Search NASA⌕ Search

SEARCH · Search NASA

Results for “model interpretability”

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 91 records · Page 5

Evaluation of Deep Learning Model Architectures for Point-of-Care Ultrasound Diagnostics

Point-of-care ultrasound imaging is a critical tool for patient triage during trauma for diagnosing injuries and prioritizing limited medical evacuation resources. Specifically, an eFAST exam evaluates if there are free fluids in the chest or abdomen but this is only possible if ultrasound scans can be accurately interpreted, a challenge in the pre-hospital setting. In this effort, we evaluated the use of artificial intelligent eFAST image interpretation models. Widely used deep learning model architectures were evaluated as well as Bayesian models optimized for six different diagnostic models: pneumothorax (i) B- or (ii) M-mode, hemothorax (iii) B- or (iv) M-mode, (v) pelvic or bladder abdominal hemorrhage and (vi) right upper quadrant abdominal hemorrhage. Models were trained using images captured in 27 swine. Using a leave-one-subject-out training approach, the MobileNetV2 and DarkNet53 models surpassed 85% accuracy for each M-mode scan site. The different B-mode models performed worse with accuracies between 68% and 74% except for the pelvic hemorrhage model, which only reached 62% accuracy for all model architectures. These results highlight which eFAST scan sites can be easily automated with image interpretation models, while other scan sites, such as the bladder hemorrhage model, will require more robust model development or data augmentation to improve performance. With these additional improvements, the skill threshold for ultrasound-based triage can be reduced, thus expanding its utility in the pre-hospital setting.

47 OTHER INSTRUMENTATION↗

Laboratory Experiments and Modeling for Interpreting Field Studies of Secondary Organic Aerosol Formation Using an Oxidation Flow Reactor

This grant was originally funded for deployment of a suite of aerosol instrumentation by our group in collaboration with other research groups and DOE/ARM to the Ganges Valley in India (GVAX) to study aerosols sources and processing. Much of the first year of this grant was focused on preparations for GVAX. That campaign was cancelled due to political reasons and with the consultation with our program manager, the research of this grant was refocused to study the applications of oxidation flow reactors (OFRs) for investigating secondary organic aerosol (SOA) formation and organic aerosol (OA) processing in the field and laboratory through a series of laboratory and modeling studies. We developed a gas-phase photochemical model of an OFR which was used to 1) explore the sensitivities of key output variables (e.g., OH exposure, O 3 , HO 2 /OH) to controlling factors (e.g., water vapor, external reactivity, UV irradiation), 2) develop simplified OH exposure estimation equations, 3) investigate under what conditions non-OH chemistry may be important, and 4) help guide design of future experiments to avoid conditions with undesired chemistry for a wide range of conditions applicable to the ambient, laboratory, and source studies. Uncertainties in the model were quantified and modeled OH exposure was compared to tracer decay measurements of OH exposure in the lab and field. Laboratory studies using OFRs were conducted to explore aerosol yields and composition from anthropogenic and biogenic VOC as well as crude oil evaporates. Various aspects of the modeling and laboratory results and tools were applied to interpretation of ambient and source measurements using OFR. Additionally, novel measurement methods were used to study gas/particle partitioning. The research conducted was highly successful and details of the key results are summarized in this report through narrative text, figures, and a complete list of publications acknowledging this grant.

54 ENVIRONMENTAL SCIENCES↗

Model-independent versus model-dependent interpretation of the SDSS-III BOSS power spectrum: Bridging the divide

The traditional clustering analyses of galaxy redshift surveys compress the clustering data into a set of late-time physical variables in a model-independent way. This approach has recently been extended by an additional shape variable encoding early-time physics information. Here we apply this new technique, ShapeFit, to SDSS-III BOSS data and show that it matches the constraining power of alternative, model-dependent approaches, which directly constrain the model’s parameters adopting a cosmological model ab initio. ShapeFit is ~30 times faster, model independent, naturally splits early- and late-time variables, and enables a better control of observational systematics.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A Model-based Interpretation of Low-frequency Changes in the Carbon Cycle during the Last 120,000 years and its Implications for the Reconstruction of Atmospheric (delta) 14-C

