Search NASA⌕ Search

SEARCH · Search NASA

Results for “predict”

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 1,135 records · Page 63

A machine learning framework for elastic constants predictions in multi-principal element alloys

On the one hand, multi-principal element alloys (MPEAs) have created a paradigm shift in alloy design due to large compositional space, whereas on the other, they have presented enormous computational challenges for theory-based materials design, especially density functional theory (DFT), which is inherently computationally expensive even for traditional dilute alloys. In this paper, we present a machine learning framework, namely PREDICT (PRedict properties from Existing Database In Complex alloys Territory), that opens a pathway to predict elastic constants in large compositional space with little computational expense. The framework only relies on the DFT database of binary alloys and predicts Voigt–Reuss–Hill Young’s modulus, shear modulus, bulk modulus, elastic constants, and Poisson’s ratio in MPEAs. We show that the key descriptors of elastic constants are the A–B bond length and cohesive energy. The framework can predict elastic constants in hypothetical compositions as long as the constituent elements are present in the database, thereby enabling property exploration in multi-compositional systems. We illustrate predictions in a FCC Ni-Cu-Au-Pd-Pt system.

Linton, Nathan (ORCID:0000000315485613)↗

Repetitive proteins that undergo large conformational changes evade structural prediction algorithms

Protein structure prediction algorithms, such as AlphaFold, have accelerated protein design and advanced the understanding of the relationship between amino acid sequence and protein structure. However, these algorithms are limited in their ability to predict the structures of conformationally dynamic, intrinsically disordered, and stimuli-responsive proteins. To evaluate sequence-to-structure predictions of such challenging proteins, we explored a class of conformationally dynamic, repeats-in-toxin (RTX) proteins. RTX proteins adopt intrinsically disordered conformations in the absence of calcium and undergo reversible folding into β-roll structures upon binding to calcium. RTX proteins are characterized by tandem repeats of the sequence GGXGXDXUX, in which X can be any amino acid and U is an aliphatic amino acid. We designed RTX sequence variants with global substitutions of nonconserved amino acids, tandem repeats of consensus sequences GGAGXDTLY, and tandem repeats of scrambled sequences GGAGXDTYL. AlphaFold2 and AlphaFold3 predicted that all of these RTX variants adopt β-roll structures, characteristic of wild-type RTX bound to calcium. However, modeling the predicted structures with molecular dynamics simulations and characterizing the protein variants with circular dichroism spectroscopy, small-angle x-ray scattering, and x-ray crystallography revealed that variants adopt diverse, sequence-dependent structures in the absence and presence of calcium. To better design proteins for applications in biotechnology and sustainability, it is critical to build predictive tools that consider intrinsically disordered protein states and validate these tools with multi-mode, multi-scale experimental data.

Chang, Marina P. [Stanford Univ., CA (United State↗

Population structure limits the use of genomic data for predicting phenotypes and managing genetic resources in forest trees

There is overwhelming evidence that forest trees are locally adapted to climate. Thus, genecological models based on population phenotypes have been used to measure local adaptation, infer genetic maladaptation to climate, and guide assisted migration. However, instead of phenotypes, there is increasing interest in using genomic data for gene resource management. We used whole-genome resequencing and common-garden experiments to understand the genetic architecture of adaptive traits in black cottonwood. We studied the potential of using genome-wide association studies (GWAS) and genomic prediction to detect causal loci, identify climate-adapted phenotypes, and inform gene resource management. We analyzed population structure by partitioning phenotypic and genomic (single-nucleotide polymorphism) variation among 840 genotypes collected from 91 stands along 16 rivers. Most phenotypic variation (60 to 81%) occurred among populations and was strongly associated with climate. Population phenotypes were predicted well using genomic data (e.g., predictive abilityr> 0.9) but almost as well using climate or geography (r> 0.8). In contrast, genomic prediction within populations was poor (r< 0.2). We identified many GWAS associations among populations, but most appeared to be spurious based on pooled within-population analyses. Hierarchical partitioning of linkage disequilibrium and haplotype sharing suggested that within-population genomic prediction and GWAS were poor because allele frequencies of causal loci and linked markers differed among populations. Given the urgent need to conserve natural populations and ecosystems, our results suggest that climate variables alone can be used to predict population phenotypes, delineate seed zones and deployment zones, and guide assisted migration.

