Justin Mikell

Dosimetric comparison of Acuros XB deterministic radiation transport method with Monte Carlo and model‐based convolution methods in heterogeneous media

PURPOSE The deterministic Acuros XB (AXB) algorithm was recently implemented in the Eclipse treatment planning system. The goal of this study was to compare AXB performance to Monte Carlo (MC) and two standard clinical convolution methods: the anisotropic analytical algorithm (AAA) and the collapsed-cone convolution (CCC) method. METHODS Homogeneous water and multilayer slab virtual phantoms were used for this study. The multilayer slab phantom had three different materials, representing soft tissue, bone, and lung. Depth dose and lateral dose profiles from AXB v10 in Eclipse were compared to AAA v10 in Eclipse, CCC in Pinnacle3, and EGSnrc MC simulations for 6 and 18 MV photon beams with open fields for both phantoms. In order to further reveal the dosimetric differences between AXB and AAA or CCC, three-dimensional (3D) gamma index analyses were conducted in slab regions and subregions defined by AAPM Task Group 53. RESULTS The AXB calculations were found to be closer to MC than both AAA and CCC for all the investigated plans, especially in bone and lung regions. The average differences of depth dose profiles between MC and AXB, AAA, or CCC was within 1.1, 4.4, and 2.2%, respectively, for all fields and energies. More specifically, those differences in bone region were up to 1.1, 6.4, and 1.6%; in lung region were up to 0.9, 11.6, and 4.5% for AXB, AAA, and CCC, respectively. AXB was also found to have better dose predictions than AAA and CCC at the tissue interfaces where backscatter occurs. 3D gamma index analyses (percent of dose voxels passing a 2%/2 mm criterion) showed that the dose differences between AAA and AXB are significant (under 60% passed) in the bone region for all field sizes of 6 MV and in the lung region for most of field sizes of both energies. The difference between AXB and CCC was generally small (over 90% passed) except in the lung region for 18 MV 10 x 10 cm2 fields (over 26% passed) and in the bone region for 5 x 5 and 10 x 10 cm2 fields (over 64% passed). With the criterion relaxed to 5%/2 mm, the pass rates were over 90% for both AAA and CCC relative to AXB for all energies and fields, with the exception of AAA 18 MV 2.5 x 2.5 cm2 field, which still did not pass. CONCLUSIONS In heterogeneous media, AXB dose prediction ability appears to be comparable to MC and superior to current clinical convolution methods. The dose differences between AXB and AAA or CCC are mainly in the bone, lung, and interface regions. The spatial distributions of these differences depend on the field sizes and energies.

Experimental validation of deterministic Acuros XB algorithm for IMRT and VMAT dose calculations with the Radiological Physics Center's head and neck phantom.

PURPOSE The purpose of this study was to verify the dosimetric performance of Acuros XB (AXB), a grid-based Boltzmann solver, in intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT). METHODS The Radiological Physics Center (RPC) head and neck (H&N) phantom was used for all calculations and measurements in this study. Clinically equivalent IMRT and VMAT plans were created on the RPC H&N phantom in the Eclipse treatment planning system (version 10.0) by using RPC dose prescription specifications. The dose distributions were calculated with two different algorithms, AXB 11.0.03 and anisotropic analytical algorithm (AAA) 10.0.24. Two dose report modes of AXB were recorded: dose-to-medium in medium (D(m,m)) and dose-to-water in medium (D(w,m)). Each treatment plan was delivered to the RPC phantom three times for reproducibility by using a Varian Clinac iX linear accelerator. Absolute point dose and planar dose were measured with thermoluminescent dosimeters (TLDs) and GafChromic® EBT2 film, respectively. Profile comparison and 2D gamma analysis were used to quantify the agreement between the film measurements and the calculated dose distributions from both AXB and AAA. The computation times for AAA and AXB were also evaluated. RESULTS Good agreement was observed between measured doses and those calculated with AAA or AXB. Both AAA and AXB calculated doses within 5% of TLD measurements in both the IMRT and VMAT plans. Results of AXB_D(m,m) (0.1% to 3.6%) were slightly better than AAA (0.2% to 4.6%) or AXB_D(w,m) (0.3% to 5.1%). The gamma analysis for both AAA and AXB met the RPC 7%/4 mm criteria (over 90% passed), whereas AXB_D(m,m) met 5%/3 mm criteria in most cases. AAA was 2 to 3 times faster than AXB for IMRT, whereas AXB was 4-6 times faster than AAA for VMAT. CONCLUSIONS AXB was found to be satisfactorily accurate when compared to measurements in the RPC H&N phantom. Compared with AAA, AXB results were equal to or better than those obtained with film measurements for IMRT and VMAT plans. The AXB_D(m,m) reporting mode was found to be closer to TLD and film measurements than was the AXB_D(w,m) mode. AXB calculation time was found to be significantly shorter (× 4) than AAA for VMAT.

