Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data requirements”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Rapid Adaptation of Chemical Named Entity Recognition Using Few-Shot Learning and LLM Distillation

Named entity recognition (NER) has been widely used in chemical text mining for the automatic identification and extraction of chemical entities. However, existing chemical NER systems primarily focus on scenarios with abundant training data, requiring significant human effort on annotations. This poses challenges for applications in the chemical field, such as catalysis, where many advancements have traditionally relied on trial-and-error investigations and incremental adjustment of variables. This hinders catalysis science and technology progress in addressing emerging energy and environmental crises. In this work, we propose a few-shot NER model that can quickly adapt to extract new types of chemical entities by using only a limited number of annotated examples. Our model employs a metric-learning approach to transfer entity similarity knowledge from high-resource chemical domains (with abundant annotations) to enable effective entity recognition in low-resource specialized domains (limited annotation). We validate the effectiveness of our model on a few-shot chemical NER benchmark built based on six existing chemical NER data sets. Experiments show that the proposed few-shot NER model can achieve reasonable performance with only 5 examples per entity type and shows consistent improvement as the number of examples increases. Furthermore, we demonstrate how the proposed model can be trained with large language model (LLM) annotated data, opening a new pathway for rapid adaptation of NER systems. Furthermore, our approach leverages the knowledge broadness of large language models for chemistry while distilling this knowledge into a lightweight model suitable for efficient and in-house use.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Brief Overview of the NCSP [Slides]

The mission of the NCSP is to provide sustainable expert leadership, direction, and the technical infrastructure necessary to develop, maintain, and disseminate the essential technical tools, training, and data required to support safe, efficient fissionable material operations within the Department of Energy. The vision of the NCSP is a continually improving, adaptable, and transparent program that communicates and collaborates globally to incorporate technology, practices, and programs to be responsive to the essential technical needs of those responsible for developing, implementing, and maintaining nuclear criticality safety.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Adaptive machine learning for time-varying systems: low dimensional latent space tuning

Machine learning (ML) tools such as encoder-decoder convolutional neural networks (CNN) can represent incredibly complex nonlinear functions which map between combinations of images and scalars. For example, CNNs can be used to map combinations of accelerator parameters and images which are 2D projections of the 6D phase space distributions of charged particle beams as they are transported between various particle accelerator locations. Despite their strengths, applying ML to time-varying systems, or systems with shifting distributions, is an open problem, especially for large systems for which collecting new data for re-training is impractical or interrupts operations. Particle accelerators are one example of large time-varying systems for which collecting detailed training data requires lengthy dedicated beam measurements which may no longer be available during regular operations. We present a novel method of adaptive ML for time-varying systems. Our approach is to map very high (N ≈ 100k) dimensional inputs (a combination of scalar parameters and images) into the low dimensional (N ≈ 2) latent space at the output of the encoder section of an encoder-decoder CNN. We then actively tune the low dimensional latent space-based representation of complex system dynamics by the addition of an adaptively tuned feedback vector directly before the decoder sections builds back up to our image-based high-dimensional phase space density representations. This method allows us to learn correlations within and to quickly tune the characteristics of incredibly large parameter space systems and to track their evolution in real time based on feedback without massive new data sets for re-training. We demonstrate that our method can accurately predict and track the phase space of charged particle beams at various locations in a particle accelerator by adaptively adjusting in real-time while the unknown input beam distribution of the accelerator is changing in shape, charge, and offset and while the RF system of the accelerator itself is also changing in an unpredictable way. For FACET-II we demonstrate that such an approach has the potential to use transverse deflecting cavity and energy spread spectrum beam measurements to accurately predict 2D projections of the 6D phase space of the electron beam at the plasma wakefield acceleration interaction point where such diagnostics are unavailable.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

Measuring higher-order moments of neutron-time-of-flight data for cryogenic inertial confinement fusion implosions on OMEGA