A main caveat in the interpretation of observed changes in atmospheric (Delta)C-l4 during the last 50,000 years is the unknown variability of the carbon cycle, which together with changes in the C-14 production rates determines the C-14 dynamics. A plausible scenario explaining glacial/interglacial dynamics seen in atmospheric CO2 and (delta)C-13 was proposed recently (Kohler et al., 2005a). A similar approach that expands its interpretation to the C-14 cycle is an important step toward a deeper understanding of (Delta)C-14 variability. This approach is based on an ocean/atmosphere/biosphere box model of the global carbon cycle (BICYCLE) to reproduce low-frequency changes in atmospheric CO2 as seen in Antarctic ice cores. The model is forced forward in time by various paleoclimatic records derived from ice and sediment cores. The simulation results of our proposed scenario match a compiled CO2 record from various ice cores during the last 120,000 years with high accuracy (r(sup 2) = 0.89). We analyze scenarios with different C-14 production rates, which are either constant or based on Be-10 measured in Greenland ice cores or the recent high-resolution geomagnetic field reconstruction GLOPIS-75 and compare them with the available (Delta)C-14 data covering the last 50,000 years. Our results suggest that during the last glacial cycle in general less than 110%0o f the increased atmospheric (Delta)C-14 is based on variations in the carbon cycle, while the largest part (5/6) of the variations has to be explained by other factors. Glacial atmospheric (Delta)C-14 larger than 700% cannot not be explained within our framework, neither through carbon cycle-based changes nor through variable C-14 production. Superimposed on these general trends might lie positive anomalies in atmospheric (Delta)C-14 of approx. 50% caused by millennial-scale variability of the northern deep water production during Heinrich events and Dansgaard/Oeschger climate fluctuations. According to our model, the dominant processes that increase glacial (Delta)C-14 are a reduced glacial ocean circulation (+ approx.40%0), a restricted glacial gas exchange between the atmosphere and the surface ocean through sea ice coverage (+ approx. 20%), and the enrichment of dissolved inorganic carbon with C-14 in the surface waters through isotopic fractionation during higher glacial marine export production caused by iron fertilization (+ approx.10%).

Koehler, Peter↗

Charmonium in lattice QCD and the non-relativistic quark model

We compare the results of a numerical lattice QCD calculation of the charmonium spectrum with the structure of a general non-relativistic potential model. To achieve this we form the non-relativistic reduction of derivative-based fermion bilinear interpolating fields used in lattice QCD calculations and compute their overlap with $c\bar{c}$ meson states at rest constructed in the non-relativistic quark model, providing a bound-state model interpretation for the lattice data. Essential gluonic components in the bound-states, usually called hybrids, are identified by considering interpolating fields that involve the gluonic field-strength tensor and which have zero overlap onto simple $c\bar{c}$ model states.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A statistical model for interpreting computerized dynamic posturography data

Computerized dynamic posturography (CDP) is widely used for assessment of altered balance control. CDP trials are quantified using the equilibrium score (ES), which ranges from zero to 100, as a decreasing function of peak sway angle. The problem of how best to model and analyze ESs from a controlled study is considered. The ES often exhibits a skewed distribution in repeated trials, which can lead to incorrect inference when applying standard regression or analysis of variance models. Furthermore, CDP trials are terminated when a patient loses balance. In these situations, the ES is not observable, but is assigned the lowest possible score--zero. As a result, the response variable has a mixed discrete-continuous distribution, further compromising inference obtained by standard statistical methods. Here, we develop alternative methodology for analyzing ESs under a stochastic model extending the ES to a continuous latent random variable that always exists, but is unobserved in the event of a fall. Loss of balance occurs conditionally, with probability depending on the realized latent ES. After fitting the model by a form of quasi-maximum-likelihood, one may perform statistical inference to assess the effects of explanatory variables. An example is provided, using data from the NIH/NIA Baltimore Longitudinal Study on Aging.