Dosimetric impact of Acuros XB deterministic radiation transport algorithm for heterogeneous dose calculation in lung cancer

PURPOSE The novel deterministic radiation transport algorithm, Acuros XB (AXB), has shown great potential for accurate heterogeneous dose calculation. However, the clinical impact between AXB and other currently used algorithms still needs to be elucidated for translation between these algorithms. The purpose of this study was to investigate the impact of AXB for heterogeneous dose calculation in lung cancer for intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT). METHODS The thorax phantom from the Radiological Physics Center (RPC) was used for this study. IMRT and VMAT plans were created for the phantom in the Eclipse 11.0 treatment planning system. Each plan was delivered to the phantom three times using a Varian Clinac iX linear accelerator to ensure reproducibility. Thermoluminescent dosimeters (TLDs) and Gafchromic EBT2 film were placed inside the phantom to measure delivered doses. The measurements were compared with dose calculations from AXB 11.0.21 and the anisotropic analytical algorithm (AAA) 11.0.21. Two dose reporting modes of AXB, dose-to-medium in medium (Dm,m) and dose-to-water in medium (Dw,m), were studied. Point doses, dose profiles, and gamma analysis were used to quantify the agreement between measurements and calculations from both AXB and AAA. The computation times for AAA and AXB were also evaluated. RESULTS For the RPC lung phantom, AAA and AXB dose predictions were found in good agreement to TLD and film measurements for both IMRT and VMAT plans. TLD dose predictions were within 0.4%-4.4% to AXB doses (both Dm,m and Dw,m); and within 2.5%-6.4% to AAA doses, respectively. For the film comparisons, the gamma indexes (± 3%∕3 mm criteria) were 94%, 97%, and 98% for AAA, AXB_Dm,m, and AXB_Dw,m, respectively. The differences between AXB and AAA in dose-volume histogram mean doses were within 2% in the planning target volume, lung, heart, and within 5% in the spinal cord. However, differences up to 8% between AXB and AAA were found at lung∕soft tissue interface regions for individual IMRT fields. AAA was found to be 5-6 times faster than AXB for IMRT, while AXB was 4-5 times faster than AAA for VMAT plan. CONCLUSIONS AXB is satisfactorily accurate for the dose calculation in lung cancer for both IMRT and VMAT plans. The differences between AXB and AAA are generally small except in heterogeneous interface regions. AXB Dw,m and Dm,m calculations are similar inside the soft tissue and lung regions. AXB can benefit lung VMAT plans by both improving accuracy and reducing computation time.

Monte Carlo calculations of AAPM Task Group Report No. 43 dosimetry parameters for the 125I I-Seed AgX100 source model

PURPOSE The purpose of this study was to determine the dosimetric parameters of the AgX100, a new (125)I brachytherapy seed model, using Monte Carlo (MC) simulations according to the protocol specified by the updated American Association of Physicists in Medicine Task Group No. 43 Report (TG-43U1) and compare these parameters with those of the established brachytherapy (125)I seed models 6711 and I25.S06. METHODS AND MATERIALS Independent verification of the new seed geometry was performed using high-resolution digital radiography and scanning electron microscopy. MCNPX v.2.5 MC simulations of the AgX100 seed were performed to derive its TG-43U1 parameters, the dose rate constant, the radial dose function, and the two- and one-dimensional anisotropy functions in liquid water. A dosimetric error propagation analysis was also performed to include uncertainty because of seed manufacturing tolerances and physics parameters. RESULTS The MC-calculated dose rate constant for the AgX100 seed was 0.943cGy·h(-1)·U(-1)±2.6% (k=1) based on the air kerma strength for a simulated point detector. Tabulated results of the radial dose function for line and point source approximations and the two-dimensional anisotropy function are also reported. CONCLUSIONS The MC-predicted dose distribution of the AgX100 seed was found to be comparable with that of the model 6711 seed but much different from the dose distribution of the model I25.S06 seeds. However, at shallow distances, there were some dosimetric differences between the AgX100 and 6711 seed, which warrant separate TG-43U1 parameters for use in clinical treatment planning systems.

