Accurate Dose Calculation Algorithm Research Articles

Development of an algorithm for proton dose calculation in magnetic fields.

The advantages of proton therapy can be further enhanced with online magnetic resonance imaging (MRI) guidance. One of the challenges in the realization of MRI-guided proton therapy (MRPT) is accurately calculating the radiation dose in the presence of magnetic fields. This study aims to develop an efficient and accurate proton dose calculation algorithm adapted to the presence of magnetic fields. An analytical-numerical radiation dose calculation algorithm, Proton and Ion Dose Engine (PRIDE), was developed. The algorithm combines the pencil beam algorithm (PBA) with a novel iterative voxel-based ray-tracing algorithm. The new ray-tracing method uses fewer assumptions and ensures broader applicability for proton beam trajectory prediction in magnetic fields, and has been compared to Wolf's method and Schellhammer's method. The accuracy of PRIDE algorithm was validated on three phantoms and two practical plans (one single-field water plan and one prostate tumor plan) in different magnetic field strengths up to 3.0 T. The validation was performed by comparing the results against the Monte Carlo (MC) simulations, using the global gamma index criteria of 2%/2mm and 3%/3mm with a 10% threshold. PRIDE showed good agreement with MC in homogeneous and slab heterogeneous phantom, achieving gamma passing rates (%GPs) above 99% for 2%/2mm criteria when magnetic field strength is not greater than 1.5 T. Although the agreement decreased for scenarios involving high proton energy (240 MeV) and strong magnetic field (3.0 T), the 2%/2mm %GPs still remained above 98%. In lateral heterogeneous phantom, the accuracy of PRIDE decreased due to the PBA's limitation. For the two practical plans in different magnetic fields, %GPs exceeded 98% and 99% for 2%/2mm and 3%/3mm criteria, respectively. PRIDE can perform efficient and accurate proton dose calculation in magnetic fields up to 3.0 T, and is expected to work as a useful tool for proton dose calculation in MRPT.

2772: Fast and accurate dose calculation algorithm based on multi-level U-Net for IMRT

2772: Fast and accurate dose calculation algorithm based on multi-level U-Net for IMRT

Technical note: A fast and accurate analytical dose calculation algorithm for 125 I seed-loaded stent applications.

The safety and clinical efficacy of 125 I seed-loaded stent for the treatment of portal vein tumor thrombosis (PVTT) have been shown. Accurate and fast dose calculation of the 125 I seeds with the presence of the stent is necessary for the plan optimization and evaluation. However, the dosimetric characteristics of the seed-loaded stents remain unclear and there is no fast dose calculation technique available. This paper aims to explore a fast and accurate analytical dose calculation method based on Monte Carlo (MC) dose calculation, which takes into account the effect of stent and tissue inhomogeneity. A detailed model of the seed-loaded stent was developed using 3D modeling software and subsequently used in MC simulations to calculate the dose distribution around the stent. The dose perturbation caused by the presence of the stent was analyzed, and dose perturbation kernels (DPKs) were derived and stored for future use. Then, the dose calculation method from AAPM TG-43 was adapted by integrating the DPK and appropriate inhomogeneity correction factors (ICF) to calculate dose distributions analytically. To validate the proposed method, several comparisons were performed with other methods in water phantom and voxelized CT phantoms for three patients. The stent has a considerable dosimetric effect reducing the dose up to 47.2% for single-seed stent and 11.9%-16.1% for 16-seed stent. In a water phantom, dose distributions from MC simulations and TG-43-DP-ICF showed a good agreement with the relative error less than 3.3%. In voxelized CT phantoms, taking MC results as the reference, the relative errors of TG-43 method can be up to 33%, while those of TG-43-DP-ICF method were less than 5%. For a dose matrix with 256 × 256 × 46 grid (corresponding to a phantom of 17.2 × 17.2 × 11.5cm3 ) for 16-seed-loaded stent, it only takes 17s for TG-43-DP-ICF to compute, compared to 25h for the full MC calculation. The combination of DPK and inhomogeneity corrections is an effective approach to handle both the presence of stent and tissue heterogeneity. Exhibiting good agreement with MC calculation and computational efficiency, the proposed TG-43-DP-ICF method is adequate for dose evaluation and optimization in seed-loaded stent implantation treatment planning.