Ion temperatures serve as an important diagnostic for inertial confinement fusion (ICF) implosions. In direct-drive ICF experiments on OMEGA, neutron-time-of-flight (nTOF) data are used to infer the ion temperature of the fusing plasma produced in the implosion experiment. The analysis of the nTOF data requires an assumption about the shape of the underlying source signal. Since the source nTOF signal is a near-replica of the neutron energy spectrum, an ideal Gaussian shape, corresponding to the neutron energy spectrum of a uniform temperature plasma, is routinely employed. However, spatial and temporal variations of the ion temperature in the plasma give rise to higher-order moments, which were first described by Munro [Nucl. Fusion 56, 036001 (2016)]. In this work, we show a simpler alternative analysis to derive moments of the neutron energy spectrum for a plasma with variations in ion temperature. We also present a revised analysis of measured nTOF signals that uses a model with an additional degree of freedom to take into account the effect of ion temperature variations on the shape of the spectrum. Compared to presently used nTOF analysis, the revised analysis yields on average ≈2× more accurate fits to the data and up to 15% higher ion temperatures for cryogenic experiments. As a result, we quantify the ion temperature inflation caused by radially symmetric fluid flows, which are present even in a symmetric implosion, and which serve as a lower bound on the ion temperature inflation in real implosions.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Quantifying heavy quark transport coefficients with an improved transport model

We report that the heavy-flavor transport coefficients contain important information on the strong interaction at finite temperatures. The extraction of these numbers from experimental data requires dynamical modeling of heavy-flavor transport that is coupled to realistic medium evolution. Furthermore, meaningful extractions necessitate both a faithful implementation of the physical inputs to be tested and the quantification of model uncertainty. For these purposes, we have developed a partonic transport model LIDO. It has an improved treatment of in-medium parton bremsstrahlung, which has been calibrated to theoretical calculations in a simple medium to reduce modeling uncertainty. Regarding the interaction between heavy quark and the medium, few-body perturbative scatterings are applied to large-momentum transfer (q) processes, while a diffusion equation models the dynamics of small-q processes. Such a separation restricts the explicit use of medium quasi-particles to large-q processes only. Another advantage is that deviations from the leading-order probe-medium coupling can be parametrized as an additional contribution to the diffusion constant. The heavy quark transport coefficients are then extracted with uncertainty estimation from a Bayesian analysis including both the RHIC and the LHC data. The results are found to be consistent with earlier extraction of the light-quark transport coefficients at high momentum and be comparable with lattice calculations of the heavy-flavor diffusion constant in the static limit at low momentum.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Technological aspects in blanket design: Effects of micro-alloying and thermo-mechanical treatments of EUROFER97 type steels after neutron irradiation

Presently available data on neutron irradiation damage raise doubts on the feasibility of using EUROFER97 steel for a water-cooled starter blanket in a DEMO reactor, since the ductile-to-brittle transition temperature (DBTT) increases significantly for irradiation temperatures below 350°C. The additional DBTT shift caused by H and He transmutation can only be estimated based on very few results with isotopically tailored EUROFER97 steel. Conservative calculations show that the DBTT of EUROFER97 steel could exceed the operating temperature in water-cooled starter blankets within a relatively short time period. This paper presents results from a EUROfusion funded irradiation campaign that was performed in the High Flux Isotope Reactor at Oak Ridge National Laboratory. The paper compares ten newly developed reduced activation ferritic-martensitic (RAFM) steels irradiated to a nominal dose of 2.5 dpa at 300°C. The post-irradiation experiments using Small Specimen Test Technology included hardness, tensile, and fracture mechanics tests combined with fractography and microstructure analysis are presented. Results show that micro-alloying EUROFER97-type steels influenced the mechanical properties but a dominating impact on irradiation damage resistance could not be identified. In contrast, specific thermo-mechanical treatments lead to better DBTT behavior. Discussion about irradiation response to heat treatment conditions is also given. Here, despite requiring data also at high dpa values, the results indicate that with these modified materials an increased lifetime and potentially also an increased operating temperature window can be achieved compared to EUROFER97.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Identification of Unique Fragmentation Patterns of Fentanyl Analog Protomers Using Structures for Lossless Ion Manipulations Ion Mobility-Orbitrap Mass Spectrometry