Practical reconstruction protocol for quantitative 90Y bremsstrahlung SPECT/CT

PURPOSE To develop a practical background compensation (BC) technique to improve quantitative (90)Y-bremsstrahlung single-photon emission computed tomography (SPECT)/computed tomography (CT) using a commercially available imaging system. METHODS All images were acquired using medium-energy collimation in six energy windows (EWs), ranging from 70 to 410 keV. The EWs were determined based on the signal-to-background ratio in planar images of an acrylic phantom of different thicknesses (2-16 cm) positioned below a (90)Y source and set at different distances (15-35 cm) from a gamma camera. The authors adapted the widely used EW-based scatter-correction technique by modeling the BC as scaled images. The BC EW was determined empirically in SPECT/CT studies using an IEC phantom based on the sphere activity recovery and residual activity in the cold lung insert. The scaling factor was calculated from 20 clinical planar (90)Y images. Reconstruction parameters were optimized in the same SPECT images for improved image quantification and contrast. A count-to-activity calibration factor was calculated from 30 clinical (90)Y images. RESULTS The authors found that the most appropriate imaging EW range was 90-125 keV. BC was modeled as 0.53× images in the EW of 310-410 keV. The background-compensated clinical images had higher image contrast than uncompensated images. The maximum deviation of their SPECT calibration in clinical studies was lowest (<10%) for SPECT with attenuation correction (AC) and SPECT with AC + BC. Using the proposed SPECT-with-AC + BC reconstruction protocol, the authors found that the recovery coefficient of a 37-mm sphere (in a 10-mm volume of interest) increased from 39% to 90% and that the residual activity in the lung insert decreased from 44% to 14% over that of SPECT images with AC alone. CONCLUSIONS The proposed EW-based BC model was developed for (90)Y bremsstrahlung imaging. SPECT with AC + BC gave improved lesion detectability and activity quantification compared to SPECT with AC only. The proposed methodology can readily be used to tailor (90)Y SPECT/CT acquisition and reconstruction protocols with different SPECT/CT systems for quantification and improved image quality in clinical settings.

Effects of image noise, respiratory motion, and motion compensation on 3D activity quantification in count-limited PET images

The aims of this study were to evaluate the effects of noise, motion blur, and motion compensation using quiescent-period gating (QPG) on the activity concentration (AC) distribution-quantified using the cumulative AC volume histogram (ACVH)-in count-limited studies such as 90Y-PET/CT. An International Electrotechnical Commission phantom filled with low 18F activity was used to simulate clinical 90Y-PET images. PET data were acquired using a GE-D690 when the phantom was static and subject to 1-4 cm periodic 1D motion. The static data were down-sampled into shorter durations to determine the effect of noise on ACVH. Motion-degraded PET data were sorted into multiple gates to assess the effect of motion and QPG on ACVH. Errors in ACVH at AC90 (minimum AC that covers 90% of the volume of interest (VOI)), AC80, and ACmean (average AC in the VOI) were characterized as a function of noise and amplitude before and after QPG. Scan-time reduction increased the apparent non-uniformity of sphere doses and the dispersion of ACVH. These effects were more pronounced in smaller spheres. Noise-related errors in ACVH at AC20 to AC70 were smaller (<15%) compared to the errors between AC80 to AC90 (>15%). The accuracy of ACmean was largely independent of the total count. Motion decreased the observed AC and skewed the ACVH toward lower values; the severity of this effect depended on motion amplitude and tumor diameter. The errors in AC20 to AC80 for the 17 mm sphere were  -25% and  -55% for motion amplitudes of 2 cm and 4 cm, respectively. With QPG, the errors in AC20 to AC80 of the 17 mm sphere were reduced to  -15% for motion amplitudes  <4 cm. For spheres with motion amplitude to diameter ratio  >0.5, QPG was effective at reducing errors in ACVH despite increases in image non-uniformity due to increased noise. ACVH is believed to be more relevant than mean or maximum AC to calculate tumor control and normal tissue complication probability. However, caution needs to be exercised when using ACVH in post-therapy 90Y imaging because of its susceptibility to image degradation from both image noise and respiratory motion.