Science & Technology - Other Topics↗

EMPIRICAL PREDICTIONS FOR (SUB-)MILLIMETER LINE AND CONTINUUM DEEP FIELDS

Modern (sub-)millimeter/radio interferometers such as ALMA, JVLA, and the PdBI successor NOEMA will enable us to measure the dust and molecular gas emission from galaxies that have luminosities lower than the Milky Way, out to high redshifts and with unprecedented spatial resolution and sensitivity. This will provide new constraints on the star formation properties and gas reservoir in galaxies throughout cosmic times through dedicated deep field campaigns targeting the CO/[C II] lines and dust continuum emission in the (sub-)millimeter regime. In this paper, we present empirical predictions for such line and continuum deep fields. We base these predictions on the deepest available optical/near-infrared Advanced Camera for Surveys and NICMOS data on the Hubble Ultra Deep Field (over an area of about 12 arcmin{sup 2}). Using a physically motivated spectral energy distribution model, we fit the observed optical/near-infrared emission of 13,099 galaxies with redshifts up to z = 5, and obtain median-likelihood estimates of their stellar mass, star formation rate, dust attenuation, and dust luminosity. We combine the attenuated stellar spectra with a library of infrared emission models spanning a wide range of dust temperatures to derive statistical constraints on the dust emission in the infrared and (sub-)millimeter which are consistent with the observed optical/near-infrared emission in terms of energy balance. This allows us to estimate, for each galaxy, the (sub-)millimeter continuum flux densities in several ALMA, PdBI/NOEMA, and JVLA bands. As a consistency check, we verify that the 850 {mu}m number counts and extragalactic background light derived using our predictions are consistent with previous observations. Using empirical relations between the observed CO/[C II] line luminosities and the infrared luminosity of star-forming galaxies, we infer the luminosity of the CO(1-0) and [C II] lines from the estimated infrared luminosity of each galaxy in our sample. We then predict the luminosities of higher CO transition lines CO(2-1) to CO(7-6) based on two extreme gas excitation scenarios: quiescent (Milky Way) and starburst (M82). We use our predictions to discuss possible deep field strategies with ALMA. The predictions presented in this study will serve as a direct benchmark for future deep field campaigns in the (sub-)millimeter regime.

79 ASTRONOMY AND ASTROPHYSICS↗

Optimization of simulated high-field side lower hybrid current drive coupling using machine learning predictions of scrape-off layer density

Lower hybrid current drive (LHCD) is a potential source of non-inductive off-axis current drive (CD) for tokamaks. Although LHCD has been successfully deployed on a number of tokamaks, it is highly sensitive to the scrape-off layer (SOL) conditions local to the LHCD launcher. Large gaps between the launcher and plasma core, SOL turbulence, or edge density perturbations due to edge-localized modes can hamper CD or cause large reflected power. These coupling issues in part motivated the installation of an LHCD launcher on the high-field side (HFS) of DIII-D. On the HFS, the SOL is less turbulent and more controllable compared to the low-field side. This quiescence may result in more predictable edge conditions and thus a more predictable CD. Here, in this work, HFS SOL reflectometry measurements are predicted from global plasma parameters using machine learning models. The SOL predictions coupled with the full-wave simulation of the LHCD launcher allow for the prediction of reflected power, directivity, and arcing risk before the discharge. Launcher performance is then optimized using multi-objective Bayesian optimization, finding the shot parameters that result in an optimal SOL density that maximizes CD while minimizing the risk of arcing. The predictions and optimizations of LHCD performance are then accelerated using a surrogate model of the full-wave LHCD simulation.

