Search NASA⌕ Search

SEARCH · Search NASA

Results for “Workload”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,351 records · Page 75

Harnessing Data Movement in Virtual Clusters for In-Situ Execution

As a result of increasing data volume and velocity, Big Data science at exascale has shifted towards the in-situ paradigm, where large scale simulations run concurrently alongside data analytics. With in-situ, data generated from simulations can be processed while still in memory, thereby avoiding the slow storage bottleneck. However, running simulations and analytics together on shared resources will likely result in substantial contention if left unmanaged, as demonstrated in this work, leading to much reduced efficiency of simulations and analytics. Recently, virtualization technologies such as Linux containers have been widely applied to data centers and physical clusters to provide highly efficient and elastic resource provisioning for consolidated workloads including scientific simulations and data analytics. In this paper, we investigate to facilitate network traffic manipulation and reduce mutual interference on the network for in-situ applications in virtual clusters. In order to dynamically allocate the network bandwidth when it is needed, we adopt SARIMA-based techniques to analyze and predict MPI traffic issued from simulations. Although this can be an effective technique, the naïve usage of network virtualization can lead to performance degradation for bursty asynchronous transmissions within an MPI job. Here, we analyze and resolve this performance degradation in virtual clusters.

97 MATHEMATICS AND COMPUTING↗

Managing Rich Metadata in High-Performance Computing Systems Using a Graph Model

High-performance computing (HPC) systems generate huge amounts of metadata about different entities such as jobs, users, and files. Existing systems can efficiently record and manage part of these metadata, mainly the POSIX metadata of data files (e.g., file size, name, and permissions mode). But another important set of metadata, referred to as "rich" metadata in this study, which record not only wider range of entities (e.g., running processes and jobs) but also more complex relationships between them, are mostly missing in current HPC systems. Yet such rich metadata are critical for supporting many advanced data management functions such as identifying data sources and parameters behind a given result; auditing data usage; or understanding details about how inputs are transformed into outputs. To uniformly and efficiently manage the rich metadata generated in HPC systems, We propose to utilize a graph model in this study. We identify the key challenges of implementing such a graph-based HPC rich metadata management system and present GraphMeta, a graph-based rich metadata management system designed and optimized for HPC platforms, to tackle these challenges. In conclusion, extensive evaluations on both synthetic and real HPC metadata workloads show its advantages in both performance and scalability compared with existing solutions.

97 MATHEMATICS AND COMPUTING↗

Customizable Scale-Out Key-Value Stores

Enterprise KV stores are often not well suited for HPC applications, and thus cumbersome end-to-end KV design customization is required to meet the needs of modern HPC applications. To this end, in this article we present bespoKV, an adaptive, extensible, and scale-out KV store framework. bespoKV decouples the KV store design into the control plane for distributed management and the data plane for local data store. For the control plane, bespoKVprovides pre-built modules, called controlets, supporting common distributed functionalities (e.g., replication, consistency, and topology) and their various combinations. This decoupling allows bespoKV to take a user-provided single-server KV store, called a datalet, and transparently enables a scalable and fault-tolerant distributed KV store service. The resulting distributed stores are also adaptive to consistency or topology requirement changes and can be easily extended for new types of services. Such specializations enable innovative uses of KV stores in HPC applications, especially for emerging applications that utilize KV-friendly workloads. We evaluate bespoKV in a local testbed as well as in a public cloud settings. Experiments show that bespoKV-enabled distributed KV stores scale horizontally to a large number of nodes, and performs comparably and sometimes 1.2× to 2.6× better than the state-of-the-art systems.