Dosimetric Comparison and Clinical Implications of Dose Calculation Algorithms for Stereotactic Body Radiation Therapy in Spine Metastases

Dose calculation algorithms for radiation oncology plans differ in their ability to handle patient specific anatomy, tissue heterogeneity, and artificial implants, which is especially crucial in spine stereotactic body radiation therapy (SBRT) as high ablative dose radiation is directed at a heterogeneous volume, often with metal. In addition, due to the sharp dose gradient required, a slight difference in dose calculation can result in overdosing of the cord or underdosing of the tumor. Better understanding of the different calculation algorithms are needed in order to apply existing historical dose objectives to the modern era. Evaluate the dosimetric differences in three commonly used calculation algorithms—collapsed cone convolution (CCC), anisotropic analytic algorithm (AAA), and Acuros XB (AXB)—for SBRT in spine tumors with and without metal implants. We retrospectively analyzed 27 consecutive patients who have received SBRT for spine metastases, eight of whom had metal hardware within the target. Fourteen plans initially calculated using CCC were recalculated with AAA and AXB while preserving the total monitor units. An additional 13 plans previously calculated with AXB were recalculated using AAA in a similar fashion. Target coverage and the maximum dose for the spinal cord were examined. Doses are expressed relative to AXB as the standard. The maximum dose to the spinal cord was overestimated with AAA by +3.4% [95% CI: -2.8%, 9.6%] and CCC by +0.34% [-7.98%, 8.66%] compared to AXB. Patients with metal hardware in the target volume showed a larger degree of overestimation: AAA by +4.1% with metal vs. +3.0% without; CCC by +1.2% with metal vs. -0.3% underestimation for those without. The maximum dose to the target was overestimated with AAA by +1.2% [-3.4%, 5.9%], whereas CCC underestimated by -4.9% [-11.0%, 1.3%]. However, for patients with metal in the target, AAA overestimated by +0.7% compared to +1.5% for those without. CCC underestimation was much larger with metal, -7.0% compared to -3.2% without metal. Both algorithms overestimated the dose covering 95% of the target volume: AAA by +2.9% [-.8%, 6.6%] and CCC by +3.9% [-1.8%, 9.6%]. The degree of overestimation was larger in patients with metal; AAA: +4.9% overestimation with metal vs. +2.1% without; CCC: +6.3% overestimation with metal vs. +2.2% without. As more accurate dose calculation algorithms (AXB) are implemented clinically, the dosimetric review of target and normal organs need to be re-examined. Historical algorithms overestimate the minimum target coverage compared with AXB, indicating potential under-dosing of those plans generated with CCC and AAA. Additionally, spinal cord tolerances established using historically less accurate algorithms need to be validated in plans calculated with AXB. The differences in the calculated doses are exaggerated in patients with metal implants, warranting additional attention.

Boosting radiotherapy dose calculation accuracy with deep learning.

In radiotherapy, a trade‐off exists between computational workload/speed and dose calculation accuracy. Calculation methods like pencil‐beam convolution can be much faster than Monte‐Carlo methods, but less accurate. The dose difference, mostly caused by inhomogeneities and electronic disequilibrium, is highly correlated with the dose distribution and the underlying anatomical tissue density. We hypothesize that a conversion scheme can be established to boost low‐accuracy doses to high‐accuracy, using intensity information obtained from computed tomography (CT) images. A deep learning‐driven framework was developed to test the hypothesis by converting between two commercially available dose calculation methods: Anisotropic analytic algorithm (AAA) and Acuros XB (AXB). A hierarchically dense U‐Net model was developed to boost the accuracy of AAA dose toward the AXB level. The network contained multiple layers of varying feature sizes to learn their dose differences, in relationship to CT, both locally and globally. Anisotropic analytic algorithm and AXB doses were calculated in pairs for 120 lung radiotherapy plans covering various treatment techniques, beam energies, tumor locations, and dose levels. For each case, the CT and the AAA dose were used as the input and the AXB dose as the “ground‐truth” output, to train and test the model. The mean squared errors (MSEs) and gamma passing rates (2 mm/2% & 1 mm/1%) were calculated between the boosted AAA doses and the “ground‐truth” AXB doses. The boosted AAA doses demonstrated substantially improved match to the “ground‐truth” AXB doses, with average (± s.d.) gamma passing rate (1 mm/1%) 97.6% (±2.4%) compared to 87.8% (±9.0%) of the original AAA doses. The corresponding average MSE was 0.11(±0.05) vs 0.31(±0.21). Deep learning is able to capture the differences between dose calculation algorithms to boost the low‐accuracy algorithms. By combining a less accurate dose calculation algorithm with a trained deep learning model, dose calculation can potentially achieve both high accuracy and efficiency.