Bayesian optimization↗

SOLAR CYCLE PROPAGATION, MEMORY, AND PREDICTION: INSIGHTS FROM A CENTURY OF MAGNETIC PROXIES

The solar cycle and its associated magnetic activity are the main drivers behind changes in the interplanetary environment and Earth's upper atmosphere (commonly referred to as space weather). These changes have a direct impact on the lifetime of space-based assets and can create hazards to astronauts in space. In recent years there has been an effort to develop accurate solar cycle predictions (with aims at predicting the long-term evolution of space weather), leading to nearly a hundred widely spread predictions for the amplitude of solar cycle 24. A major contributor to the disagreement is the lack of direct long-term databases covering different components of the solar magnetic field (toroidal versus poloidal). Here, we use sunspot area and polar faculae measurements spanning a full century (as our toroidal and poloidal field proxies) to study solar cycle propagation, memory, and prediction. Our results substantiate predictions based on the polar magnetic fields, whereas we find sunspot area to be uncorrelated with cycle amplitude unless multiplied by area-weighted average tilt. This suggests that the joint assimilation of tilt and sunspot area is a better choice (with aims to cycle prediction) than sunspot area alone, and adds to the evidence in favor of active region emergence and decay as the main mechanism of poloidal field generation (i.e., the Babcock-Leighton mechanism). Finally, by looking at the correlation between our poloidal and toroidal proxies across multiple cycles, we find solar cycle memory to be limited to only one cycle.

79 ASTRONOMY AND ASTROPHYSICS↗

Genetic mapping and prediction of flowering time and plant height in a maize Stiff Stalk MAGIC population

Abstract The Stiff Stalk heterotic pool is a foundation of US maize seed parent germplasm and has been heavily utilized by both public and private maize breeders since its inception in the 1930s. Flowering time and plant height are critical characteristics for both inbred parents and their test crossed hybrid progeny. To study these traits, a 6-parent multiparent advanced generation intercross population was developed including maize inbred lines B73, B84, PHB47 (B37 type), LH145 (B14 type), PHJ40 (novel early Stiff Stalk), and NKH8431 (B73/B14 type). A set of 779 doubled haploid lines were evaluated for flowering time and plant height in 2 field replicates in 2016 and 2017, and a subset of 689 and 561 doubled haploid lines were crossed to 2 testers, respectively, and evaluated as hybrids in 2 locations in 2018 and 2019 using an incomplete block design. Markers were derived from a practical haplotype graph built from the founder whole genome assemblies and genotype-by-sequencing and exome capture-based sequencing of the population. Genetic mapping utilizing an update to R/qtl2 revealed differing profiles of significant loci for both traits between 635 of the DH lines and 2 sets of 570 and 471 derived hybrids. Genomic prediction was used to test the feasibility of predicting hybrid phenotypes based on the per se data. Predictive abilities were highest on direct models trained using the data they would predict (0.55–0.63), and indirect models trained using per se data to predict hybrid traits had slightly lower predictive abilities (0.49–0.55). Overall, this finding is consistent with the overlapping and nonoverlapping significant quantitative trait loci found within the per se and hybrid populations and suggests that selections for phenology traits can be made effectively on doubled haploid lines before hybrid data is available.

Genetics & Heredity↗

TEM Approaches for Microstructure-Informed Prediction of Mechanical Properties in Structural Alloys

Predicting the mechanical performance of structural alloys from their evolving microstructure remains a major challenge in materials science, particularly for nuclear structural materials, where irradiation-induced defects span multiple types and length scales and interact through complex mechanisms. The dispersed barrier hardening (DBH) [1] and Friedel–Kroupa–Hirsch (FKH) [2,3] models have been widely used to evaluate the hardening contributions of individual obstacles and to estimate tensile strength from quantified microstructures; however, when multiple size-dependent obstacles coexist and evolve, predicting temperature-dependent tensile strength becomes significantly more complex, and a fully consistent hardening model is still lacking. Transmission electron microscopy (TEM) plays a central role in refining hardening models and enabling predictive assessments of tensile strength evolution by providing quantitative characterization of dislocations, irradiation-induced defects (e.g., dislocation loops and cavities), precipitates, and grain structure (Fig. 1.). These experimentally measured defect densities are incorporated into physically based hardening models with size- and shape- dependent obstacle strengths [4], using root-sum-square superposition for obstacles of comparable strength and linear superposition for dissimilar ones [5]. In addition, recent advances in TEM [6-8], including high-resolution imaging, 4D-STEM strain mapping, EDS/EELS elemental analysis, and flash-polishing-based TEM specimen preparation and extraction-replica methods (Fig. 2), further improve the accuracy of microstructural quantification. By comparison with prior studies as well as our own results, we show that when TEM-derived microstructural information is carefully integrated with physically grounded hardening models, yield strength (or irradiation-induced hardening) measured at room temperature can be predicted with good quantitative agreement across multiple alloy classes. In-situ TEM combined with high-temperature mechanical testing represents an important next step for refining hardening models by directly probing dislocation–obstacle interactions across varying irradiation doses and temperatures [9]. Because the barrier strength factor (α) depends on both temperature and obstacle size, it should not be treated as a constant fitting parameter; rather, it must be explicitly evaluated to achieve physically meaningful predictions of mechanical behaviour at operating temperatures. This presentation therefore discusses why all strengthening contributions (e.g., Peierls stress, solid-solution strengthening, voids, bubbles, dislocation loops, dislocation lines, and grain boundaries) must be considered collectively, why appropriate superposition methods are essential when obstacles possess different barrier strength factors, how hardness measurements can be meaningfully related to tensile properties, and how TEM-derived microstructural information can be systematically incorporated into hardening models. More broadly, it outlines a pathway toward microstructure-informed prediction of mechanical properties and supports the goal of establishing science-based tools for evaluating structural materials in extreme environments [10].