97 MATHEMATICS AND COMPUTING↗

Enabling Scalable and Extensible Memory-mapped Datastores in Userspace

Exascale workloads are expected to incorporate data-intensive processing in close coordination with traditional physics simulations. These emerging scientific, data-analytics and machine learning applications need to access a wide variety of datastores in flat files and structured databases. Programmer productivity is greatly enhanced by mapping datastores into the application process's virtual memory space to provide a unified “in-memory” interface. Currently, memory mapping is provided by system software primarily designed for generality and reliability. However, scalability at high concurrency is a formidable challenge on exascale systems. Also, there is a need for extensibility to support new datastores potentially requiring HPC data transfer services. In this article, we present UMap , a scalable and extensible userspace service for memory-mapping datastores. Furthermore, through decoupled queue management, concurrency aware adaptation, and dynamic load balancing, UMap enables application performance to scale even at high concurrency. We evaluate UMap in data-intensive applications, including sorting, graph traversal, database operations, and metagenomic analytics. Our results show that UMap as a userspace service outperforms an optimized kernel-based service across a wide range of intra-node concurrency by 1.22-1.9 × . We performed two case studies to demonstrate UMap 's extensibility. First, a new datastore residing in remote memory is incorporated into UMap as an application-specific plugin. Second, we present a persistent memory allocator Metall built atop UMap for unified storage/memory.

97 MATHEMATICS AND COMPUTING↗

GSoFa: Scalable Sparse Symbolic LU Factorization on GPUs

Decomposing a matrix $\mathbf {A}$ into a lower matrix $\mathbf {L}$ and an upper matrix $\mathbf {U}$, which is also known as LU decomposition, is an essential operation in numerical linear algebra. For a sparse matrix, LU decomposition often introduces more nonzero entries in the $\mathbf {L}$ and $\mathbf {U}$ factors than in the original matrix. A symbolic factorization step is needed to identify the nonzero structures of $\mathbf {L}$ and $\mathbf {U}$ matrices. Attracted by the enormous potentials of the Graphics Processing Units (GPUs), an array of efforts have surged to deploy various LU factorization steps except for the symbolic factorization, to the best of our knowledge, on GPUs. This article introduces gSoFa, the first GPU-based symbolic factorization design with the following three optimizations to enable scalable LU symbolic factorization for nonsymmetric pattern sparse matrices on GPUs. First, here we introduce a novel fine-grained parallel symbolic factorization algorithm that is well suited for the Single Instruction Multiple Thread (SIMT) architecture of GPUs. Second, we tailor supernode detection into a SIMT friendly process and strive to balance the workload, minimize the communication and saturate the GPU computing resources during supernode detection. Third, we introduce a three-pronged optimization to reduce the excessive space consumption problem faced by multi-source concurrent symbolic factorization. Taken together, gSoFa achieves up to 31× speedup from 1 to 44 Summit nodes (6 to 264 GPUs) and outperforms the state-of-the-art CPU project, on average, by 5×. Notably, gSoFa also achieves up to 47 percent of the peak memory throughput of a V100 GPU in the Summit Supercomputer.

97 MATHEMATICS AND COMPUTING↗

IRIS: A Performance-Portable Framework for Cross-Platform Heterogeneous Computing

From edge to exascale, computer architectures are becoming more heterogeneous and complex. The systems typically have fat nodes, with multicore CPUs and multiple hardware accelerators such as GPUs, FPGAs, and DSPs. This complexity is causing a crisis in programming systems and performance portability. Several programming systems are working to address these challenges, but the increasing architectural diversity is forcing software stacks and applications to be specialized for each architecture. As we show, all of these approaches critically depend on their software framework for discovery, execution, scheduling, and data orchestration. To address this challenge, we believe that a more agile and proactive software framework is essential to increase performance portability and improve user productivity. To this end, we have designed and implemented IRIS: a performance-portable framework for cross-platform heterogeneous computing. IRIS can discover available resources, manage multiple diverse programming platforms (e.g., CUDA, Hexagon, HIP, Level Zero, OpenCL, OpenMP) simultaneously in the same execution, respect data dependencies, orchestrate data movement proactively, and provide for user-configurable scheduling. To simplify data movement, IRIS introduces a shared virtual device memory with relaxed consistency among different heterogeneous devices. IRIS also adds an automatic kernel workload partitioning technique using the polyhedral model so that it can resize kernels for a wide range of devices. Our evaluation on three architectures, ranging from Qualcomm Snapdragon to a Summit supercomputer node, shows that IRIS improves portability across a wide range of diverse heterogeneous architectures with negligible overhead.

