dose distributions obtained in lung SRT.

Multisource modeling of flattening filter free (FFF) beam and the optimization of model parameters

PURPOSE

With the introduction of flattening filter free (FFF) linear accelerators to radiation oncology, new analytical source models for a FFF beam applicable to current treatment planning systems is needed. In this work, a multisource model for the FFF beam and the optimization of involved model parameters were designed.

METHODS

The model is based on a previous three source model proposed by Yang et al. ["A three-source model for the calculation of head scatter factors," Med. Phys. 29, 2024-2033 (2002)]. An off axis ratio (OAR) of photon fluence was introduced to the primary source term to generate cone shaped profiles. The parameters of the source model were determined from measured head scatter factors using a line search optimization technique. The OAR of the photon fluence was determined from a measured dose profile of a 40 x 40 cm2 field size with the same optimization technique, but a new method to acquire gradient terms for OARs was developed to enhance the speed of the optimization process. The improved model was validated with measured dose profiles from 3 x 3 to 40 x 40 cm2 field sizes at 6 and 10 MV from a TrueBeam STx linear accelerator. Furthermore, planar dose distributions for clinically used radiation fields were also calculated and compared to measurements using a 2D array detector using the gamma index method.

RESULTS

All dose values for the calculated profiles agreed with the measured dose profiles within 0.5% at 6 and 10 MV beams, except for some low dose regions for larger field sizes. A slight overestimation was seen in the lower penumbra region near the field edge for the large field sizes by 1%-4%. The planar dose calculations showed comparable passing rates (> 98%) when the criterion of the gamma index method was selected to be 3%/3 mm.

CONCLUSIONS

The developed source model showed good agreements between measured and calculated dose distributions. The model is easily applicable to any other linear accelerator using FFF beams as the required data include only the measured PDD, dose profiles, and output factors for various field sizes, which are easily acquired during conventional beam commissioning process.

GPU-accelerated Monte Carlo convolution/superposition implementation for dose calculation

PURPOSE

Dose calculation is a key component in radiation treatment planning systems. Its performance and accuracy are crucial to the quality of treatment plans as emerging advanced radiation therapy technologies are exerting ever tighter constraints on dose calculation. A common practice is to choose either a deterministic method such as the convolution/superposition (CS) method for speed or a Monte Carlo (MC) method for accuracy. The goal of this work is to boost the performance of a hybrid Monte Carlo convolution/superposition (MCCS) method by devising a graphics processing unit (GPU) implementation so as to make the method practical for day-to-day usage.

METHODS

Although the MCCS algorithm combines the merits of MC fluence generation and CS fluence transport, it is still not fast enough to be used as a day-to-day planning tool. To alleviate the speed issue of MC algorithms, the authors adopted MCCS as their target method and implemented a GPU-based version. In order to fully utilize the GPU computing power, the MCCS algorithm is modified to match the GPU hardware architecture. The performance of the authors' GPU-based implementation on an Nvidia GTX260 card is compared to a multithreaded software implementation on a quad-core system.

RESULTS

A speedup in the range of 6.7-11.4x is observed for the clinical cases used. The less than 2% statistical fluctuation also indicates that the accuracy of the authors' GPU-based implementation is in good agreement with the results from the quad-core CPU implementation.

CONCLUSIONS

This work shows that GPU is a feasible and cost-efficient solution compared to other alternatives such as using cluster machines or field-programmable gate arrays for satisfying the increasing demands on computation speed and accuracy of dose calculation. But there are also inherent limitations of using GPU for accelerating MC-type applications, which are also analyzed in detail in this article.

Postlumpectomy focal brachytherapy for simultaneous treatment of surgical cavity and draining lymph nodes

PURPOSE

The primary objective was to investigate a novel focal brachytherapy technique using lipid nanoparticle (liposome)-carried β-emitting radionuclides (rhenium-186 [(186)Re]/rhenium-188 [(188)Re]) to simultaneously treat the postlumpectomy surgical cavity and draining lymph nodes.

METHODS AND MATERIALS

Cumulative activity distributions in the lumpectomy cavity and lymph nodes were extrapolated from small animal imaging and human lymphoscintigraphy data. Absorbed dose calculations were performed for lumpectomy cavities with spherical and ellipsoidal shapes and lymph nodes within human subjects by use of the dose point kernel convolution method.

RESULTS

Dose calculations showed that therapeutic dose levels within the lumpectomy cavity wall can cover 2- and 5-mm depths for (186)Re and (188)Re liposomes, respectively. The absorbed doses at 1 cm sharply decreased to only 1.3% to 3.7% of the doses at 2 mm for (186)Re liposomes and 5 mm for (188)Re liposomes. Concurrently, the draining sentinel lymph nodes would receive a high focal therapeutic absorbed dose, whereas the average dose to 1 cm of surrounding tissue received less than 1% of that within the nodes.

CONCLUSIONS

Focal brachytherapy by use of (186)Re/(188)Re liposomes was theoretically shown to be capable of simultaneously treating the lumpectomy cavity wall and draining sentinel lymph nodes with high absorbed doses while significantly lowering dose to surrounding healthy tissue. In turn, this allows for dose escalation to regions of higher probability of containing residual tumor cells after lumpectomy while reducing normal tissue complications.

A GPU-based framework for modeling real-time 3D lung tumor conformal dosimetry with subject-specific lung tumor motion

In this paper, we present a graphics processing unit (GPU)-based simulation framework to calculate the delivered dose to a 3D moving lung tumor and its surrounding normal tissues, which are undergoing subject-specific lung deformations. The GPU-based simulation framework models the motion of the 3D volumetric lung tumor and its surrounding tissues, simulates the dose delivery using the dose extracted from a treatment plan using Pinnacle Treatment Planning System, Phillips, for one of the 3DCTs of the 4DCT and predicts the amount and location of radiation doses deposited inside the lung. The 4DCT lung datasets were registered with each other using a modified optical flow algorithm. The motion of the tumor and the motion of the surrounding tissues were simulated by measuring the changes in lung volume during the radiotherapy treatment using spirometry. The real-time dose delivered to the tumor for each beam is generated by summing the dose delivered to the target volume at each increase in lung volume during the beam delivery time period. The simulation results showed the real-time capability of the framework at 20 discrete tumor motion steps per breath, which is higher than the number of 4DCT steps (approximately 12) reconstructed during multiple breathing cycles.

Towards real-time radiation therapy: GPU accelerated superposition/convolution

We demonstrate the use of highly parallel graphics processing units (GPUs) to accelerate the superposition/convolution (S/C) algorithm to interactive rates while reducing the number of approximations. S/C first transports the incident fluence to compute the total energy released per unit mass (TERMA) grid. Dose is then calculated by superimposing the dose deposition kernel at each point in the TERMA grid and summing the contributions to the surrounding voxels. The TERMA algorithm was enhanced with physically correct multi-spectral attenuation and a novel inverse formulation for increased performance, accuracy and simplicity. Dose deposition utilized a tilted poly-energetic inverse cumulative-cumulative kernel, with the novel option of using volumetric mip-maps to approximate solid angle ray casting. Exact radiological path ray casting decreased discretization errors. We achieved a speedup of 34x-98x over a highly optimized CPU implementation.