Treatment planning for proton therapy: what is needed in the next 10 years?

Treatment planning is the process where the prescription of the radiation oncologist is translated into a deliverable treatment. With the complexity of contemporary radiotherapy, treatment planning cannot be performed without a computerized treatment planning system. Proton therapy (PT) enables highly conformal treatment plans with a minimum of dose to tissues outside the target volume, but to obtain the most optimal plan for the treatment, there are a multitude of parameters that need to be addressed. In this review areas of ongoing improvements and research in the field of PT treatment planning are identified and discussed. The main focus is on issues of immediate clinical and practical relevance to the PT community highlighting the needs for the near future but also in a longer perspective. We anticipate that the manual tasks performed by treatment planners in the future will involve a high degree of computational thinking, as many issues can be solved much better by e.g. scripting. More accurate and faster dose calculation algorithms are needed, automation for contouring and planning is required and practical tools to handle the variable biological efficiency in PT is urgently demanded just to mention a few of the expected improvements over the coming 10 years.

Impact of TPS calculation algorithms on dose delivered to the patient in proton therapy treatments

To estimate the impact of dose calculation approaches adopted in different treatment planning systems (TPSs) on proton therapy dose delivered with pencil beam scanning (PBS).Treatment plans for six regular volumes in water and 15 clinical cases were optimized with Syngo-VC13 and exported for forward recalculation with Raystation-V7.0 pencil beam (RS-PBA) and Monte Carlo (RS-MC) algorithms and with the independent Fluka-MC engine. To verify clinical consistency between the two TPS dosimetric outcomes, the average percentage variations of clinical target volume (CTV) D98%, D50% and D2%, adopted for plan prescription and evaluation, were considered. Ionization chamber measurements served as a further reference for comparison in homogeneous conditions. CTV dose volume histogram (DVH) analysis and gamma evaluation with 3 mm—3% agreement criteria quantified the dose deviation of TPS calculation algorithms, in heterogeneous conditions, against the Fluka-MC code.CTV D50%, representing the plan dose prescription goal, was higher on average over H&N cases of (3.9 ± 0.9)% and (2.3 ± 0.6)% as calculated with RS-PBA and RS-MC, respectively, compared to Syngo. For tumors located in the pelvis district, average D50% variations of (1.6 ± 0.7)% and (1.2 ± 0.7)% were found. Syngo underestimated target near maximum doses with respect to all computation systems. Calculation accuracy in heterogeneous conditions of RS-PBA H&N plans resulted poor when a range shifter was required. Target DVH and γ-analysis showed excellent agreement between RS-MC and Fluka-MC, with γ-pass rates >98% for all patient groups.Different TPS dose calculation approaches mainly affected dose delivered in H&N proton treatments, while minor deviations were found for pelvic tumors. RS-MC proved to be the most accurate TPS dose calculation algorithm when compared to an independent MC simulation code.

Validation and application of a fast Monte Carlo algorithm for assessing the clinical impact of approximations in analytical dose calculations for pencil beam scanning proton therapy.