97 MATHEMATICS AND COMPUTING↗

Analog In-Memory Computing for the Synthetic Aperture Radar Polar Format Algorithm

As the utility of synthetic aperture radar (SAR) systems increases in autonomous vehicles, satellites, and other power- and space-constrained edge applications, there is a growing need for processors that can form SAR images at low power. In recent years, analog in-memory compute (AIMC) has shown immense promise for accelerating neural networks and other matrix-vector multiplication (MVM) heavy workloads at the edge. Here, in this work, we examine how the polar format algorithm (PFA), a popular SAR image formation algorithm, can be mapped to these AIMC systems. The PFA maps readily onto analog MVMs because it primarily consists of two linear operations: interpolation of frequency-domain data to a Cartesian grid, followed by a 2-D Fourier transform. This work presents two approaches to map the interpolation operation onto MVMs in analog hardware: a chirp transform and a modified form of sinc interpolation. These mappings introduce algorithmic errors, and their effect on the quality of SAR image formation is examined, both quantitatively and qualitatively. In addition, the impact of errors introduced by the analog hardware is explored to determine which approach is optimal under varying assumptions about the underlying analog memory devices and circuits.

Analog computing↗

A Scalable Hybrid Scheme for Ray-Casting of Unstructured Volume Data

We present an algorithm for parallel volume rendering that is a hybrid between classical object order and image order techniques. The algorithm operates on unstructured grids (and structured ones), and thus can deal with block boundaries interleaving in complex ways. It also deals effectively with cases that are prone to load imbalance, i.e., cases where cell sizes differ dramatically, either because of the nature of the input data, or because of the effects of the camera transformation. The algorithm divides work over resources such that each phase of its processing is bounded in the amount of computation it can perform. We demonstrate its efficacy through a series of studies, varying over camera position, data set size, transfer function, image size, and processor count. At its biggest, our experiments scaled up to 8,192 processors and operated on data sets with more than one billion cells. In total, we find that our hybrid algorithm performs well in all cases. This is because our algorithm naturally adapts its computation based on workload, and can operate like either an object order technique or an image order technique in scenarios where those techniques are efficient.

97 MATHEMATICS AND COMPUTING↗

A Study on the Impact of Temperature-Dependent Ferroelectric Switching Behavior in 3D Memory Architecture

The flourishing development of neural networks that require exponentially growing amounts of data has presented an elevated demand for memory footprint. To address this, researchers have been exploring hardware accelerators with innovative memory architectures like 3D memory. These 3D memory architectures offer enhanced storage capacity and processing capabilities, at a cost of rising on-chip temperature during operation. Hafnium Zirconium Oxide (HZO) based Ferroelectric Random Access Memory (FeRAM) is a promising nonvolatile memory candidate in neural network hardware accelerators for its outstanding write performance and reliability. However, its implementation in the architecture regarding the temperature-dependent ferroelectric switching behavior has not been well studied. In this work, we study the thermal impacts on polarization switching through experimental devices and simulation results. We conduct the circuit and architecture-level simulations to showcase that one can exploit this temperature rise to reduce FeRAM's write voltage and write energy due to its unique temperature-activated polarization switching mechanisms. As the on-chip temperature increases to 351K (ambient temperature at 300K) due to neural network workloads, the access energy per bit can be reduced by 27.6% when a dynamic write voltage is applied.

