Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “INTERPOLATION”

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 577 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

Power system disturbance location determination based on rate of change of frequency

A method for determining a location of a disturbance in a power system is provided. The method includes receiving data from a plurality of sensors distributed across the power system; determining, with one or more processors, for each sensor, an extremum of the data corresponding to the sensor; interpolating, with the one or more processors, the extrema; and presenting, on one or more display units, a location corresponding to a global maximum of the interpolated extrema as the location of the disturbance. The method may further include, if a difference between the largest extremum and each of remaining extremum is greater than a predetermined threshold, skipping the interpolating step and the presenting step, and presenting a location of the sensor corresponding to the largest extremum as the location of the disturbance.

You, Shutang↗
DOE OSTI2020

Development and Validation of Passive Yaw in the Open-Source WEC-Sim Code: Preprint

A passive yaw implementation is developed, validated, and explored for the Wave Energy Converter Simulator (WEC-Sim), an open-source wave energy converter modeling tool that works within MATLAB/Simulink. The Reference Model 5 (RM5) is selected for this investigation, and a WEC-Sim model of the device is modified to allow yaw motion. A boundary element method (BEM) code was used to calculate the excitation force coefficients for a range of wave headings. An algorithm was implemented in WEC-Sim to determine the equivalent wave heading from a body’s instantaneous yaw angle and interpolate the appropriate excitation coefficients to ensure the correct time-domain excitation force. This approach is able to determine excitation force for a body undergoing large yaw displacement. For the mathematically simple case of regular wave excitation, the dynamic equation was integrated numerically and found to closely approximate the results from this implementation in WEC-Sim. A case study is presented for the same device in irregular waves. In this case, computation time is increased by 32x when this interpolation is performed at every time step. To reduce this expense, a threshold yaw displacement can be set to reduce the number of interpolations performed. A threshold of 0.01 was found to increase computation time by only 22x without significantly affecting time domain results. Similar amplitude spectra for yaw force and displacements are observed for all threshold values less than or equal to 1, for which computation time is only increased by 2.2x.

50 EE - Wind and Water Power Program - Water (EE-4↗
DOE OSTI2023

Dynamic Model Agnostic Reliability Evaluation of Machine-Learning Models Integrated in Instrumentation & Control Systems

In recent years, the field of data-driven neural network-based machine learning (ML) algorithms has grown significantly and spurred research in its applicability to instrumentation and control systems. While they are promising in operational contexts, the trustworthiness of such algorithms is not adequately assessed. Failures of ML-integrated systems are poorly understood; the lack of comprehensive risk modeling can degrade the trustworthiness of these systems. In recent reports by the National Institute for Standards and Technology, trustworthiness in ML is a critical barrier to adoption and will play a vital role in intelligent systems' safe and accountable operation. Thus, in this work, we demonstrate a real-time model-agnostic method to evaluate the relative reliability of ML predictions by incorporating out-of-distribution detection on the training dataset. It is well documented that ML algorithms excel at interpolation (or near-interpolation) tasks but significantly degrade at extrapolation. This occurs when new samples are "far" from training samples. The method, referred to as the Laplacian distributed decay for reliability (LADDR), determines the difference between the operational and training datasets, which is used to calculate a prediction's relative reliability. LADDR is demonstrated on a feedforward neural network-based model used to predict safety significant factors during different loss-of-flow transients. LADDR is intended as a "data supervisor" and determines the appropriateness of well-trained ML models in the context of operational conditions. Ultimately, LADDR illustrates how training data can be used as evidence to support the trustworthiness of ML predictions when utilized for conventional interpolation tasks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Generating An Advanced Cross-section Library For HTGR Pebble Bed Depletion Calculations Using Reduced-Order Model Generation Techniques