Monte Carlo (MC) dose calculation is generally superior to analytical dose calculation (ADC) used in commercial TPS to model the dose distribution especially for heterogeneous sites, such as lung and head/neck patients. The purpose of this study was to provide a validated, fast, and open-source MC code, MCsquare, to assess the impact of approximations in ADC on clinical pencil beam scanning (PBS) plans covering various sites. First, MCsquare was validated using tissue-mimicking IROC lung phantom measurements as well as benchmarked with the general purpose Monte Carlo TOPAS for patient dose calculation. Then a comparative analysis between MCsquare and ADC was performed for a total of 50 patients with 10 patients per site (including liver, pelvis, brain, head-and-neck, and lung). Differences among TOPAS, MCsquare, and ADC were evaluated using four dosimetric indices based on the dose-volume histogram (target Dmean, D95, homogeneity index, V95), a 3D gamma index analysis (using 3%/3mm criteria), and estimations of tumor control probability (TCP). Comparison between MCsquare and TOPAS showed less than 1.8% difference for all of the dosimetric indices/TCP values and resulted in a 3D gamma index passing rate for voxels within the target in excess of 99%. When comparing ADC and MCsquare, the variances of all the indices were found to increase as the degree of tissue heterogeneity increased. In the case of lung, the D95s for ADC were found to differ by as much as 6.5% from the corresponding MCsquare statistic. The median gamma index passing rate for voxels within the target volume decreased from 99.3% for liver to 75.8% for lung. Resulting TCP differences can be large for lung (≤10.5%) and head-and-neck (≤6.2%), while smaller for brain, pelvis and liver (≤1.5%). Given the differences found in the analysis, accurate dose calculation algorithms such as Monte Carlo simulations are needed for proton therapy, especially for disease sites with high heterogeneity, such as head-and-neck and lung. The establishment of MCsquare can facilitate patient plan reviews at any institution and can potentially provide unbiased comparison in clinical trials given its accuracy, speed and open-source availability.

P259] Radiobiological method for the clinical validation of models dose-to-water versus dose-to-meduim in radiotherapy

Purpose A present challenge in medical physics is to implement the most accurate dose calculation algorithm, close to Monte Carlo simulations, with the highest safety to obtain the expected tumour control probability (TCP) and acceptable normal tissue complication probabilities (NTCP). The purpose is to propose dosimetric and radiobiological criteria to safely implement new algorithms in radiotherapy. Methods Lung tumors are taken as examples to demonstrate this transition. The doses were calculated using three types of algorithm: type (a) as pencil beam kernels model, like PBC; type (b) as point kernels model, like AAA and type (c) as Acuros-XB based on the deterministic resolution of the Boltzmann linear transport equation. Acuros-XB offers results close to Monte Carlo simulations and propose two options: dose to water (AXB-Dw) and dose to medium (AXB-Dm). The monitor units (MUs) are retrieved from the reference plan to recalculate the dose distribution with the new one. The equivalent uniform dose (EUD), TCP, NTCP and Uncomplicated Tumor Control Probability (UTCP) were calculated to measure the clinical benefit – toxicity. Results The more advanced algorithms showed that irradiating the lung tumor according to the same prescription, would increase the MUs (5–10%) when moving from pencil kernel to point kernel, and ∼ 2% when moving from AXB-Dw to AXB-Dm, p Conclusion The adjustment of prescribed dose depends on transition types and prescription method. In the lung, a reduction of prescription is advisable when moving from pencil beam model to point kernels model (e.g, 3 × 18 Gy instead of 3 × 20 Gy for SBRT and 5% for 3DCRT). In addition, an optimization should be carried out to obtain the same UTCP. To keep the same TCP and NTCP, we recommend respectively that the EUD for the target ⩾ prescription dose and EUD for healthy lung

Film-based dose validation of Monte Carlo algorithm for Cyberknife system with a CIRS thorax phantom.

Monte Carlo (MC) simulation, as the most accurate dose calculation algorithm, is available in the MultiPlan treatment planning system for Cyberknife. The main purpose of this work was to perform experiments to thoroughly investigate the accuracy of the MC dose calculation algorithm. Besides the basic MC beam commissioning, two test scenarios were designed. First, single beam tests were performed with a solid water phantom to verify the MC source model in simple geometry. Then, a lung treatment plan on a CIRS thorax phantom was created to mimic the clinical patient treatment. The plan was optimized and calculated using ray tracing (RT) algorithm and then recalculated using MC algorithm. Measurements were performed in both a homogeneous phantom and a heterogeneous phantom (CIRS). Ion‐chamber and radiochromic film were used to obtain absolute point dose and dose distributions. Ion‐chamber results showed that the differences between measured and MC calculated dose were within 3% for all tests. On the film measurements, MC calculation results showed good agreements with the measured dose for all single beam tests. As for the lung case, the gamma passing rate between measured and MC calculated dose was 98.31% and 97.28% for homogeneous and heterogeneous situation, respectively, using 3%/2 mm criteria. However, RT algorithm failed with the passing rate of 79.25% (3%/2 mm) for heterogeneous situation. These results demonstrated that MC dose calculation algorithm in the Multiplan system is accurate enough for patient dose calculation. It is strongly recommended to use MC algorithm in heterogeneous media.