The opioid crisis in the United States is being fueled by the rapid emergence of new fentanyl analogs and precursors that can elude traditional library-based screening methods, which require data from known reference compounds. Since reference compounds are unavailable for new fentanyl analogs, we examined if fentanyls (fentanyl + fentanyl analogs) could be identified in a reference-free manner using a combination of electrospray ionization (ESI), high-resolution ion mobility (IM) spectrometry, high-resolution mass spectrometry (MS), and higher-energy collision-induced dissociation (MS/MS). We analyzed a mixture containing nine fentanyls and W-15 (a structurally similar molecule) and found that the protonated forms of all fentanyls uniquely exhibited two baseline separated IM distributions that produced different MS/MS patterns. Upon fragmentation, both IM distributions of all fentanyls produced two high intensity fragments resulting from amine site cleavages. The higher mobility distributions of all fentanyls also produced several low intensity fragments, but surprisingly, these same fragments exhibited much greater intensities in the lower mobility distributions. This observation demonstrates that many fragments of fentanyls predominantly originate from one of two different gas-phase structures (suggestive of protomers). Furthermore, increasing the water concentration in the ESI solution increased the intensity of the lower mobility distribution relative to the higher mobility distribution, which further supports that fentanyls exist as two gas-phase protomers. In conclusion, our new observations on the IM and MS/MS properties of fentanyls can be exploited to positively identify them as fentanyls without requiring reference libraries and will hopefully assist first responders and law enforcement in combating new and emerging fentanyls.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Semi-Analytical Hierarchical Bayesian Inference of Nonlinear Model Structure in Stochastic Dynamics: Applied to Compartmental Models of Infectious Diseases

A Bayesian computational framework for parsimonious inference in stochastic nonlinear dynamical systems is presented. This framework enables the concurrent estimation of system states, time-varying parameters, time-invariant parameters, and the optimal sparsity structure of the model parameters. Because differential equation-based models are often simplified mechanistic or phenomenological representations, robust inference from noisy measurement data requires explicit treatment of model error and uncertainty. Model error and time-varying parameters can be represented as random processes, enabling inference while making minimal assumptions about the underlying sources of discrepancy and variability. Adopting stochastic differential equation representations affords the model significant flexibility, but can also render it susceptible to overfitting during statistical inversion, where the inferred model may track noise rather than the underlying signal. To alleviate the effects of overfitting and to enable the discovery of the optimal sparse representation of the time-invariant parameters, a Bayesian sparse learning algorithm is embedded within the framework. This sparse learning framework adopts an approximate hierarchical Bayesian setting defined by a series of semi-analytical expressions. The model structure inference framework is validated using a stochastic compartmental model for tracking and forecasting active cases of an infectious disease. Compartmental models describe population-level infectious disease dynamics through interactions among population fractions grouped by disease state. Mathematically, such models consist of a system of coupled ordinary differential equations. This example adopts an expressive compartmental model that includes multiple possible interactions between disease states, motivated by early uncertainty surrounding COVID-19 reinfection dynamics and their implications for long-term epidemic forecasting. The sparse learning exercise permits the inference of a priori unknown epidemiological dynamics from simulated public health data, discovering the nested compartmental model that optimizes the trade-off between average data-fit and model complexity. It is shown that inducing sparsity among the model parameters eliminates redundant interactions between compartments, equivalently revealing the optimal coupling structure between differential equations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Bayesian chain graph models to characterize microbe-environment dynamics

Microbiome data require statistical models that can simultaneously decode microbes' reaction to the environment and interactions among microbes. While a multiresponse linear regression model seems like a straight-forward solution, we argue that treating it as a graphical model is problematic given that the regression coefficient matrix does not encode the conditional dependence structure between response and predictor nodes. This observation is especially important in biological settings when we have prior knowledge on the edges from specific experimental interventions that can only be properly encoded under a conditional dependence model. Here, we propose a chain graph model with two sets of nodes (predictors and responses) whose solution yields a graph with edges that indeed represent conditional dependence, thus agreeing with the experimenter's intuition on the average behavior of nodes under treatment. The solution to our model is sparse via the Bayesian linear regression (LASSO). In addition, we propose an adaptive extension so that different shrinkages can be applied to different edges to incorporate edge-specific prior knowledge. Our model is computationally inexpensive through an efficient Gibbs sampling algorithm and can account for binary, counting, and compositional responses via an appropriate hierarchical structure. We test the performance of our model in a variety of simulated datasets, thereby showing superior performance to state-of-the-art approaches. We further apply our model to human gut and soil microbial compositional datasets, and we highlight that CG-LASSO can estimate biologically meaningful network structures in the data.

compositional data↗
DOE OSTI2026

Data for: A hybrid biophysical-machine learning framework for diurnal surface energy flux estimation using proximal sensing