NASA Discipline Neuroscience↗

Deep learning models for interpretation of point of care ultrasound in military working dogs

Introduction: Military working dogs (MWDs) are essential for military operations in a wide range of missions. With this pivotal role, MWDs can become casualties requiring specialized veterinary care that may not always be available far forward on the battlefield. Some injuries such as pneumothorax, hemothorax, or abdominal hemorrhage can be diagnosed using point of care ultrasound (POCUS) such as the Global FAST® exam. This presents a unique opportunity for artificial intelligence (AI) to aid in the interpretation of ultrasound images. In this article, deep learning classification neural networks were developed for POCUS assessment in MWDs. Methods: Images were collected in five MWDs under general anesthesia or deep sedation for all scan points in the Global FAST® exam. For representative injuries, a cadaver model was used from which positive and negative injury images were captured. A total of 327 ultrasound clips were captured and split across scan points for training three different AI network architectures: MobileNetV2, DarkNet-19, and ShrapML. Gradient class activation mapping (GradCAM) overlays were generated for representative images to better explain AI predictions. Results: Performance of AI models reached over 82% accuracy for all scan points. The model with the highest performance was trained with the MobileNetV2 network for the cystocolic scan point achieving 99.8% accuracy. Across all trained networks the diaphragmatic hepatorenal scan point had the best overall performance. However, GradCAM overlays showed that the models with highest accuracy, like MobileNetV2, were not always identifying relevant features. Conversely, the GradCAM heatmaps for ShrapML show general agreement with regions most indicative of fluid accumulation. Discussion: Overall, the AI models developed can automate POCUS predictions in MWDs. Preliminarily, ShrapML had the strongest performance and prediction rate paired with accurately tracking fluid accumulation sites, making it the most suitable option for eventual real-time deployment with ultrasound systems. Further integration of this technology with imaging technologies will expand use of POCUS-based triage of MWDs.

59 BASIC BIOLOGICAL SCIENCES↗

Model‐Based Interpretation of Solute Exports and Carbon Partitioning During Shale Weathering in a Mountainous Hillslope

The weathering of sedimentary rocks in high-elevation catchments influences freshwater quality and the global carbon cycle. While individual biogeochemical mechanisms involved in this process are relatively well understood, quantifying their contributions to solute export and carbon fluxes under natural, transient conditions remains challenging. Here, we implement a numerical multidimensional and multiphase model to simulate coupled hydrological and biogeochemical processes in a shale-underlain, snow-dominated hillslope in the Rocky Mountains, Colorado. The model captures the dynamic interplay between soil respiration, mineral weathering, and climate-driven hydrological forcing, reproducing observed soil CO 2 dynamics, groundwater chemistry, and subsurface flow. Our results reveal that seasonal snowmelt enhances carbonate weathering by promoting the infiltration of CO 2 -rich water to depth, while pyrite oxidation is primarily sensitive to low water saturation that facilitates O 2 diffusion through the regolith. Topography modulates the spatial distribution of shale weathering, as steeper slopes enhance lateral drainage, favoring the delivery of reactants to greater depths. While shale weathering at our site acts as a transient carbon sink, with silicates and carbonates buffering acidity and promoting atmospheric CO 2 consumption (1% of soil-derived CO 2 ), the exported dissolved inorganic carbon is predominantly geogenic (∼73%). Consequently, when accounting for long-term marine carbonate precipitation. The current weathering regime represents a net source of carbon to the atmosphere. The oxidation of pyrite and petrogenic organic carbon together release approximately 0.9 mol·m −2 ·yr −1 of CO 2 . Our findings highlight the role of topography, hydroclimate, and the coupling between acid-base reactions in shaping the carbon balance and the solute exports in mountainous critical zones.