Critical Appraisal of the Risk of Secondary Cancer Induction From Breast Radiation Therapy With Volumetric Modulated Arc Therapy Relative to 3D Conformal Therapy

To evaluate the excess absolute risk (EAR) comparing volumetric modulated arc therapy (VMAT) and 3-dimensional (3D) conformal radiation therapy (CRT) in breast cancer radiation therapy treatment. Two VMAT arrangements (VMAT_tang and VMAT_full, i.e. partial arcs with and without a sector of 0 Monitor Unit, respectively) and a 3D CRT (field-in-field [FinF]) plan were calculated with an accurate dose calculation algorithm, Acuros, in 20 patients presenting with early-stage breast cancer. The dose prescription was 40.05Gy in 15 fractions. The planning aim was to maximize the dose reduction in the lungs, contralateral breast, heart, and coronary artery. EAR was estimated using different models: linear, linear-exponential, plateau, and full model, which better uses a carcinogenesis model and epidemiologic data for carcinoma induction and which accounts for cell repopulation or repair during the radiation therapy dose fractionation. EAR was computed for contralateral structures-breast and lung-as well as the ipsilateral lung. Normal tissue complication probability (NTCP) was computed to estimate the ipsilateral lung, heart, and skin toxicity, to balance with respect to second cancer induction. The planning objectives were fulfilled with all the planning techniques. EAR for contralateral breast carcinoma induction, estimated with the most accurate model, was 1.7, 2.4, and 8.5 (per 10,000 patients per year) with FinF, VMAT_tang, and VMAT_full, respectively. For the contralateral lung, these figures were 1.5, 1.6, and 7.3 (per 10,000 patients per year), respectively. NTCP for all the analyzed endpoints was significantly higher with FinF relative to both VMAT settings, with VMAT_full presenting the lowest toxicity risk. VMAT, in particular with the VMAT_tang setting, could have the same risk of second cancer induction as 3D CRT delivered with the FinF setting for the contralateral organs while reducing acute and late NTCP for the ipsilateral organs. VMAT might be considered a safe technique for breast cancer treatment for those aspects.

Monte Carlo study of the influence of energy spectra, mesh size, high Z element on dose and PVDR based on 1-D and 3-D heterogeneous mouse head phantom for Microbeam Radiation Therapy

PurposeTo evaluate the influence of energy spectra, mesh sizes, high Z element on dose and PVDR in Microbeam Radiation Therapy (MRT) based on 1-D analogy-mouse-head-model (1-D MHM) and 3-D voxel-mouse-head-phantom (3-D VMHP) by Monte Carlo simulation. MethodsA Microbeam-Array-Source-Model was implemented into EGSnrc/DOSXYZnrc. The microbeam size is assumed to be 25μm, 50μm or 75μm in thickness and fixed 1mm in height with 200μmc-t-c. The influence of the energy spectra of ID17@ESRF and BMIT@CLS were investigated. The mesh size was optimized. PVDR in 1-D MHM and 3-D VMHP was compared with the homogeneous water phantom. The arc influence of 3-D VMHP filled with water (3-D VMHWP) was compared with the rectangle phantom. ResultsPVDR of the lower BMIT@CLS spectrum is 2.4times that of ID17@ESRF for lower valley dose. The optimized mesh is 5µm for 25µm, and 10µm for 50µm and 75µm microbeams with 200µmc-t-c. A 500μm skull layer could make PVDR difference up to 62.5% for 1-D MHM. However this influence is limited (<5%) for the farther homogeneous media (e.g. 600µm). The peak dose uniformity of 3-D VMHP at the same depth could be up to 8% for 1.85mm×1mm irradiation field, whereas that of 3-D VMHWP is<1%. The high Z element makes the dose uniformity enhance in target. The surface arc could affect the superficial PVDR (from 44% to 21% in 0.2mm depth), whereas this influence is limited for the more depth (<1%). ConclusionAn accurate MRT dose calculation algorithm should include the influence of 3-D heterogeneous media.

Basics of Monte-Carlo Simulation: Focusing on Dose-to-medium and Dose-to-water.