36 MATERIALS SCIENCE↗

Geostatistical mapping of Cs-137 contamination depth in building structures by integrating ISOCS measurements of different spatial supports

Reliable methods to determine the contamination depth in nuclear building structures are very much needed for minimizing the radioactive waste volume and the decontamination workload. This paper investigates the geostatistical integration of in situ gamma-ray spectroscopy measurements of different spatial supports. A case study is presented from the BR3 decommissioning project, yielding an estimated reduction of waste volume of ∼35%, and recommendations are made for future application of the proposed methodology. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

An enhanced sector integration model for output and dose distribution calculation of irregular concave shaped electron beams

A comprehensive method of output factor and dose distribution calculation for electron beams has been developed. It allows one to calculate the output factors and isodose distributions in water of arbitrary shaped electron fields with excellent accuracy even for the cases of concaved, small, elongated beams, and extended source to surface distances (SSDs). The method requires two sets of data: Depth dose distribution per monitor unit for circular cutouts and depth dose distributions per monitor unit for circular blocks (plugs), both for two SSDs, one reference of 100 cm and second extended one. The method has been extensively tested using a combination of different irregular cutouts and various SSDs for the 6 and 9 MeV electron beams. The calculated values agreed with the measured data well within 1% for output factors and below 1 for {gamma} (gamma test) for isodose distributions. The computer program has been developed to facilitate the method for practical application. The method has been used for almost 8 years considerably cutting workload in the department.

61 RADIATION PROTECTION AND DOSIMETRY↗

The inter- and intrafraction reproducibilities of three common IMRT delivery techniques

Purpose: Intensity modulated radiation therapy (IMRT) treatment delivery requires higher precision than conventional 3D treatment delivery because of the sensitivity of the resulting dose to small geometric misalignment of the modulated beamlets. The chosen treatment delivery technique will affect the treatment precision in different ways, based on the characteristics of the delivery method. Delivery using a multileaf collimator (MLC) can reduce treatment time and therapist workload, but typically requires a greater number of monitor units and the fields are prone to both systematic and random leaf positioning errors. An alternative to MLC-based fields, patient specific brass compensators, do not suffer from these leaf positioning errors. In our study, we set out to investigate which delivery method will provide the highest levels of dosimetric reproducibility and the minimum amount of interfraction variability. Methods: In our study, a seven field IMRT plan for a head and neck treatment was created using the Pinnacle{sup 3} treatment planning system and the intensity maps for each field were obtained. The intensity maps of the fields were delivered with a Varian 2100C/D linear accelerator, using solid compensators and sliding window (SW) and step-and-shoot (SS) MLC segments. Three fields were selected from the seven-beam IMRT plan for comparison. Analysis was carried out using the MatriXX ion chamber array, radiochromic film, and Varian dynalog files. Results: Our results show that the error in MLC leaf positioning has no gantry angle dependence. The compensator and SW deliveries showed excellent agreement, even when stricter than usual gamma criteria were applied. However, we noted that under these strict conditions, the SS fields had at least ten times more pixels out of range than did the compensators. When using step-and-shoot MLC fields, it was observed that the increase in dose rate or the increase of MU/segment degrades the quality of the plan. Analysis of the dynalog files showed that while each individual field had its own propensity for error, all fields showed the same trend: a greater percentage of time the leaves are out of position as dose rate increases, MUs decrease, or both. Conclusions: The compensator-based field and both types of MLC-based fields have MatriXX results that are within the clinically acceptable tolerance of 3% dose difference and 2 mm DTA. However, when the criteria are tightened, it becomes evident that the compensators have a definite advantage over their comparable MLC-based competitors in terms of interfraction reproducibility. Fewer monitor units are required to deliver each portal, potentially improving patient outcomes and reducing unwanted side effects to both patients and therapists. In centers without MLC, compensators represent a simple and cost effective way to offer patients state of the art treatment. Based on the results of this study, compensator-based IMRT is a reliable, viable option for use in clinics both with and without MLC-equipped linacs.