Lin, Yan-Ru [ORNL] (ORCID:0000000339991473)↗

Accurate machine-learning predictions of coercivity in high-performance permanent magnets

Increased demand for high-performance permanent magnets in the electric vehicle and wind-turbine industries has prompted the search for cost-effective alternatives. Discovering magnetic materials with the desired intrinsic and extrinsic permanent magnet properties presents a significant challenge to researchers because of issues with the global supply of rare-earth elements, material stability, and a low maximum magnetic energy product BH max . While first-principles density functional theory (DFT) predicts materials’ magnetic moments, magnetocrystalline anisotropy constants, and exchange interactions, it cannot compute extrinsic properties such as coercivity (H c ). Although it is possible to calculate H c theoretically with micromagnetic simulations, the predicted value is larger than the experiment by almost an order of magnitude due to the Brown paradox. To circumvent these issues, we employ machine-learning (ML) methods on an extensive database obtained from experiments, DFT calculations, and micromagnetic modeling. The use of a large experimental dataset enables realistic H c predictions for materials such as Ce-doped Nd 2 ⁢Fe 14 ⁢B, comparing favorably against micromagnetically simulated coercivities. Remarkably, our ML model accurately identifies uniaxial magneto-crystalline anisotropy as the primary contributor to H c . With DFT calculations, we predict the Nd-site-dependent magnetic anisotropy behavior in Nd 2 ⁢Fe 14 ⁢B, confirming that Nd 4⁢g sites mainly contribute to uniaxial magnetocrystalline anisotropy, and also calculate the Curie temperature (T c ). Finally, both calculated results are in good agreement with the experiments. The coupled experimental dataset and ML modeling with DFT input predict H c with far greater accuracy and speed than was previously possible using micromagnetic modeling. Further, we reverse engineer the grain-boundary and intergrain exchange coupling with micromagnetic simulations by employing the ML predictions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

First-principles predictions of temperature-dependent infrared dielectric function of polar materials by including four-phonon scattering and phonon frequency shift

Recently, first-principles calculations based on density functional theory have been widely used to predict the temperature-dependent infrared spectrum of polar materials, but the calculations are usually limited to the harmonic frequency (0 K) and three-phonon scattering damping for the zone-center infrared-active optical phonon modes, and fail to predict the high-temperature infrared optical properties of materials such as sapphire (α–Al 2 O 3 ), GaAs, TiO 2 , etc., due to the neglect of high-order phonon scattering damping and phonon frequency shift. In this work, we implemented first-principles calculations to predict the temperature-dependent infrared dielectric function of polar materials by including four-phonon scattering and phonon frequency shift. The temperature-dependent phonon damping by including three- and four-phonon scattering as well as the phonon frequency shift by including cubic and quartic anharmonicity and the thermal expansion effect are calculated based on anharmonic lattice dynamics method. The infrared dielectric function of α–Al 2 O 3 is parameterized, and then the temperature-dependent infrared optical reflectance is determined. We find that our predictions agree better with the experimental data than the previous density functional theory-based methods. Lastly, this work will help to effectively predict the thermal radiative properties of polar materials at elevated temperature, which is generally difficult to measure, and will enable predictive design of new materials for radiative applications.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