Treatment planning systems with highly accurate dose calculation algorithms such as Monte-Carlo method and linear Boltzmann transport equation are becoming popular thanks to a development of the computer technology. These algorithms use new concepts, dose-to-medium and dose-to-water. However, introducing these concepts can cause confusion in clinical sites. Basic knowledges about Monte-Carlo simulation and other corresponding algorithms were explained in this article such as the principles, the parameters and words of caution.

SU‐C‐BRC‐04: Efficient Dose Calculation Algorithm for FFF IMRT with a Simplified Bivariate Gaussian Source Model

Purpose:To develop an efficient and accurate independent dose calculation algorithm with a simplified analytical source model for the quality assurance and safe delivery of Flattening Filter Free (FFF)‐IMRT on an Elekta Versa HD.Methods:The source model consisted of a point source and a 2D bivariate Gaussian source, respectively modeling the primary photons and the combined effect of head scatter, monitor chamber backscatter and collimator exchange effect. The in‐air fluence was firstly calculated by back‐projecting the edges of beam defining devices onto the source plane and integrating the visible source distribution. The effect of the rounded MLC leaf end, tongue‐and‐groove and interleaf transmission was taken into account in the back‐projection. The in‐air fluence was then modified with a fourth degree polynomial modeling the cone‐shaped dose distribution of FFF beams. Planar dose distribution was obtained by convolving the in‐air fluence with a dose deposition kernel (DDK) consisting of the sum of three 2D Gaussian functions. The parameters of the source model and the DDK were commissioned using measured in‐air output factors (Sc) and cross beam profiles, respectively. A novel method was used to eliminate the volume averaging effect of ion chambers in determining the DDK. Planar dose distributions of five head‐and‐neck FFF‐IMRT plans were calculated and compared against measurements performed with a 2D diode array (MapCHECK™) to validate the accuracy of the algorithm.Results:The proposed source model predicted Sc for both 6MV and 10MV with an accuracy better than 0.1%. With a stringent gamma criterion (2%/2mm/local difference), the passing rate of the FFF‐IMRT dose calculation was 97.2±2.6%.Conclusion:The removal of the flattening filter represents a simplification of the head structure which allows the use of a simpler source model for very accurate dose calculation. The proposed algorithm offers an effective way to ensure the safe delivery of FFF‐IMRT.

SU‐E‐T‐252: Developing a Pencil Beam Dose Calculation Algorithm for CyberKnife System

Purpose:Currently there are two dose calculation algorithms available in the Cyberknife planning system: ray‐tracing and Monte Carlo, which is either not accurate or time‐consuming for irregular field shaped by the MLC that was recently introduced. The purpose of this study is to develop a fast and accurate pencil beam dose calculation algorithm which can handle irregular field.Methods:A pencil beam dose calculation algorithm widely used in Linac system is modified. The algorithm models both primary (short range) and scatter (long range) components with a single input parameter: TPR20/10. The TPR20/20/10 value was first estimated to derive an initial set of pencil beam model parameters (PBMP). The agreement between predicted and measured TPRs for all cones were evaluated using the root mean square of the difference (RMSTPR), which was then minimized by adjusting PBMPs. PBMPs are further tuned to minimize OCR RMS (RMSocr) by focusing at the outfield region. Finally, an arbitrary intensity profile is optimized by minimizing RMSocr difference at infield region. To test model validity, the PBMPs were obtained by fitting to only a subset of cones (4) and applied to all cones (12) for evaluation.Results:With RMS values normalized to the dmax and all cones combined, the average RMSTPR at build‐up and descending region is 2.3% and 0.4%, respectively. The RMSocr at infield, penumbra and outfield region is 1.5%, 7.8% and 0.6%, respectively. Average DTA in penumbra region is 0.5mm. There is no trend found in TPR or OCR agreement among cones or depths.Conclusion:We have developed a pencil beam algorithm for Cyberknife system. The prediction agrees well with commissioning data. Only a subset of measurements is needed to derive the model. Further improvements are needed for TPR buildup region and OCR penumbra. Experimental validations on MLC shaped irregular field needs to be performed.This work was partially supported by the National Natural Science Foundation of China (61171005) and the China Scholarship Council (CSC).

A deterministic solution of the first order linear Boltzmann transport equation in the presence of external magnetic fields.