For code development, Advanced Reactor Technologies - Gas Cooled Reactors Program (ART-GCR) rely on a collaboration with the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program, but the cross sections generation and the methodology definition is part of this program area goals. Based on previous studies in FY23, the size of microscopic cross section libraries increases rapidly with the number of tabulations, requiring significant amount of memory and drastically slowing down the Griffin calculations when evaluating cross sections via the multivariate linear interpolation approach. Rising to these challenges, this work investigates constructing Reduced-order Models (ROMs) for the multi-group microscopic cross sections to accelerate the cross section evaluation in Griffin. A database of multigroup cross sections is first collected considering all possible parameters that a designer could change for optimization. Down-selection of the ROM techniques afterward shows Deep Neural Network (DNN) as the best candidate when jointly consider memory efficiency, predictive accuracy, computational cost, scalability, flexibility and ease of implementation of the algorithms in comparison to the multidimensional interpolation. This work develops a specific interface that enables the cross section predictions using pre-trained DNN models into Griffin leveraging the existing ROM capabilities. DNNs have been trained for all isotopes for use in Griffin. Preliminary Griffin testing shows that DNNs exhibit exceptional predictive accuracy and the use of DNNs provides orders of magnitude improvement in memory efficiency compared to conventional interpolation techniques. With such ROM techniques, it holds great promise to further increase the fidelity of the Pebble Bed Reactor (PBR) simulation by increasing the number of tabulations/state variables during cross section evaluation, while maintaining the computational cost affordable in Griffin.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Reduce-Order Modeling of Multigroup Neutron Cross Sections for High-Temperature Gas-cooled Reactors

Deterministic neutronics calculations rely on multigroup neutron cross section libraries, which usually consists of a database of tabulated values, used to calculate the cross sections through multivariate linear interpolation. However, interpolation of the multidimensional cross section data becomes memory inefficient and time consuming as the number of tabulations increases, significantly slowing down the neutronics calculation, especially in the case of micro cross section libraries where every isotope (on the order of hundreds) has its own set of specific reactions and cross sections. To address this challenge, this work constructs efficient and robust reduced-order models (ROMs) of the multi-group cross sections to support the Griffin simulation of high-temperature gas-cooled reactors (HTGRs). The first part of the study investigates the linearity of the multi-group cross section data across isotopes, reaction types and energy groups on pre-generated datasets for the purpose of dimensionality reduction. Secondly, a down-selection of ROM techniques is presented on representative classical machine learning (ML) techniques, including variants of linear regression, kernel-based methods, tree-based algorithms, and artificial neural networks. The selection criteria jointly consider the memory efficiency, predictive accuracy, prediction speed, and scalability in comparison to the multidimensional interpolation. Among all the ML techniques, deep neural networks (DNNs) have proven to be the best selection with sufficient accuracy, high robustness, good memory efficiency, great scalability, and superior flexibility. DNNs for have been trained for all isotopes in this work and systematic Griffin testing is ongoing at this moment to ensure the feasibility of this ROM technique for cross section predictions.

42 - ENGINEERING↗
DOE OSTI2024

Reduced-Order Modeling of Multigroup Neutron Cross Sections for High-Temperature Gas-cooled Reactors

Abstract – Deterministic neutronics calculations rely on multigroup neutron cross section libraries, which consist of databases of tabulated values, used to calculate the neutron cross sections through multivariate linear interpolation. However, interpolation of the multidimensional cross section data becomes memory inefficient and time consuming as the number of tabulations increases, significantly slowing down the neutronics calculation, especially in the case of microscopic cross section libraries where every isotope (on the order of hundreds) has its own set of specific reactions and cross sections. In order to address this challenge, this work constructs efficient and robust reduced-order models (ROMs) of the multi-group cross sections to support the Griffin simulation of high-temperature gas-cooled reactors (HTGRs). The first part of the study investigates the linearity of the multigroup cross section data across isotopes, reaction types, and energy groups on pre-generated datasets for the purpose of dimensionality reduction. Secondly, a down-selection of ROM techniques is presented on representative classical machine learning (ML) techniques, including variants of linear regression, kernel-based methods, tree-based algorithms, and artificial neural networks. The selection criteria jointly consider the memory efficiency, predictive accuracy, prediction speed, and scalability in comparison to the multidimensional interpolation. Among all the ML techniques, deep neural networks (DNNs) have proven to be the best selection with sufficient accuracy, high robustness, good memory efficiency, great scalability, and superior flexibility. DNNs have been trained for all isotopes in this work and systematic Griffin testing is ongoing to ensure the feasibility of this ROM technique for predicting cross section and reducing memory requirements without a significant sacrifice in computational performance.