CuXASNet: Rapid and accurate prediction of copper L-edge x-ray absorption spectra using machine learning

In this work, we have developed CuXASNet, a dense neural network that predicts simulated Cu -edge x-ray absorption spectra (XAS) from atomic structures. Featurization of the Cu local environment is performed using a component of M3GNet, a graph neural network developed for predicting the potential energy surface. CuXASNet is trained on simulated spectra from FEFF9 at the multiple scattering level of theory, and can predict the and edges for Cu sites to quantitative accuracy. To validate our approach, we compare 14 experimental spectra extracted from the literature with the predictions of CuXASNet. The agreement of CuXASNet with experiments is shown by an average mean absolute error of 0.125 and an average Spearman's correlation coefficient of 0.891, which is comparable to FEFF9's values of 0.131 and 0.898 for the same metrics. As such, CuXASNet can rapidly predict a large number of -edge XAS spectra at the same accuracy as FEFF9 simulations. This can be used as a drop-in replacement for multiple scattering codes for fast screening of candidate atomic structure models of a measured system. This model establishes a general framework for Cu XAS prediction, and can be extended to more computationally expensive levels of theory and to other transition metal edges.

36 MATERIALS SCIENCE↗

Proteome-scale Deployment of Protein Structure Prediction Workflows on the Summit Supercomputer

Deep learning has contributed to major advances in the prediction of protein structure from sequence, a fundamental problem in structural bioinformatics. With predictions now approaching the accuracy of crystallographic experiments, and with accelerators like GPUs and TPUs making inference using large models rapid, genome-level structure prediction becomes an obvious aim. Leadership-class computing resources can be used to perform genome-scale protein structure prediction using state-of-the-art deep learning models, providing a wealth of new data for systems biology applications. Here we describe our efforts to efficiently deploy the AlphaFold v.2 program, for full-proteome structure prediction, at scale on the Oak Ridge Leadership Computing Facility's resources, including the Summit supercomputer. We performed inference to produce the predicted structures for 40,526 protein sequences, corresponding to four prokaryotic proteomes and one plant proteome, using under 4,400 total Summit node hours, equivalent to using the majority of the supercomputer for a little over one hour. We also designed an optimized structure refinement that reduced the time for the relaxation stage of the AlphaFold pipeline by over 10X for longer sequences. We demonstrate the types of analyses that can be performed on proteome-scale collections of sequences, including a search for novel quaternary structures and implications for functional annotation.

Gao, Mu↗

Three-Stage Adjusted Regression Forecasting for Software Defect Prediction

In this paper, a three-stage adjusted regression forecasting model is proposed to forecast the local regression model [6]. This is a growth curve approximation model that predicts the parameters for a future linear model based on a sliding window of previous linear models. The three stages of the model are as follows: • Initial fit: train regression models on a sliding window of the date and record model coefficients. • Prediction: fit new regression models to the coefficient lists and predict the value of the next coefficient. • Error correction: correct coefficient prediction error using the residual of the last point of the coefficient list and moving average. The resulting model from the multi-stage process is a forecast of the local regression model that represents the future window of data and is referred to as the predicted line. Results suggest the three-stage model demonstrates better prediction capability compared to existing solutions.