Accurate radiotherapy dose calculation algorithms are essential to any successful radiotherapy program, considering the high level of dose conformity and modulation in many of today's treatment plans. As technology continues to progress, such as is the case with novel MRI-guided radiotherapy systems, the necessity for dose calculation algorithms to accurately predict delivered dose in increasingly challenging scenarios is vital. To this end, a novel deterministic solution has been developed to the first order linear Boltzmann transport equation which accurately calculates x-ray based radiotherapy doses in the presence of magnetic fields. The deterministic formalism discussed here with the inclusion of magnetic fields is outlined mathematically using a discrete ordinates angular discretization in an attempt to leverage existing deterministic codes. It is compared against the EGSnrc Monte Carlo code, utilizing the emf_macros addition which calculates the effects of electromagnetic fields. This comparison is performed in an inhomogeneous phantom that was designed to present a challenging calculation for deterministic calculations in 0, 0.6, and 3 T magnetic fields oriented parallel and perpendicular to the radiation beam. The accuracy of the formalism discussed here against Monte Carlo was evaluated with a gamma comparison using a standard 2%/2 mm and a more stringent 1%/1 mm criterion for a standard reference 10 × 10 cm(2) field as well as a smaller 2 × 2 cm(2) field. Greater than 99.8% (94.8%) of all points analyzed passed a 2%/2 mm (1%/1 mm) gamma criterion for all magnetic field strengths and orientations investigated. All dosimetric changes resulting from the inclusion of magnetic fields were accurately calculated using the deterministic formalism. However, despite the algorithm's high degree of accuracy, it is noticed that this formalism was not unconditionally stable using a discrete ordinate angular discretization. The feasibility of including magnetic field effects in a deterministic solution to the first order linear Boltzmann transport equation is shown. The results show a high degree of accuracy when compared against Monte Carlo calculations in all magnetic field strengths and orientations tested.

A medical image-based graphical platform—Features, applications and relevance for brachytherapy

Brachytherapy dose calculation is commonly performed using the Task Group-No 43 Report-Updated protocol (TG-43U1) formalism. Recently, a more accurate approach has been proposed that can handle tissue composition, tissue density, body shape, applicator geometry, and dose reporting either in media or water. Some model-based dose calculation algorithms are based on Monte Carlo (MC) simulations. This work presents a software platform capable of processing medical images and treatment plans, and preparing the required input data for MC simulations. The A Medical Image-based Graphical platfOrm-Brachytherapy module (AMIGOBrachy) is a user interface, coupled to the MCNP6 MC code, for absorbed dosecalculations. The AMIGOBrachy was first validated in water for a high-dose-rate (192)Ir source. Next, dose distributions were validated in uniform phantoms consisting of different materials. Finally, dose distributions were obtained in patient geometries. Results were compared against a treatment planning system including a linear Boltzmann transport equation (LBTE) solver capable of handling nonwater heterogeneities. The TG-43U1 source parameters are in good agreement with literature with more than 90% of anisotropy values within 1%. No significant dependence on the tissue composition was observed comparing MC results against an LBTE solver. Clinical cases showed differences up to 25%, when comparing MC results against TG-43U1. About 92% of the voxels exhibited dose differences lower than 2% when comparing MC results against an LBTE solver. The AMIGOBrachy can improve the accuracy of the TG-43U1 dose calculation by using a more accurate MC dose calculation algorithm. The AMIGOBrachy can be incorporated in clinical practice via a user-friendly graphical interface.

SU‐E‐T‐465: Dose Calculation Method for Dynamic Tumor Tracking Using a Gimbal‐Mounted Linac

Purpose:Dynamic tumor tracking using the gimbal‐mounted linac (Vero4DRT, Mitsubishi Heavy Industries, Ltd., Japan) has been available when respiratory motion is significant. The irradiation accuracy of the dynamic tumor tracking has been reported to be excellent. In addition to the irradiation accuracy, a fast and accurate dose calculation algorithm is needed to validate the dose distribution in the presence of respiratory motion because the multiple phases of it have to be considered. A modification of dose calculation algorithm is necessary for the gimbal‐mounted linac due to the degrees of freedom of gimbal swing. The dose calculation algorithm for the gimbal motion was implemented using the linear transformation between coordinate systems.Methods:The linear transformation matrices between the coordinate systems with and without gimbal swings were constructed using the combination of translation and rotation matrices. The coordinate system where the radiation source is at the origin and the beam axis along the z axis was adopted. The transformation can be divided into the translation from the radiation source to the gimbal rotation center, the two rotations around the center relating to the gimbal swings, and the translation from the gimbal center to the radiation source. After operating the transformation matrix to the phantom or patient image, the dose calculation can be performed as the no gimbal swing. The algorithm was implemented in the treatment planning system, PlanUNC (University of North Carolina, NC). The convolution/superposition algorithm was used. The dose calculations with and without gimbal swings were performed for the 3 × 3 cm2 field with the grid size of 5 mm.Results:The calculation time was about 3 minutes per beam. No significant additional time due to the gimbal swing was observed.Conclusions:The dose calculation algorithm for the finite gimbal swing was implemented. The calculation time was moderate.