carbon cycling↗

The Northwest Geysers EGS Demonstration Project, California. Pre-stimulation Modeling and Interpretation of the Stimulation

The Northwest Geysers Enhanced Geothermal System (EGS) demonstration project aims to create an EGS by directly and systematically injecting cool water at relatively low pressure into a known High Temperature (280–400 °C) Zone (HTZ) located under the conventional (240 °C) geothermal steam reservoir at The Geysers geothermal field in California. Here we report that , the results of coupled thermal, hydraulic, and mechanical (THM) analyses made using a model developed as part of the pre-stimulation phase of the EGS demonstration project is presented. The model simulations were conducted in order to investigate injection strategies and the resulting effects of cold-water injection upon the EGS system; in particular to predict the extent of the stimulation zone for a given injection schedule. The actual injection began on October 6, 2011, and in this paper a comparison of pre-stimulation model predictions with micro-earthquake (MEQ) monitoring data over the first few months of a one-year injection program is presented. The results show that, by using a calibrated THM model based on historic injection and MEQ data at a nearby well, the predicted extent of the stimulation zone (defined as a zone of high MEQ density around the injection well) compares well with observed seismicity. The modeling indicates that the MEQ events are related to shear reactivation of preexisting fractures, which is triggered by the combined effects of injection-induced cooling around the injection well and small changes in steam pressure as far as half a kilometer away from the injection well. Pressure-monitoring data at adjacent wells and satellite-based ground-surface deformation data were also used to validate and further calibrate reservoir-scale hydraulic and mechanical model properties. The pressure signature monitored from the start of the injection was particularly useful for a precise back-calculation of reservoir porosity. Ultimately, the first few months of reservoir pressure and surface deformation data were useful for estimating the reservoir-rock permeability and elastic modulus. Finally, although the extent of the calculated stimulation zone matches the field observations over the first few months of injection, the observed surface deformations and MEQ evolution showed more heterogeneous behavior as a result of more complex geology, including minor faults and fracture zones that are important for consideration in the analysis of energy production and the long-term evolution of the EGS system.

58 GEOSCIENCES↗

Ground-state oxygen holes and the metal–insulator transition in the negative charge-transfer rare-earth nickelates

The metal-insulator transitions and the intriguing physical properties of rare-earth perovskite nickelates have attracted considerable attention in recent years. However, a complete understanding of these materials remains elusive. Here, taking a NdNiO 3 thin film as a representative example, we utilize a combination of x-ray absorption (XAS) and resonant inelastic x-ray scattering (RIXS) spectroscopies to resolve important aspects of the complex electronic structure of the rare-earth nickelates. The unusual coexistence of bound and continuum excitations observed in the RIXS spectra provides strong evidence for the abundance of oxygen 2p holes in the ground state of these materials. Using cluster calculations and Anderson impurity model interpretation, we show that these distinct spectral signatures arise from a Ni 3d 8 configuration along with holes in the oxygen 2p valence band, confirming suggestions that these materials do not obey a “conventional” positive charge-transfer picture, but instead exhibit a negative charge-transfer energy, in line with recent models interpreting the metal to insulator transition in terms of bond disproportionation.