61 RADIATION PROTECTION AND DOSIMETRY↗

iCycle: Integrated, multicriterial beam angle, and profile optimization for generation of coplanar and noncoplanar IMRT plans

Purpose: To introduce iCycle, a novel algorithm for integrated, multicriterial optimization of beam angles, and intensity modulated radiotherapy (IMRT) profiles. Methods: A multicriterial plan optimization with iCycle is based on a prescription called wish-list, containing hard constraints and objectives with ascribed priorities. Priorities are ordinal parameters used for relative importance ranking of the objectives. The higher an objective priority is, the higher the probability that the corresponding objective will be met. Beam directions are selected from an input set of candidate directions. Input sets can be restricted, e.g., to allow only generation of coplanar plans, or to avoid collisions between patient/couch and the gantry in a noncoplanar setup. Obtaining clinically feasible calculation times was an important design criterium for development of iCycle. This could be realized by sequentially adding beams to the treatment plan in an iterative procedure. Each iteration loop starts with selection of the optimal direction to be added. Then, a Pareto-optimal IMRT plan is generated for the (fixed) beam setup that includes all so far selected directions, using a previously published algorithm for multicriterial optimization of fluence profiles for a fixed beam arrangement Breedveld et al.[Phys. Med. Biol. 54, 7199-7209 (2009)]. To select the next direction, each not yet selected candidate direction is temporarily added to the plan and an optimization problem, derived from the Lagrangian obtained from the just performed optimization for establishing the Pareto-optimal plan, is solved. For each patient, a single one-beam, two-beam, three-beam, etc. Pareto-optimal plan is generated until addition of beams does no longer result in significant plan quality improvement. Plan generation with iCycle is fully automated. Results: Performance and characteristics of iCycle are demonstrated by generating plans for a maxillary sinus case, a cervical cancer patient, and a liver patient treated with SBRT. Plans generated with beam angle optimization did better meet the clinical goals than equiangular or manually selected configurations. For the maxillary sinus and liver cases, significant improvements for noncoplanar setups were seen. The cervix case showed that also in IMRT with coplanar setups, beam angle optimization with iCycle may improve plan quality. Computation times for coplanar plans were around 1-2 h and for noncoplanar plans 4-7 h, depending on the number of beams and the complexity of the site. Conclusions: Integrated beam angle and profile optimization with iCycle may result in significant improvements in treatment plan quality. Due to automation, the plan generation workload is minimal. Clinical application has started.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Shielding implications for secondary neutrons and photons produced within the patient during IMPT

Purpose: Intensity modulated proton therapy (IMPT) uses a combination of computer controlled spot scanning and spot-weight optimized planning to irradiate the tumor volume uniformly. In contrast to passive scattering systems, secondary neutrons and photons produced from inelastic proton interactions within the patient represent the major source of emitted radiation during IMPT delivery. Various published studies evaluated the shielding considerations for passive scattering systems but did not directly address secondary neutron production from IMPT and the ambient dose equivalent on surrounding occupational and nonoccupational work areas. Thus, the purpose of this study was to utilize Monte Carlo simulations to evaluate the energy and angular distributions of secondary neutrons and photons following inelastic proton interactions within a tissue-equivalent phantom for incident proton spot energies between 70 and 250 MeV.Methods: Monte Carlo simulation methods were used to calculate the ambient dose equivalent of secondary neutrons and photons produced from inelastic proton interactions in a tissue-equivalent phantom. The angular distribution of emitted neutrons and photons were scored as a function of incident proton energy throughout a spherical annulus at 1, 2, 3, 4, and 5 m from the phantom center. Appropriate dose equivalent conversion factors were applied to estimate the total ambient dose equivalent from secondary neutrons and photons.Results: A reference distance of 1 m from the center of the patient was used to evaluate the mean energy distribution of secondary neutrons and photons and the resulting ambient dose equivalent. For an incident proton spot energy of 250 MeV, the total ambient dose equivalent (3.6 Multiplication-Sign 10{sup -3} mSv per proton Gy) was greatest along the direction of the incident proton spot (0 Degree-Sign -10 Degree-Sign ) with a mean secondary neutron energy of 71.3 MeV. The dose equivalent decreased by a factor of 5 in the backward direction (170 Degree-Sign -180 Degree-Sign ) with a mean energy of 4.4 MeV. An 8 Multiplication-Sign 8 Multiplication-Sign 8 cm{sup 3} volumetric spot distribution (5 mm FWHM spot size, 4 mm spot spacing) optimized to produce a uniform dose distribution results in an ambient dose equivalent of 4.5 Multiplication-Sign 10{sup -2} mSv per proton Gy in the forward direction.Conclusions: This work evaluated the secondary neutron and photon emission due to monoenergetic proton spots between 70 and 250 MeV, incident on a tissue equivalent phantom. Example calculations were performed to estimate concrete shield thickness based upon appropriate workload and shielding design assumptions. Although lower than traditional passive scattered proton therapy systems, the ambient dose equivalent from secondary neutrons produced by the patient during IMPT can be significant relative to occupational and nonoccupational workers in the vicinity of the treatment vault. This work demonstrates that Monte Carlo simulations are useful as an initial planning tool for studying the impact of the treatment room and maze design on surrounding occupational and nonoccupational work areas.