SU‐E‐T‐584: Evaluation of Collapsed Cone Dose Calculation Algorithm for HDR Brachytherapy Using Monte Carlo

Purpose: In clinical routine, dose calculation for HDR brachytherapy plans is usually based on the AAPM‐TG43 formalism. Because the patient's anatomy is not taken into account, the resulting dose calculation may be inaccurate. Monte Carlo (MC) methods are able to calculate dose distributions accurately, but due to long computation times they are mainly used for benchmarking purposes. In this work, a test version of the collapsed cone (CC) algorithm from Nucletron was evaluated using MC. Methods: The Nucletron brachytherapy source micro Selectron‐HDR v2 has been implemented in the MC environment. CC and MC calculated dose distributions for a single source placed in a water phantom are validated against published AAPM‐TG43 data. Furthermore, dose distributions for clinical breast cancer patient treatment plans are calculated with CC and MC. Primary and scatter dose components are separated and compared individually. Results: The local differences between the MC calculated dose distributions in the water phantom and the AAPM‐TG43 data are less than 1%, while the differences between MC and CC for distances &gt; 2 cm from the source are up to 2% of the reference dose (dose at 1 cm from the source in its equatorial plane). These differences are assigned to the first and residual scatter dose components. For the clinical breast cancer patients, CC underestimates the MC dose up to 2% in the high‐dose region (&gt; 100% of the prescribed dose). This deviation is correlated with the scan‐length of the patient CT data set. If the CT data set extends the PTV by an appropriate amount, the underestimation falls below 1%. Conclusion: The CC algorithm is an accurate dose calculation algorithm and improves the dose calculation accuracy in brachytherapy compared to AAPM‐TG43 based algorithms. This work was supported by Nucletron. This work was supported by Nucletron.

A comparison between anisotropic analytical and multigrid superposition dose calculation algorithms in radiotherapy treatment planning

Monte Carlo (MC) simulation is currently the most accurate dose calculation algorithm in radiotherapy planning but requires relatively long processing time. Faster model-based algorithms such as the anisotropic analytical algorithm (AAA) by the Eclipse treatment planning system and multigrid superposition (MGS) by the XiO treatment planning system are 2 commonly used algorithms. This study compared AAA and MGS against MC, as the gold standard, on brain, nasopharynx, lung, and prostate cancer patients. Computed tomography of 6 patients of each cancer type was used. The same hypothetical treatment plan using the same machine and treatment prescription was computed for each case by each planning system using their respective dose calculation algorithm. The doses at reference points including (1) soft tissues only, (2) bones only, (3) air cavities only, (4) soft tissue-bone boundary (Soft/Bone), (5) soft tissue-air boundary (Soft/Air), and (6) bone-air boundary (Bone/Air), were measured and compared using the mean absolute percentage error (MAPE), which was a function of the percentage dose deviations from MC. Besides, the computation time of each treatment plan was recorded and compared. The MAPEs of MGS were significantly lower than AAA in all types of cancers (p<0.001). With regards to body density combinations, the MAPE of AAA ranged from 1.8% (soft tissue) to 4.9% (Bone/Air), whereas that of MGS from 1.6% (air cavities) to 2.9% (Soft/Bone). The MAPEs of MGS (2.6%±2.1) were significantly lower than that of AAA (3.7%±2.5) in all tissue density combinations (p<0.001). The mean computation time of AAA for all treatment plans was significantly lower than that of the MGS (p<0.001). Both AAA and MGS algorithms demonstrated dose deviations of less than 4.0% in most clinical cases and their performance was better in homogeneous tissues than at tissue boundaries. In general, MGS demonstrated relatively smaller dose deviations than AAA but required longer computation time.