36 MATERIALS SCIENCE↗

Machine Learning for Improving Surface-Layer-Flux Estimates

Abstract Flows in the atmospheric boundary layer are turbulent, characterized by a large Reynolds number, the existence of a roughness sublayer and the absence of a well-defined viscous layer. Exchanges with the surface are therefore dominated by turbulent fluxes. In numerical models for atmospheric flows, turbulent fluxes must be specified at the surface; however, surface fluxes are not known a priori and therefore must be parametrized. Atmospheric flow models, including global circulation, limited area models, and large-eddy simulation, employ Monin–Obukhov similarity theory (MOST) to parametrize surface fluxes. The MOST approach is a semi-empirical formulation that accounts for atmospheric stability effects through universal stability functions. The stability functions are determined based on limited observations using simple regression as a function of the non-dimensional stability parameter representing a ratio of distance from the surface and the Obukhov length scale (Obukhov in Trudy Inst Theor Geofiz AN SSSR 1:95–115, 1946), $$z/L$$ z / L . However, simple regression cannot capture the relationship between governing parameters and surface-layer structure under the wide range of conditions to which MOST is commonly applied. We therefore develop, train, and test two machine-learning models, an artificial neural network (ANN) and random forest (RF), to estimate surface fluxes of momentum, sensible heat, and moisture based on surface and near-surface observations. To train and test these machine-learning algorithms, we use several years of observations from the Cabauw mast in the Netherlands and from the National Oceanic and Atmospheric Administration’s Field Research Division tower in Idaho. The RF and ANN models outperform MOST. Even when we train the RF and ANN on one set of data and apply them to the second set, they provide more accurate estimates of all of the fluxes compared to MOST. Estimates of sensible heat and moisture fluxes are significantly improved, and model interpretability techniques highlight the logical physical relationships we expect in surface-layer processes.

Meteorology & Atmospheric Sciences↗

USING REALISTIC MHD SIMULATIONS FOR THE MODELING AND INTERPRETATION OF QUIET-SUN OBSERVATIONS WITH THE SOLAR DYNAMICS OBSERVATORY HELIOSEISMIC AND MAGNETIC IMAGER

The solar atmosphere is extremely dynamic, and many important phenomena develop on small scales that are unresolved in observations with the Helioseismic and Magnetic Imager (HMI) instrument on the Solar Dynamics Observatory. For correct calibration and interpretation of the observations, it is very important to investigate the effects of small-scale structures and dynamics on the HMI observables, such as Doppler shift, continuum intensity, spectral line depth, and width. We use 3D radiative hydrodynamics simulations of the upper turbulent convective layer and the atmosphere of the Sun, and a spectro-polarimetric radiative transfer code to study observational characteristics of the Fe i 6173 Å line observed by HMI in quiet-Sun regions. We use the modeling results to investigate the sensitivity of the line Doppler shift to plasma velocity, and also sensitivities of the line parameters to plasma temperature and density, and determine effective line formation heights for observations of solar regions located at different distances from the disk center. These estimates are important for the interpretation of helioseismology measurements. In addition, we consider various center-to-limb effects, such as convective blueshift, variations of helioseismic travel-times, and the “concave” Sun effect, and show that the simulations can qualitatively reproduce the observed phenomena, indicating that these effects are related to a complex interaction of the solar dynamics and radiative transfer.

79 ASTRONOMY AND ASTROPHYSICS↗

Using snowflake surface-area-to-volume ratio to model and interpret snowfall triple-frequency radar signatures

Abstract. The snowflake microstructure determines the microwave scattering properties of individual snowflakes and has a strong impact on snowfall radar signatures. In this study, individual snowflakes are represented by collections of randomly distributed ice spheres where the size and number of the constituent ice spheres are specified by the snowflake mass and surface-area-to-volume ratio (SAV) and the bounding volume of each ice sphere collection is given by the snowflake maximum dimension. Radar backscatter cross sections for the ice sphere collections are calculated at X-, Ku-, Ka-, and W-band frequencies and then used to model triple-frequency radar signatures for exponential snowflake size distributions (SSDs). Additionally, snowflake complexity values obtained from high-resolution multi-view snowflake images are used as an indicator of snowflake SAV to derive snowfall triple-frequency radar signatures. The modeled snowfall triple-frequency radar signatures cover a wide range of triple-frequency signatures that were previously determined from radar reflectivity measurements and illustrate characteristic differences related to snow type, quantified through snowflake SAV, and snowflake size. The results show high sensitivity to snowflake SAV and SSD maximum size but are generally less affected by uncertainties in the parameterization of snowflake mass, indicating the importance of snowflake SAV for the interpretation of snowfall triple-frequency radar signatures.