61 RADIATION PROTECTION AND DOSIMETRY↗

Vision 20/20: Perspectives on automated image segmentation for radiotherapy

Due to rapid advances in radiation therapy (RT), especially image guidance and treatment adaptation, a fast and accurate segmentation of medical images is a very important part of the treatment. Manual delineation of target volumes and organs at risk is still the standard routine for most clinics, even though it is time consuming and prone to intra- and interobserver variations. Automated segmentation methods seek to reduce delineation workload and unify the organ boundary definition. In this paper, the authors review the current autosegmentation methods particularly relevant for applications in RT. The authors outline the methods’ strengths and limitations and propose strategies that could lead to wider acceptance of autosegmentation in routine clinical practice. The authors conclude that currently, autosegmentation technology in RT planning is an efficient tool for the clinicians to provide them with a good starting point for review and adjustment. Modern hardware platforms including GPUs allow most of the autosegmentation tasks to be done in a range of a few minutes. In the nearest future, improvements in CT-based autosegmentation tools will be achieved through standardization of imaging and contouring protocols. In the longer term, the authors expect a wider use of multimodality approaches and better understanding of correlation of imaging with biology and pathology.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

SU-D-BRE-07: Neutron Shielding Assessment for a Compact Proton Therapy Vault

Purpose: To perform a neutron shielding assessment of a commercially available compact proton therapy system. Methods: TOPAS (TOol for PArticle Simulation) beta release was used to model beam line components for Mevion S250 proton treatment system the design of which is that the cyclotron is present in the treatment room. Three neutron production sources were taken into account in the simulation. These are the cyclotron, the treatment nozzle and the patient itself, respectively. The cyclotron was modeled as a cylindrical iron target (r =5 cm, length = 8 cm). A water phantom (10 cm ×10 cm ×60 cm) was used to model the patient and various structures (scattering foils, range modulator wheel, applicator and compensator) defaulted in TOPAS were used to model the passive scattering treatment nozzle. Neutron fluences and energy spectra were counted in a spherical scoring geometry per incident proton in 18 angular bins (10 degree each). Fluence to dose conversion factors from ICRU publication 74 were used to acquire neutron ambient dose equivalent H*(10). A point source line of sight model was then used to calculate neutron dose at eight locations beyond shielding barriers. Results: The neutron ambient dose equivalent was calculated at the 8 points of interest around the proton treatment vault. The highest dose was found to be less than 0.781 mSv/year outside south barrier wall. However, the dose is less than 0.05 mSv/year at the control room area of the proton vault. Conclusion: All Points of interest were well under annual dose limits. This suggests that the shielding design of this compact proton therapy system is sufficient for radiation protection purpose. However, it is important to note that the workload and the occupancy factors are direct multipliers for dose calculations beyond the barrier and must be accurately estimated for validation of our results.