42 - ENGINEERING↗
DOE OSTI2024

Advanced Cross Section Library Generation using Reduced Order Models

Deterministic neutronics calculations rely on multigroup neutron cross section libraries, which consist of databases of tabulated values, used to calculate the neutron cross sections through multivariate linear interpolation. However, interpolation of the multidimensional cross section data becomes memory inefficient and time consuming as the number of tabulations increases, significantly slowing down the neutronics calculation, especially in the case of microscopic cross section libraries where every isotope (on the order of hundreds) has its own set of specific reactions and cross sections. In order to address this challenge, this work constructs efficient and robust reduced-order models (ROMs) of the multi-group cross sections to support the Griffin simulation of high-temperature gas-cooled reactors (HTGRs). The first part of the study investigates the linearity of the multigroup cross section data across isotopes, reaction types, and energy groups on pre-generated datasets for the purpose of dimensionality reduction. Secondly, a down-selection of ROM techniques is presented on representative classical machine learning (ML) techniques, including variants of linear regression, kernel-based methods, tree-based algorithms, and artificial neural networks. The selection criteria jointly consider the memory efficiency, predictive accuracy, prediction speed, and scalability in comparison to the multidimensional interpolation. Among all the ML techniques, deep neural networks (DNNs) have proven to be the best selection with sufficient accuracy, high robustness, good memory efficiency, great scalability, and superior flexibility. DNNs have been trained for all isotopes in this work and systematic Griffin testing is ongoing to ensure the feasibility of this ROM technique for predicting cross section and reducing memory requirements without a significant sacrifice in computational performance.

42 - ENGINEERING↗
DOE OSTI2018

Superconvergence analysis for nonlinear parabolic equation with EQ{sub 1}{sup rot} nonconforming finite element

EQ{sub 1}{sup rot} nonconforming finite element method (FEM) applied to a class of nonlinear parabolic equation is discussed. First, by use of two typical characters of this element (one is that the associated FE interpolation operator is identical to its traditional Ritz projection operator; the other is that the consistency error is of order O(h{sup 2}), one order higher than the interpolation error, when the exact solution of the problems belongs to H{sup 3}(Ω)), the supercloseness of order O(h{sup 2}) in broken H{sup 1}-norm for semi-discrete scheme is obtained without the boundedness of numerical solution in L{sup ∞}-norm through splitting the numerical solution into several parts. Moreover, we get the desired result with requirement of u,u{sub t}∈H{sup 3}(Ω) only. Secondly, a linearized Crank–Nicolson fully discrete scheme is proposed and the superclose property of order O(h{sup 2}+τ{sup 2}) is derived by constructing a suitable auxiliary problem. Finally, a numerical example is carried out to confirm our theoretical analysis. Here, h is the subdivision parameter and τ is the time step.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

New discretization of Caputo–Fabrizio derivative

We derive a numerical approximation, namely L1–2 formula, to the Caputo–Fabrizio derivative by using a quadratic interpolation. Quadratic and cubic convergence rates are achieved for L1 and L1–2 formulas using Lagrange interpolation, respectively. We compute Caputo–Fabrizio derivatives of some known functions both theoretically and numerically. In addition, we solve non/linear sub-diffusion equations to test theoretical findings. Numerical results confirm the theoretically observed convergence rates.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

H{sub 2}{sup +}, HeH and H{sub 2}: Approximating potential curves, calculating rovibrational states