54 ENVIRONMENTAL SCIENCES↗

Interpretative transport modeling of the WEST boundary plasma: main plasma and light impurities

Understanding impurity transport in tokamak plasmas is crucial to control radiative losses and material migration in future magnetic fusion reactors. In this work we deploy the SolEdge2D-EIRENE code to model the boundary plasma in a WEST discharge, satisfactorily reproducing measurements of both upstream and divertor plasma conditions. Additionally, the spatial distribution of oxygen, studied here as a representative light impurity, is compared to vacuum ultraviolet spectroscopy measurements acquired with an oscillating line of sight. The simulation captures a key feature of the experiment, namely a factor of sime2 higher oxygen brightness in the inner divertor region compared to the outer one. This spatial asymmetry in oxygen concentration is interpreted by analyzing the balance of friction forces and thermal gradient forces that the light impurity exchanges with the main plasma.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Toward Smart, Automated Junctional Tourniquets—AI Models to Interpret Vessel Occlusion at Physiological Pressure Points

Hemorrhage is the leading cause of preventable death in both civilian and military medicine. Junctional hemorrhages are especially difficult to manage since traditional tourniquet placement is often not possible. Ultrasound can be used to visualize and guide the caretaker to apply pressure at physiological pressure points to stop hemorrhage. However, this process is technically challenging, requiring the vessel to be properly positioned over rigid boney surfaces and applying sufficient pressure to maintain proper occlusion. As a first step toward automating this life-saving intervention, we demonstrate an artificial intelligence algorithm that classifies a vessel as patent or occluded, which can guide a user to apply the appropriate pressure required to stop flow. Neural network models were trained using images captured from a custom tissue-mimicking phantom and an ex vivo swine model of the inguinal region, as pressure was applied using an ultrasound probe with and without color Doppler overlays. Using these images, we developed an image classification algorithm suitable for the determination of patency or occlusion in an ultrasound image containing color Doppler overlay. Separate AI models for both test platforms were able to accurately detect occlusion status in test-image sets to more than 93% accuracy. In conclusion, this methodology can be utilized for guiding and monitoring proper vessel occlusion, which, when combined with automated actuation and other AI models, can allow for automated junctional tourniquet application.

60 APPLIED LIFE SCIENCES↗

Shrinkage and Swelling of Coal Induced by Desorption and Sorption of Fluids: Theoretical Model and Interpretation of a Field Project

Geologic sequestration in deep unmineable coal seams and enhanced coalbed methane production is a promising choice, economically and environmentally, to reduce anthropogenic gases such as carbon dioxide in the atmosphere. Unmineable coal seams are typically known to adsorb large amounts of carbon dioxide in comparison to the sizeable amounts of sorbed methane, which raises the potential for large scale sequestration projects. During the process of sequestration, carbon dioxide is injected into the coalbed and desorbed methane is produced. The coal matrix is believed to shrink when a gas is desorbed and swell when a gas is sorbed, sometimes causing profound changes in the cleat porosity and permeability of the coal seam. These changes may have significant impact on the reservoir performance. Therefore, it is necessary to understand the combined influence of swelling and shrinkage, and geomechanical properties including elastic modulus, cleat porosity, and permeability of the reservoir. The present paper deals with the influence of swelling and shrinkage on the reservoir performance, and the geomechanical response of the reservoir system during the process of geologic sequestration of carbon dioxide and enhanced coalbed methane production in an actual field project located in northern New Mexico. A three-dimensional swelling and shrinkage model was developed and implemented into an existing reservoir model to understand the influence of geomechanical parameters, as well as swelling and shrinkage properties, on the reservoir performance. Numerical results obtained from the modified simulator were compared to available measured values from that site and previous studies. Results show that swelling and shrinkage, and the combination of geomechanical and operational parameters, have a significant influence on the performance of the reservoir system.

01 COAL, LIGNITE, AND PEAT↗