Thermal-based remote sensing of surface energy fluxes has traditionally relied on high spatial resolution satellite data with revisit frequencies on the order of weeks. In this study, we evaluate a biophysics-based analytical surface energy balance model for predicting latent energy (LE) and sensible heat (H) fluxes using proximal sensing observations. The Surface Temperature Initiated Closure (STIC1.2) model has been extensively validated across a wide range of spatial and temporal scales using various satellite-derived thermal datasets. Here we extend this validation by applying STIC at sub-hourly temporal resolution over multiple growing seasons for four distinct agricultural systems. We further develop and evaluate novel STIC variants that incorporate machine learning (ML) techniques to eliminate the need for specific surface energy balance observations, specifically net radiation and soil heat flux, thereby enhancing model applicability in data-sparse settings. The integration of an ML component to estimate surface available energy is shown to have strong predictive performance for both LE (R2 = 0.81-0.94) and H (R2 = 0.46-0.72) across all agricultural systems examined here, demonstrating the potential of hybrid biophysical – machine learning approaches for surface energy balance modeling with minimal data requirements. This study concludes with a novel application of explainable machine learning (exML) to diagnose sources of model error. This exML framework attributes residual prediction errors to both model input variables and environmental drivers not explicitly included in the simulation experiments. This approach provides a new pathway for improving model design and integrating previously overlooked yet influential variables into future model iterations.

Agricultural Sciences↗
DOE OSTI2025

High‐Asymmetry Metasurface: A New Solution for Terahertz Resonance via Active Learning‐Augmented Diffusion Model

Terahertz (THz) metamaterials with high‐figure‐of‐merit (high‐FoM) performance resonance are essential for advancing sensors, detectors, and imagers. Conventional designs focus on symmetric or low‐asymmetry geometric structures, leaving high‐asymmetry designs largely unexplored due to the inefficiency of trial‐and‐error‐based rational design. Recent deep learning techniques offer automation and acceleration but are constrained by the need for large datasets inherent to their data‐driven nature. Here, a novel prior knowledge‐guided generative model augmented by a physics‐constrained active learning mechanism to design high‐asymmetry metamaterials. An advanced diffusion model learns features from a small set of classical structures with high‐FoM THz resonance and generates new high‐asymmetry structures. To mitigate the limited number of classical structures, the generated high‐asymmetry structures are actively selected and integrated into the initial training dataset based on their physical characteristics. Experimental results demonstrate the superior resonance performance of the generated high‐asymmetry metamaterials over classical designs, exhibiting improvements exceeding 30% in key resonance metrics. Remarkably, this performance is attained using only 68 classical structures as the initial training dataset, significantly reducing the data requirements for deep learning‐based metamaterial design. The proposed scheme for generating high‐asymmetry structures provides a new effective and efficient solution for high‐FoM resonance, expanding applications in high‐sensitivity THz metadevices.

diffusion model↗
DOE OSTI2024

Imitating the “breeder's eye”: Predicting grain yield from measurements of non‐yield traits

Abstract Plant breeding relies on information gathered from field trials to select promising new crop varieties for release to farmers and to develop genomic prediction models that can enhance the efficiency of genetic improvement in future breeding cycles. However, generating the genetic marker data required to apply genomic prediction at the early stages of a breeding program remains costly for many public‐sector breeding programs as well as for many plant breeders operating in developing countries. As the pace of climate change intensifies, the time lag of developing and deploying new crop varieties requires plant breeders to make selection decisions without knowing the future environments those crop varieties will encounter in farmers’ fields. Therefore, both lower cost and higher accuracy methods for prediction of crop performance are essential for creating and maintaining resilient agricultural systems in the latter half of the 21 st century. To address this challenge, we conducted linked yield trials of 752 public maize ( Zea mays ) genotypes in two distinct environments. We developed and trained a phenotypic prediction model to predict yield from manually scored plant traits. The phenotypic prediction approach we employed outperformed genomic prediction in predicting yields in a second environment, with 8.7%–63% higher R 2 and 4%–13% less root mean square error than the genomic prediction. The phenotypic prediction has the potential to be applied to a wider range of breeding programs, including those that lack the resources to genotype large populations, such as programs in the developing world, breeding programs for specialty crops, and public sector programs.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Predicting chatter using machine learning and acoustic signals from low-cost microphones