Analytic consideration of the Bohr–Oppenheimer (BO) potential curves for diatomic molecules is proposed: accurate analytic interpolation for a potential curve consistent with its rovibrational spectra is found. It is shown that in the BO approximation for four lowest electronic states 1sσ{sub g} and 2pσ{sub u}, 2pπ{sub u} and 3dπ{sub g} of H{sub 2}{sup +}, the ground state X{sup 2}Σ{sup +} of HeH and the two lowest states {sup 1}Σ{sub g}{sup +} and {sup 3}Σ{sub u}{sup +} of H{sub 2}, the potential curves can be analytically interpolated in full range of internuclear distances R with not less than 4–5–6 s.d. Approximation based on matching the Laurant-type expansion at small R and a combination of the multipole expansion with one-instanton type contribution at large distances R is given by two-point Padé approximant. The position of minimum, when exists, is predicted within 1% or better. For the molecular ion H{sub 2}{sup +} in the Lagrange mesh method, the spectra of vibrational, rotational and rovibrational states (ν,L) associated with 1sσ{sub g} and 2pσ{sub u}, 2pπ{sub u} and 3dπ{sub g} potential curves are calculated. In general, it coincides with spectra found via numerical solution of the Schrödinger equation (when available) within six s.d. It is shown that 1sσ{sub g} curve contains 19 vibrational states (ν,0), while 2pσ{sub u} curve contains a single one (0,0) and 2pπ{sub u} state contains 12 vibrational states (ν,0). In general, 1sσ{sub g} electronic curve contains 420 rovibrational states, which increases up to 423 when we are beyond BO approximation. For the state 2pσ{sub u} the total number of rovibrational states (all with ν=0) is equal to 3, within or beyond Bohr–Oppenheimer approximation. As for the state 2pπ{sub u} within the Bohr–Oppenheimer approximation the total number of the rovibrational bound states is equal to 284. The state 3dπ{sub g} is repulsive, no rovibrational state is found. It is confirmed in Lagrange mesh formalism the statement that the ground state potential curve of the heteronuclear molecule HeH does not support rovibrational states. Accurate analytical expression for the potential curves of the hydrogen molecule H{sub 2} for the states {sup 1}Σ{sub g}{sup +} and {sup 3}Σ{sub u}{sup +} is presented. The ground state {sup 1}Σ{sub g}{sup +} contains 15 vibrational states (ν,0),ν=0–14. In general, this state supports 301 rovibrational states. The potential curve of the state {sup 3}Σ{sub u}{sup +} has a shallow minimum: it does not support any rovibrational state, it is repulsive.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Optimizing shift selection in multilevel Monte Carlo for disconnected diagrams in lattice QCD

The calculation of disconnected diagram contributions to physical signals is a computationally expensive task in Lattice QCD. To extract the physical signal, the trace of the inverse Lattice Dirac operator, a large sparse matrix, must be stochastically estimated. Because the variance of the stochastic estimator is typically large, variance reduction techniques must be employed. Multilevel Monte Carlo (MLMC) methods reduce the variance of the trace estimator by utilizing a telescoping sequence of estimators. Frequency Splitting is one such method that uses a sequence of inverses of shifted operators to estimate the trace of the inverse lattice Dirac operator, however there is no a priori way to select the shifts that minimize the cost of the multilevel trace estimation. Here we present a sampling and interpolation scheme that is able to predict the variances associated with Frequency Splitting under displacements of the underlying space time lattice. The interpolation scheme is able to predict the variances to high accuracy and therefore chooses shifts that correspond to an approximate minimum of the cost for the trace estimation. We show that Frequency Splitting with the chosen shifts displays significant speedups over multigrid deflation, and that these shifts can be used for multiple configurations within the same ensemble with no penalty to performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Data-driven models of nonautonomous systems

Nonautonomous dynamical systems are characterized by time-dependent inputs, which complicates the discovery of predictive models describing the spatiotemporal evolution of the state variables of quantities of interest from their temporal snapshots. When dynamic mode decomposition (DMD) is used to infer a linear model, this difficulty manifests itself in the need to approximate the time-dependent Koopman operators. Our approach is to approximate the original nonautonomous system with a modified system derived via a local parameterization of the time-dependent inputs. The modified system comprises a sequence of local parametric systems, which are subsequently approximated by a parametric surrogate model using the DRIPS (dimension reduction and interpolation in parameter space) framework. The offline step of DRIPS relies on DMD to build a linear surrogate model, endowed with reduced-order bases for the observables mapped from training data. The online step interpolates on suitable manifolds to construct a sequence of iterative parametric surrogate models; the target/test parameter points on these manifolds are specified by a local parameterization of the test time-dependent inputs. Here, we use numerical experimentation to demonstrate the robustness of our method and compare its performance with that of deep neural networks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Soft x-ray spectrum unfold of K-edge filtered x-ray diode arrays using cubic splines

Cubic spline interpolation is able to recover temporally and spectrally resolved soft x-ray flux from an array of K-edge filtered x-ray diodes without the need for a priori assumptions about the spectrum or the geometry of the emitting volume. The mathematics of the cubic spline interpolation is discussed in detail. The analytic nature of the cubic spline solution allows for analytical error propagation, and the method of calculating the error for radiation temperature, spectral power, and confidence intervals of the unfolded spectrally resolved flux is explained. An unfold of a blackbody model demonstrates the accuracy of the cubic spline unfold. Tests of cubic spline performance using spectrally convolved detailed atomic model simulation results have been performed to measure the method's ability to conserve spectral power to within a factor of 2 or better in line-dominated regimes. The unfold is also demonstrated to work when information from the x-ray diode array is limited due to high signal-to-noise ratios or lack of signal due to over-attenuation or over-filtration of the x-ray diode signal. Here, the robustness of the unfold with respect to background subtraction and raw signal processing, signal alignment between diode traces, limited signal information, and initial conditions is discussed. Results from an example analysis of a halfraum drive are presented to demonstrate the capabilities of the unfold in comparison with previously established methods.