Pritchard, Shadow↗

Prediction of DIII-D Pedestal Structure From Externally Controllable Parameters

The sharp increase of pressure at the edge of a high confinement mode (H-mode) plasma, the pedestal, strongly impacts overall plasma performance. Predicting the pedestal is a necessity to control and optimize tokamak operations. Here, an experimental data-driven machine learning (ML) approach is presented that predicts the pedestal heights and widths of electron density (n e ) and electron temperature (T e ) profiles as well as the separatrix ne from externally controllable parameters such as the plasma shape, heating method and power, and gas puff rate and integrated gas puff. The OMFIT framework was used with DIII-D data to efficiently, robustly, and automatically build a database of pedestal parameters to train machine learning models. Database creation was enabled by the search engine tool for DIII-D data, TokSearch, which parallelizes data fetching, enabling fast searches through basic signals of thousands of DIII-D shots and selection of relevant time intervals. Principal Component Analysis (PCA) separated the database into three clusters that represent classes of plasma shapes that are regularly used in DIII-D. The most important parameters for setting the pedestal structure were plasma current (I p ), toroidal magnetic field (B Φ ), neutral beam heating power (P NBI ) and shaping quantities. The Deep Jointly Informed Neural Networks (DJINN) algorithm was applied to identify suitable neural network (NN) architectures that appropriately capture the features of the pedestal database. Separate NNs were implemented for each pedestal parameter, and ensembling methods were used to improve the prediction accuracy and allowed estimation of the prediction uncertainty. The pedestal predictions of the test dataset lie within the measurement uncertainties of the pedestal parameters. The NN outperformed simple Linear Regression (LR) analysis, indicating non-linear dependencies in the pedestal structure. The presented achievements illustrate a promising path for future research, using feature extraction to infer experimental trends and thereby improve pedestal models as well as deploying NN for a fast pedestal prediction in DIII-D scenario development.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Sensitivity of Thermal Predictions to Uncertain Surface Tension Data in Laser Additive Manufacturing

To understand the process-microstructure relationships in additive manufacturing (AM), it is necessary to predict the solidification characteristics in the melt pool. This study investigates the influence of Marangoni driven fluid flow on the predicted melt pool geometry and solidification conditions using a continuum finite volume model. A calibrated laser absorptivity was determined by comparing the model predictions (neglecting fluid flow) against melt pool dimensions obtained from single laser melt experiments on a nickel super alloy 625 (IN625) plate. Using this calibrated efficiency, predicted melt pool geometries agree well with experiments across a range of process conditions. When fluid mechanics is considered, a surface tension gradient recommended for IN625 tends to overpredict the influence of convective heat transfer, but the use of an intermediate value reported from experimental measurements of a similar nickel super alloy produces excellent experimental agreement. Despite its significant effect on the melt pool geometry predictions, fluid flow was found to have a small effect on the predicted solidification conditions compared to processing conditions. We find that this result suggests that under certain circumstances, a model only considering conductive heat transfer is sufficient for approximating process-microstructure relationships in laser AM. Extending the model to multiple laser passes further showed that fluid flow also has a small effect on the solidification conditions compared to the transient variations in the process. Furthermore, limitations of the current model and areas of improvement, including uncertainties associated with the phenomenological model inputs are discussed.

36 MATERIALS SCIENCE↗

MACHINE LEARNING-ENABLED PREDICTION OF TRANSIENT INJECTION MAP IN AUTOMOTIVE INJECTORS WITH UNCERTAINTY QUANTIFICATION

Accurate prediction of injection profiles is a critical aspect of linking injector operation with engine performance and emissions. However, highly resolved injector simulations can take one to two weeks of wall-clock time, which is incompatible with engine design cycles with desired turnaround times of less than a day. Hence, it is important to reduce the time-to-solution of the internal flow simulations by several orders of magnitude to make it compatible with engine simulations. This work demonstrates a data-driven approach for tackling the computational overhead of injector simulations, whereby the transient injection profiles are emulated for a side-oriented, single-hole diesel injector using a Bayesian machine-learning framework. First, an interpretable Bayesian learning strategy was employed to understand the effect of design parameters on the total void fraction field. Then, autoencoders are utilized for efficient dimensionality reduction of the flowfields. Gaussian process models are finally used to predict the spatiotemporal void fraction field at the injector exit for unknown operating conditions. The Gaussian process models produce principled uncertainty estimates associated with the emulated flowfields, which provide the engine designer with valuable information of where the data-driven predictions can be trusted in the design space. The Bayesian flowfield predictions are compared with the corresponding predictions from a deep neural network, which has been transfer-learned from static needle simulations from a previous work by the authors. The emulation framework can predict the void fraction field at the exit of the orifice within a few seconds, thus achieving a speed-up factor of up to 38 x 10(6) over the traditional simulation-based approach of generating transient injection maps.