Machining chatter is a phenomenon resulting from self-oscillation between a machining tool and workpiece. This self-oscillation results in variation on the machined product that reduces the ability to meet desired specifications. Chatter is a widely studied topic as it directly relates to the quality of machined products. Here, this study details the application of a Random Forest (RF) classifier with Recursive Feature Elimination (RFE) to machining audio collected by a single microphone during down-milling operations. This approach allows straightforward feature elimination that results in an easily understood set of analyzed dimensions. Stability is predicted solely based on the classification output of the RF classifier. Our approach proves highly predictive with consistent machining setup and a small sample set. We also review transferability between machining setups and present key findings. Our RF approach demonstrates the ability to analyze and classify chatter through a low-cost approach with limited training data required. The motivation for using a single microphone is to enable detection on machines without other sensors, such as accelerometers, present in the machining setup. The value of the in-process sensor and chatter classifier is highlighted because the machining setup included asymmetric dynamics that reduced the accuracy of the traditional analytical stability solution. We see a natural progression to deploying this audio-only methodology with real-time processing and classification using either a laptop or smartphone. This progression will allow visual indicators during the machining process that can alert machinists of progression into unstable machining processes.

42 ENGINEERING↗
DOE OSTI2024

Phase Composition and Phase Transformation of Additively Manufactured Nickel Alloy 718 AM Bench Artifacts

Additive manufacturing (AM) technologies offer unprecedented design flexibility but are limited by a lack of understanding of the material microstructure formed under their extreme and transient processing conditions and its subsequent transformation during post-build processing. As part of the 2022 AM Bench Challenge, sponsored by the National Institute of Standards and Technology, this study focuses on the phase composition and phase evolution of AM nickel alloy 718, a nickel-based superalloy, to provide benchmark data essential for the validation of computational models for microstructural predictions. Here, we employed high-energy synchrotron X-ray diffraction, in situ synchrotron X-ray scattering, as well as high-resolution transmission electron microscopy for our analyses. The study uncovers critical aspects of the microstructure in its as-built state, its transformation during homogenization, and its phase evolution during subsequent aging heat treatment. Specifically, we identified secondary phases, monitored the dissolution and coarsening of microstructural elements, and observed the formation and stability of γ ’ and γ ” phases. The results provide the rigorous benchmark data required to understand the atomic and microstructural transformations of AM nickel alloy 718, thereby enhancing the reliability and applicability of AM models for predicting phase evolution and mechanical properties.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Deep operator network surrogate for phase-field modeling of metal grain growth during solidification

A deep operator network (DeepONet) has been constructed that generates accurate representations of phase-field model simulations for evolving two dimensional metal grain morphology growing from melt. These representations serve as lower resolution, computationally efficient stand-ins for quick parameter space exploration of solutions to the the Allen-Cahn equations that dictate the phase-field model simulations. The experimental target for the phase-field model is a uranium casting system cooling a 434 g uranium charge from a maximum temperature of 1400° C at an average rate of 30° C / min , traversing the crystallographic phases of the pure metal. Experimental parameters inform the phase-field model, whose higher resolution computational model solutions are used to train the DeepONet in a given parameter space with the aim of developing a faster, more efficient method for predicting the solidifying metal's microstructure at different potential experimental values. The final DeepONet generates high accuracy, lower resolution predictions with cumulative relative approximation error over all timesteps of less than 0.5%, while ensuring solutions remain within physically feasible ranges. Further, these relative error values are comparable with other state-of-the-art DeepONet models for microstructure evolution, while significantly reducing the amount of training data required. Training a convolutional neural network simultaneously with the DeepONet, enforcing realistic values at the complex metal grain boundaries, and mathematically encoding boundary conditions into the structure of the DeepONet improved prediction accuracy and computational efficiency over a standard DeepONet model.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Effects of different vegetation drag parameterizations on the tidal propagation in coastal marshlands