47 OTHER INSTRUMENTATION↗
DOE OSTI2015

Better alternatives to “astronomical silicate”: Laboratory-based optical functions of chondritic/solar abundance glass with application to HD 161796

“Astronomical” or “circumstellar” silicate optical functions (real and imaginary indices of refraction n(λ) and k(λ)) have previously been derived from compositionally and structurally disparate samples; past values were compiled from different sources in the literature, and are essentially kluges of observational, laboratory, and extrapolated or interpolated values. These synthetic optical functions were created because astronomers lack the quantitative data on amorphous silicates at all wavelengths needed for radiative transfer modeling. This paper provides optical functions that (1) are created with a consistent methodology, (2) use the same sample across all wavelengths, and (3) minimize interpolation and extrapolation wherever possible. We present electronic data tables of optical functions derived from mid-ultraviolet to far-infrared (FIR) laboratory transmission spectra for two materials: iron-free glass with chondritic/solar atmospheric abundances, and metallic iron. We compare these optical functions to other popular n, k data used to model amorphous silicates (e.g., “astronomical” or “circumstellar” silicate), both directly and in application to a simple system: the dust shell of the post-AGB star HD 161796. Using the new optical functions, we find that the FIR profile of model spectral energy distributions are significantly affected by the ratio of glass to iron. Our case study on HD 161796 shows that in modeling with our new optical functions, the mineralogy is markedly different from that derived using synthetic optical functions and suggests a new scenario of crystalline silicate formation.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Cosmic microwave background constraints for global strings and global monopoles

We present the first cosmic microwave background (CMB) power spectra from numerical simulations of the global O( N ) linear σ-model, with N =2,3, which have global strings and monopoles as topological defects. In order to compute the CMB power spectra we compute the unequal time correlators (UETCs) of the energy-momentum tensor, showing that they fall off at high wave number faster than naive estimates based on the geometry of the defects, indicating non-trivial (anti-)correlations between the defects and the surrounding Goldstone boson field. We obtain source functions for Einstein-Boltzmann solvers from the UETCs, using a recently developed method that improves the modelling at the radiation-matter transition. We show that the interpolation function that mimics the transition is similar to other defect models, but not identical, confirming the non-universality of the interpolation function. The CMB power spectra for global strings and global monopoles have the same overall shape as those obtained using the non-linear σ-model approximation, which is well captured by a large- N calculation. However, the amplitudes are larger than the large- N calculation would naively predict, and in the case of global strings much larger: a factor of 20 at the peak. Finally we compare the CMB power spectra with the latest CMB data in other to put limits on the allowed contribution to the temperature power spectrum at multipole l = 10 of 1.7% for global strings and 2.4% for global monopoles. These limits correspond to symmetry-breaking scales of 2.9× 10{sup 15} GeV (6.3× 10{sup 14} GeV with the expected logarithmic scaling of the effective string tension between the simulation time and decoupling) and 6.4× 10{sup 15} GeV respectively. The bound on global strings is a significant one for the ultra-light axion scenario with axion masses m {sub a} ∼< 10{sup −28} eV . These upper limits indicate that gravitational waves from global topological defects will not be observable at the gravitational wave observatory LISA.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2015