machine learning↗

Comprehensive fluence model for absolute portal dose image prediction

Amorphous silicon (a-Si) electronic portal imaging devices (EPIDs) continue to be investigated as treatment verification tools, with a particular focus on intensity modulated radiation therapy (IMRT). This verification could be accomplished through a comparison of measured portal images to predicted portal dose images. A general fluence determination tailored to portal dose image prediction would be a great asset in order to model the complex modulation of IMRT. A proposed physics-based parameter fluence model was commissioned by matching predicted EPID images to corresponding measured EPID images of multileaf collimator (MLC) defined fields. The two-source fluence model was composed of a focal Gaussian and an extrafocal Gaussian-like source. Specific aspects of the MLC and secondary collimators were also modeled (e.g., jaw and MLC transmission factors, MLC rounded leaf tips, tongue and groove effect, interleaf leakage, and leaf offsets). Several unique aspects of the model were developed based on the results of detailed Monte Carlo simulations of the linear accelerator including (1) use of a non-Gaussian extrafocal fluence source function, (2) separate energy spectra used for focal and extrafocal fluence, and (3) different off-axis energy spectra softening used for focal and extrafocal fluences. The predicted energy fluence was then convolved with Monte Carlo generated, EPID-specific dose kernels to convert incident fluence to dose delivered to the EPID. Measured EPID data were obtained with an a-Si EPID for various MLC-defined fields (from 1x1 to 20x20 cm{sup 2}) over a range of source-to-detector distances. These measured profiles were used to determine the fluence model parameters in a process analogous to the commissioning of a treatment planning system. The resulting model was tested on 20 clinical IMRT plans, including ten prostate and ten oropharyngeal cases. The model predicted the open-field profiles within 2%, 2 mm, while a mean of 96.6% of pixels over all IMRT fields was in agreement with the 2%, 3 mm criteria. This model demonstrates accuracy commensurate to existing methods for IMRT pretreatment verification with portal dose image prediction of complex clinical examples (<2%, 3 mm).

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A two-dimensional matrix correction for off-axis portal dose prediction errors

Purpose: This study presents a follow-up to a modified calibration procedure for portal dosimetry published by Bailey et al. ['An effective correction algorithm for off-axis portal dosimetry errors,' Med. Phys. 36, 4089-4094 (2009)]. A commercial portal dose prediction system exhibits disagreement of up to 15% (calibrated units) between measured and predicted images as off-axis distance increases. The previous modified calibration procedure accounts for these off-axis effects in most regions of the detecting surface, but is limited by the simplistic assumption of radial symmetry. Methods: We find that a two-dimensional (2D) matrix correction, applied to each calibrated image, accounts for off-axis prediction errors in all regions of the detecting surface, including those still problematic after the radial correction is performed. The correction matrix is calculated by quantitative comparison of predicted and measured images that span the entire detecting surface. The correction matrix was verified for dose-linearity, and its effectiveness was verified on a number of test fields. The 2D correction was employed to retrospectively examine 22 off-axis, asymmetric electronic-compensation breast fields, five intensity-modulated brain fields (moderate-high modulation) manipulated for far off-axis delivery, and 29 intensity-modulated clinical fields of varying complexity in the central portion of the detecting surface. Results: Employing the matrix correction to the off-axis test fields and clinical fields, predicted vs measured portal dose agreement improves by up to 15%, producing up to 10% better agreement than the radial correction in some areas of the detecting surface. Gamma evaluation analyses (3 mm, 3% global, 10% dose threshold) of predicted vs measured portal dose images demonstrate pass rate improvement of up to 75% with the matrix correction, producing pass rates that are up to 30% higher than those resulting from the radial correction technique alone. As in the 1D correction case, the 2D algorithm leaves the portal dosimetry process virtually unchanged in the central portion of the detector, and thus these correction algorithms are not needed for centrally located fields of moderate size (at least, in the case of 6 MV beam energy).Conclusion: The 2D correction improves the portal dosimetry results for those fields for which the 1D correction proves insufficient, especially in the inplane, off-axis regions of the detector. This 2D correction neglects the relatively smaller discrepancies that may be caused by backscatter from nonuniform machine components downstream from the detecting layer.

60 APPLIED LIFE SCIENCES↗