WE‐D‐BRB‐02: Comparing a Grid‐Based Boltzmann Solver with Monte Carlo Simulation for Voxel‐Based Therapeutic Radionuclide Dose Calculations

Purpose: To compare the accuracy and speed of a deterministic grid‐based Boltzmann solver (GBBS) with Monte Carlo(MC) simulations for calculating voxel‐based absorbed dose rates from SPECT/CT imaging.Methods: A SPECT/CT image of a breast cancer patient with metastatic osteosarcoma was obtained using a tracer administration of 153 Sm EDTMP. The therapeutic activity administered was determined from MIRD estimates. DOSXYZnrc and the GBBS Acuros were used to calculate dose rate maps from the activity distribution over the full CT grid. The GBBS photondose rate was calculated using the collisional KERMA approximation rather than explicitly transporting the generated electrons; an energy cutoff was used to neglect spatial transport of electrons below a threshold of 200 keV for Acuros and 189 keV for DOSXYZnrc. The photon, beta‐particle, and total absorbed dose rates calculated using GBBS were compared with the gold standard MC simulations using the gamma index. Gamma index parameters evaluated were 3%/3mm and 5%/5mm; both used a step size of 0.5mm with a 10 mm search distance. A patient mask was created from the CT to report pixels within/near the patient. The gamma failure points (γ>1.0) within the patient mask were viewed overlaid on the activity map and on the CT for the calculated beta, photon, and total dose rates. Results: For total dose rate, 90.1% and 99.6% of pixels within the patient mask had γ⩽1.0 for 3%/3mm and 5%/5mm respectively. γ failures were noticed at the edges of the activity distribution for beta‐particles, and at various interfaces for the photons. For Acuros, the beta‐particle and photon transport required about 10 minutes each. DOSXYZnrc took approximately 15,000 times longer. Conclusions: GBBS has the potential to provide accuracy similar to MC in a much shorter time; this could be useful for voxel‐based radionuclide absorbed dose estimates in a clinical setting.

Evaluation of a deterministic grid-based Boltzmann solver (GBBS) for voxel-level absorbed dose calculations in nuclear medicine

To evaluate the 3D Grid-based Boltzmann Solver (GBBS) code ATTILA (®) for coupled electron and photon transport in the nuclear medicine energy regime for electron (beta, Auger and internal conversion electrons) and photon (gamma, x-ray) sources. Codes rewritten based on ATTILA are used clinically for both high-energy photon teletherapy and (192)Ir sealed source brachytherapy; little information exists for using the GBBS to calculate voxel-level absorbed doses in nuclear medicine. We compared DOSXYZnrc Monte Carlo (MC) with published voxel-S-values to establish MC as truth. GBBS was investigated for mono-energetic 1.0, 0.1, and 0.01 MeV electron and photon sources as well as (131)I and (90)Y radionuclides. We investigated convergence of GBBS by analyzing different meshes ([Formula: see text]), energy group structures ([Formula: see text]) for each radionuclide component, angular quadrature orders ([Formula: see text], and scattering order expansions ([Formula: see text]-[Formula: see text]); higher indices imply finer discretization. We compared GBBS to MC in (1) voxel-S-value geometry for soft tissue, lung, and bone, and (2) a source at the interface between combinations of lung, soft tissue, and bone. Excluding Auger and conversion electrons, MC agreed within  ≈5% of published source voxel absorbed doses. For the finest discretization, most GBBS absorbed doses in the source voxel changed by less than 1% compared to the next finest discretization along each phase space variable indicating sufficient convergence. For the finest discretization, agreement with MC in the source voxel ranged from  -3% to  -20% with larger differences at lower energies (-3% for 1 MeV electron in lung to  -20% for 0.01 MeV photon in bone); similar agreement was found for the interface geometries. Differences between GBBS and MC in the source voxel for (90)Y and (131)I were  -6%. The GBBS ATTILA was benchmarked against MC in the nuclear medicine regime. GBBS can be a viable alternative to MC for voxel-level absorbed doses in nuclear medicine. However, reconciliation of the differences between GBBS and MC at lower energies requires further investigation of energy deposition cross-sections.