Vegetation drag is a fundamental quantity directly affecting results for both long- and short-term coastal marsh and geomorphological studies. The vegetation drag in coastal marshland has been modeled by various two-dimensional (2D) and threedimensional (3D) numerical parameterizations. 2D parameterizations treat coastal marshes as bottom roughness elements, while 3D parameterizations resolve the vertically-variable vegetation drag through the water column. However, differences in tidal propagation arising from different drag parameterizations within a single model are largely unknown, and clear guidance on parameterization selection is still missing. In this study, we implemented four vegetation drag parameterizations into the Model for Prediction Across Scales-Ocean (MPAS-O), which include 1) a 2D parameterization using land-cover type-determined Manning’s n (2DLM); 2) a 2D parameterization using vegetation-determined Manning’s n (2DVM); 3) a 3D parameterization for stiff vegetation (3DSV); and 4) a 3D parameterization for flexible vegetation (3DFV). Estimates of the flow resistance effects from these parameterizations were compared using a series of idealized tidal propagation simulations. Given the same tidal condition, flooding depth and flooding distance are the largest in the 2DLM simulations and the smallest in the 3DSV simulations. 2DVM results are the closest to the 2DLM results. 3DFV results are the closest to the average of 2DVM, 3DSV, and 3DFV results. 2DVM and 3DSV results are the least and most sensitive to the vegetation aboveground biomass, respectively. Based on the input data requirement and computational efficiency of each parameterization, a comparison summary is provided to help inform parameterization selection for specific applications. Here, the effects of these parameterizations on coastal geomorphology are further discussed, and the results demonstrate that estimates of the long-term evolution of coastal marshes and coastal morphology depend upon the selection of the vegetation drag parameterization

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Using X-ray spectrum imaging to quantify fission gas in high-burnup UO2 fuel

Analysis of fission gas bubbles (FGBs) in light-water reactor (LWR) fuel is needed to improve the understanding and predictive capability of fuel evolution under normal- and transient conditions. One of the most important parameters of a FGB is the pressure of the gas, primarily Xe, inside it. Bubble volume and location (inter- vs. intragranular) are important considerations as well. However, experimental analysis of such bubbles is challenging due to their small size and embedded nature, and usefulness of the data requires large numbers be analyzed. This paper proposes a method to measure the pressure of Xe bubbles using X-ray spectrum imaging (XSI) in scanning transmission electron microscopy (STEM). From X-ray generation and instrumental parameters, the number of Xe atoms yielding a given number of Xe L-series X-ray counts is estimated, and then from the estimated volume of the bubble the Xe density is estimated, which can then be converted to pressure via an equation-of-state. We apply the method to XSI datasets from high burnup (HBu) fuel from North Anna 1 reactor in as-irradiated and post-LOCA-test conditions and find Xe pressures in the FGBs clustered around 1 GPa.

Parish, Chad [ORNL] (ORCID:0000000312097439)↗
DOE OSTI2025

Fast-RF-Shimming: Accelerate RF shimming in 7T MRI using deep learning

Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) offers an elevated signal-to-noise ratio (SNR), enabling exceptionally high spatial resolution that benefits both clinical diagnostics and advanced research. However, the jump to higher fields introduces complications, particularly transmit radiofrequency (RF) field ($B^{+}_{1}$) inhomogeneities, manifesting as uneven flip angles and image intensity irregularities. These artifacts can degrade image quality and impede broader clinical adoption. Traditional RF shimming methods, such as Magnitude Least Squares (MLS) optimization, effectively mitigate $B^{+}_{1}$ inhomogeneity, but remain time-consuming. Recent machine learning approaches, including RF Shim Prediction by Iteratively Projected Ridge Regression and other deep learning architectures, suggest alternative pathways. Although these approaches show promise, challenges such as extensive training periods, limited network complexity, and practical data requirements persist. In this paper, we introduce a holistic learning-based framework called Fast-RF-Shimming, which achieves a 5000 ​× ​speed-up compared to the traditional MLS method. In the initial phase, we employ random-initialized Adaptive Moment Estimation (Adam) to derive the desired reference shimming weights from multi-channel $B^{+}_{1}$ fields. Next, we train a Residual Network (ResNet) to map $B^{+}_{1}$ fields directly to the ultimate RF shimming outputs, incorporating the confidence parameter into its loss function. Finally, we design Non-uniformity Field Detector (NFD), an optional post-processing step, to ensure the extreme non-uniform outcomes are identified. Comparative evaluations with standard MLS optimization underscore notable gains in both processing speed and predictive accuracy, which indicates that our technique shows a promising solution for addressing persistent inhomogeneity challenges.

Deep learning↗