07 ISOTOPE AND RADIATION SOURCES↗

SU-E-T-88: Comprehensive Automated Daily QA for Hypo- Fractionated Treatments

Purpose: The trend towards more SBRT treatments with fewer high dose fractions places increased importance on daily QA. Patient plan specific QA with 3%/3mm gamma analysis and daily output constancy checks may not be enough to guarantee the level of accuracy required for SBRT treatments. But increasing the already extensive amount of QA procedures that are required is a daunting proposition. We performed a feasibility study for more comprehensive automated daily QA that could improve the diagnostic capabilities of QA without increasing workload. Methods: We performed the study on a Siemens Artiste linear accelerator using the integrated flat panel EPID. We included square fields, a picket fence, overlap and representative IMRT fields to measure output, flatness, symmetry, beam center, and percent difference from the standard. We also imposed a set of machine errors: MLC leaf position, machine output, and beam steering to compare with the standard. Results: Daily output was consistent within +/− 1%. Change in steering current by 1.4% and 2.4% resulted in a 3.2% and 6.3% change in flatness. 1 and 2mm MLC leaf offset errors were visibly obvious in difference plots, but passed a 3%/3mm gamma analysis. A simple test of transmission in a picket fence can catch a leaf offset error of a single leaf by 1mm. The entire morning QA sequence is performed in less than 30 minutes and images are automatically analyzed. Conclusion: Automated QA procedures could be used to provide more comprehensive information about the machine with less time and human involvement. We have also shown that other simple tests are better able to catch MLC leaf position errors than a 3%/3mm gamma analysis commonly used for IMRT and modulated arc treatments. Finally, this information could be used to watch trends of the machine and predict problems before they lead to costly machine downtime.

60 APPLIED LIFE SCIENCES↗

SU-E-T-433: Calibration Accuracy in Mailed High-Resolution 3D Dosimetry Service for SRS/SBRT QA

Purpose: SRS/SBRT combines hypofractionation with excellent dose distributions. However, extremely steep gradients across the target along with dose escalation, if not administered accurately, may lead to serious complications, recurrences, or even fatalities. Existing commercial QA products either lack adequate spatial resolution or the 3D aspect. By contrast, the new CrystalBall™ mailed high-resolution 3D dosimetry service removes the above limitations while reducing the overall workload on medical physics staff. The exposed dosimeters, which change optical density in proportion to local dose, are sent back to the manufacturer (MGS Research Inc., Madison, CT) for sub-millimeter-resolution laser-CT scanning and QA data analysis. QA report is returned electronically within 24 hours. The purpose of this study was to evaluate the dose calibration accuracy in this system. Methods: Two spherical CrystalBall™ polymer gel dosimeters from the same batch, 166 mm diameter, with embedded 3D image registration markers, were mounted in a special phantom designed for reproducible positioning. For full end to end testing, the optical guidance array was mounted onto the phantom and a CT was taken. Two separate Rapid Arc SRS plans were designed. Varian Medical Systems optical guidance system was used to position the phantom and the SRS treatment plans were delivered to the two spheres on Varian's Trilogy Accelerator. Exposed dosimeters were mailed back to the manufacturer for laser CT scanning and analysis. Results: For each plan, 3D gamma passing rate was 100% for 2%/2mm distance-to-agreement criteria above 50% isodose level. The two calibration curves, generated using volumetric dose and optical density data, showed excellent mutual agreement (max difference 2.2%, median difference 0.75%). Conclusion: The clinical utility of new CrystalBall™ mailed QA service for SRS/SBRT and high accuracy of dose calibration have been validated. The workflow associated with the use of the CrystalBall™ in clinical setting was found to be minimal. The presenting author is the founder of and has an ownership in MGS Research Inc which manufactures the CrystalBall system for 3D dosimetry.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