Hepatocellular Carcinoma Tumor Dose Response After 90Y-radioembolization With Glass Microspheres Using 90Y-SPECT/CT-Based Voxel Dosimetry

PURPOSE To investigate hepatocellular carcinoma tumor dose-response characteristics based on voxel-level absorbed doses (D) and biological effective doses (BED) using quantitative 90Y-single-photon emission computed tomography (SPECT)/computed tomography (CT) after 90Y-radioembilization with glass microspheres. We also investigated the relationship between normal liver D and toxicities. METHODS AND MATERIALS 90Y-radioembolization activity distributions for 34 patients were based on quantitative 90Y-bremsstrahlung SPECT/CT. D maps were generated using a local-deposition algorithm. Contrast-enhanced CT or magnetic resonance imaging scans of the liver were registered to 90Y-SPECT/CT, and all tumors larger than 2.5 cm diameter (53 tumors) were segmented. Tumor mean D and BED (Dmean and BEDmean) and dose volume coverage from 0% to 100% in 10% steps (D0-D100 and BED0-BED100) were extracted. Tumor response was evaluated on follow-up using World Health Organization (WHO), Response Evaluation Criteria in Solid Tumors (RECIST), and modified RECIST (mRECIST) criteria. Differences in dose metrics for responders and nonresponders were assessed using the Mann-Whitney U test. A univariate logistic regression model was used to determine tumor dose metrics that correlated with tumor response. Correlations among tumor size, tumor Dmean, and tumor dose heterogeneity (defined as the coefficient of variation) were assessed. RESULTS The objective response rates were 14 of 53, 15 of 53, and 30 of 53 for WHO, RECIST, and mRECIST criteria, respectively. WHO and RECIST response statuses did not correlate with D or BED. For mRECIST responders and nonresponders, D and BED were significantly different for Dmean, D20 to D80, BEDmean, and BED0 to BED80. Threshold doses (and the 95% confidence interval) for 50% probability of mRECIST response (D50%) were 160 Gy (123-196 Gy) for Dmean and 214 Gy (146-280 Gy) for BEDmean. Tumor dose heterogeneity significantly correlated with tumor volume. No statistically significant association between Dmean to normal liver and complications related to bilirubin, albumin, or ascites was observed. CONCLUSIONS Hepatocellular carcinoma tumor dose-response curves after 90Y-radioembolization with glass microspheres showed Dmean of 160 Gy and BEDmean of 214 Gy for D50% with a positive predictive value of ∼70% and a negative predictive value of ∼62%. No complications were observed in our patient cohort for normal liver Dmean less than 44 Gy.

TH-AB-BRA-10: Clinical Implementation of a Grid-Based Boltzmann Solver with Adaptive Meshing for Nuclear Medicine Dosimetry

Purpose: We previously compared the Grid-Based Boltzmann Solver (GBBS) with DOSXYZnrc Monte Carlo (MC) for beta and gamma sources in uniform materials and along material interfaces. The current standard in nuclear medicine is MIRD, which only calculates mean organ doses assuming uniform uptake in anthropomorphic phantoms. This work extends GBBS to clinical nuclear medicine for fast and accurate patient-specific voxel-level dosimetry. Methods: We compared the GBBS with MC using a patient’s post-therapy quantitative 90Y microsphere bremsstrahlung SPECT/CT.The GBBS (Attila™ v8.0.0) requires a tetrahedral mesh as input instead of voxels. Adaptive tetrahedral meshes were generated with TetGen v1.5.0 by folding CT and SPECT images into a mesh sizing function; the function was defined at each node on a background mesh with nodes defined at each SPECT/CT voxel. Target tetrahedral edge lengths varied from 0.5 to 8 cm and were adapted on SPECT activity level, CT material, and material gradient.GBBS used 30 electron energy groups, 32 angles, and 67,000 tetrahedrons. GBBS reported point data aligned with the center of SPECT/CT voxels which match MC. We report: 1) run time of the GBBS, 2) average difference in non-zero absorbed dose voxels, and 3) percent difference in mean absorbed dose to tumor and normal liver. DVHs and isodose curves were visually compared, and the tetrahedral mesh was compared to SPECT/CT. Results: GBBS calculation required 2 minutes. The average difference over all non-zero absorbed dose voxels (N=318,000) was −0.2 Gy. GBBS mean doses agreed within 1% of MC for both normal liver (39.1 Gy vs 39.3 Gy) and tumor (317 Gy vs 320 Gy). Adaptive meshing allowed sufficient material assignment and yielded smaller tetrahedrons near high activity regions and bone. DVHs had excellent visual agreement. Conclusions: GBBS with adaptive meshing is practical for fast and accurate clinical nuclear medicine dosimetry. Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA138986. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.