SU-E-T-485: In Vivo Dosimetry with EBT3 Radiochromic Films for TBI Treatments

Purpose: Total body irradiation (TBI) is a technique that requires special equipment to control “in vivo” the dose to the patient because it is a complex technique performed in extraordinary conditions. There are several devices to perform this task (diodes, TLDs, ionization chambers, MOSFET). In this paper we study the possibility of performing these measurements with radiochromic films EBT3 properly calibrated. This method has been compared to the PTW diodes system for TBI. Methods: Once made the TC to the patients, we measured different thicknesses of the relevant areas of the body (head, neck, chest with or without arms, umbilicus area, knees and ankles); for each of these thicknesses we measured dose rate (cGy / UM) in RW3 phantom, in TBI conditions, with ionization chamber in the center; in turn, the input diode and the output of each configuration is placed to assign dose to each set of diodes. Movie calibration is performed according to manufacturer’s recommendations but TBI conditions. The dose at the center of each thickness compared to a linear interpolation of the dose at the entrance and exit, resulting in an adequate approximation. Finally in each session for each patient put a piece of film (2×2 cm2) at the entrance and another at the exit in each area, obtaining these readings and interpolating the estimated center dose, as with the diodes. Results: These results show a greater homogeneity in the distribution for use with film and validate the use of the same for this task and, if necessary, to avoid purchasing diode group if they have not. Conclusion: By using radiochromic films for this technique gives us a proper calculation of the dose received by the patient in the absence of other methods, or gives us a second additional track that already used normally.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

SU-F-BRA-10: Fricke Dosimetry: Determination of the G-Value for Ir-192 Energy Based On the NRC Methodology

Purpose: Use the methodology developed by the National Research Council Canada (NRC), for Fricke Dosimetry, to determine the G-value used at Ir-192 energies. Methods: In this study the Radiology Science Laboratory of Rio de Janeiro State University (LCR),based the G-value determination on the NRC method, using polyethylene bags. Briefly, this method consists of interpolating the G-values calculated for Co-60 and 250 kV x-rays for the average energy of Ir-192 (380 keV). As the Co-60 G-value is well described at literature, and associated with low uncertainties, it wasn’t measured in this present study. The G-values for 150 kV (Effective energy of 68 keV), 250 kV (Effective energy of 132 keV)and 300 kV(Effective energy of 159 keV)were calculated using the air kerma given by a calibrated ion chamber, and making it equivalent to the absorbed to the Fricke solution, using a Monte Carlo calculated factor for this conversion. Instead of interpolations, as described by the NRC, we displayed the G-values points in a graph, and used the line equation to determine the G- value for Ir-192 (380 keV). Results: The measured G-values were 1.436 ± 0.002 µmol/J for 150 kV, 1.472 ± 0.002 µmol/J for 250 kV, 1.497 ± 0.003 µmol/J for 300 kV. The used G-value for Co-60 (1.25 MeV) was 1,613 µmol/J. The R-square of the fitted regression line among those G-value points was 0.991. Using the line equation, the calculate G-value for 380 KeV was 1.542 µmol/J. Conclusion: The Result found for Ir-192 G-value is 3,1% different (lower) from the NRC value. But it agrees with previous literature results, using different methodologies to calculate this parameter. We will continue this experiment measuring the G-value for Co-60 in order to compare with the NRC method and better understand the reasons for the found differences.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2016

Assessment of metal artifact reduction methods in pelvic CT

Purpose: Metal artifact reduction (MAR) produces images with improved quality potentially leading to confident and reliable clinical diagnosis and therapy planning. In this work, the authors evaluate the performance of five MAR techniques for the assessment of computed tomography images of patients with hip prostheses. Methods: Five MAR algorithms were evaluated using simulation and clinical studies. The algorithms included one-dimensional linear interpolation (LI) of the corrupted projection bins in the sinogram, two-dimensional interpolation (2D), a normalized metal artifact reduction (NMAR) technique, a metal deletion technique, and a maximum a posteriori completion (MAPC) approach. The algorithms were applied to ten simulated datasets as well as 30 clinical studies of patients with metallic hip implants. Qualitative evaluations were performed by two blinded experienced radiologists who ranked overall artifact severity and pelvic organ recognition for each algorithm by assigning scores from zero to five (zero indicating totally obscured organs with no structures identifiable and five indicating recognition with high confidence). Results: Simulation studies revealed that 2D, NMAR, and MAPC techniques performed almost equally well in all regions. LI falls behind the other approaches in terms of reducing dark streaking artifacts as well as preserving unaffected regions (p < 0.05). Visual assessment of clinical datasets revealed the superiority of NMAR and MAPC in the evaluated pelvic organs and in terms of overall image quality. Conclusions: Overall, all methods, except LI, performed equally well in artifact-free regions. Considering both clinical and simulation studies, 2D, NMAR, and MAPC seem to outperform the other techniques.

60 APPLIED LIFE SCIENCES↗