Search NASA⌕ Search

SEARCH · Search NASA

Results for “Model Counting”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

The Next Generation Virgo Cluster Survey. XXVIII. Characterization of the Galactic White Dwarf Population

We use three different techniques to identify hundreds of white dwarf (WD) candidates in the Next Generation Virgo Cluster Survey (NGVS) based on photometry from the NGVS and GUViCS, and proper motions derived from the NGVS and the Sloan Digital Sky Survey (SDSS). Photometric distances for these candidates are calculated using theoretical color–absolute magnitude relations, while effective temperatures are measured by fitting their spectral energy distributions. Disk and halo WD candidates are separated using a tangential velocity cut of 200 km s{sup −1} in a reduced proper motion diagram, which leads to a sample of six halo WD candidates. Cooling ages, calculated for an assumed WD mass of 0.6 M {sub ⊙}, range between 60 Myr and 6 Gyr, although these estimates depend sensitively on the adopted mass. Luminosity functions for the disk and halo subsamples are constructed and compared to previous results from the SDSS and SuperCOSMOS survey. We compute a number density of (2.81 ± 0.52) × 10{sup −3} pc{sup −3} for the disk WD population—consistent with previous measurements. We find (7.85 ± 4.55) × 10{sup −6} pc{sup −3} for the halo, or 0.3% of the disk. Observed stellar counts are also compared to predictions made by the TRILEGAL and Besançon stellar population synthesis models. The comparison suggests that the TRILEGAL model overpredicts the total number of WDs. The WD counts predicted by the Besançon model agree with the observations, although a discrepancy arises when comparing the predicted and observed halo WD populations; the difference is likely due to the WD masses in the adopted model halo.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Analysis of overlapping count data

Counts of a specific characteristic were obtained within regions defined on an object that was manufactured in a proprietary setting. The count regions were altered during production and resulted in misaligned or overlapping count data. A closed-formula maximum likelihood estimator (MLE) of the new region means is derived using all of the available count data and an independent Poisson model. The MLE is shown to be preferable to estimators constructed using generalized linear models for the overlapping data setting. This closed-form estimator extends to over-dispersed overlapping count data as the quasi-MLE and also performs well with correlated overlapping count data. Standard errors for the estimator are approximated and are validated with a simulation study. Additionally, the methods are extended to overlapping multinomial data. Illustrative examples of the methods are provided throughout the paper and are reproducible with the supplemental R code. Additionally, proofs of the paper’s results are also included in the supplemental material.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

WE-DE-207B-01: Optimization for Contrast-Enhanced Spectral Mammography Based On Photon-Counting Detectors

Purpose: To investigate the feasibility of optimizing the imaging parameters for contrast-enhanced spectral mammography based on Si strip photon-counting detectors. Methods: A computer simulation model using polyenergetic spectra from a tungsten anode x-ray tube and a Si-based photon-counting detector was evaluated for contrast-enhanced spectral mammography. The simulation traces the emission of photons from the x-ray source, attenuation through the breast and subsequent absorption in the detector. The breast was modeled as a mixture of adipose and mammary gland tissues with a breast density of 30%. A 4 mm iodine signal with a concentration of 4 mg/ml was used to simulate the enhancement of a lesion. Quantum efficiency of the detector was calculated based on the effective attenuation length in the Si strips. The figure-of-merit (FOM), which was defined as the decomposed iodine signal-to-noise ratio (SNR) with respect to the square root of the mean glandular dose (MGD), was chosen to optimize the imaging parameters, in terms of beam energy, splitting energy, and pre-filtrations for breast of various thicknesses and densities. Results: The optimal imaging parameters, which lead to the highest FOM, were found at a beam energy of 45 kVp with a splitting energy at 34 keV for an averaged breast thickness of 4 cm with a standard 0.75 mm Al pre-filtration. The optimal tube voltage varied slightly from 46 to 44 kVp as the breast thickness increases from 2 to 8 cm. The optimal tube voltage decreased to 42 kVp when the Al pre-filtration was increased to 3 mm. Conclusion: This simulation study predicted the optimal imaging parameters for application of photon-counting spectral mammography to contrast-enhanced imaging. The simulation results laid the ground work for future phantom and clinical studies. Grant funding from Philips Medical Systems.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Deep Learning Approach for High-accuracy Electron Counting of Monolithic Active Pixel Sensor-type Direct Electron Detectors at Increased Electron Dose

Abstract Electron counting can be performed algorithmically for monolithic active pixel sensor direct electron detectors to eliminate readout noise and Landau noise arising from the variability in the amount of deposited energy for each electron. Errors in existing counting algorithms include mistakenly counting a multielectron strike as a single electron event, and inaccurately locating the incident position of the electron due to lateral spread of deposited energy and dark noise. Here, we report a supervised deep learning (DL) approach based on Faster region-based convolutional neural network (R-CNN) to recognize single electron events at varying electron doses and voltages. The DL approach shows high accuracy according to the near-ideal modulation transfer function (MTF) and detector quantum efficiency for sparse images. It predicts, on average, 0.47 pixel deviation from the incident positions for 200 kV electrons versus 0.59 pixel using the conventional counting method. The DL approach also shows better robustness against coincidence loss as the electron dose increases, maintaining the MTF at half Nyquist frequency above 0.83 as the electron density increases to 0.06 e−/pixel. Thus, the DL model extends the advantages of counting analysis to higher dose rates than conventional methods.

Materials Science↗
DOE OSTI2012

CORRELATIONS IN THE (SUB)MILLIMETER BACKGROUND FROM ACT Multiplication-Sign BLAST

We present measurements of the auto- and cross-frequency correlation power spectra of the cosmic (sub)millimeter background at 250, 350, and 500 {mu}m (1200, 860, and 600 GHz) from observations made with the Balloon-borne Large Aperture Submillimeter Telescope (BLAST); and at 1380 and 2030 {mu}m (218 and 148 GHz) from observations made with the Atacama Cosmology Telescope (ACT). The overlapping observations cover 8.6 deg{sup 2} in an area relatively free of Galactic dust near the south ecliptic pole. The ACT bands are sensitive to radiation from the cosmic microwave background, to the Sunyaev-Zel'dovich effect from galaxy clusters, and to emission by radio and dusty star-forming galaxies (DSFGs), while the dominant contribution to the BLAST bands is from DSFGs. We confirm and extend the BLAST analysis of clustering with an independent pipeline and also detect correlations between the ACT and BLAST maps at over 25{sigma} significance, which we interpret as a detection of the DSFGs in the ACT maps. In addition to a Poisson component in the cross-frequency power spectra, we detect a clustered signal at 4{sigma}, and using a model for the DSFG evolution and number counts, we successfully fit all of our spectra with a linear clustering model and a bias that depends only on redshift and not on scale. Finally, the data are compared to, and generally agree with, phenomenological models for the DSFG population. This study demonstrates the constraining power of the cross-frequency correlation technique to constrain models for the DSFGs. Similar analyses with more data will impose tight constraints on future models.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Uncertainty-aware Continuous Implicit Neural Representations for Remote Sensing Object Counting

Many existing object counting methods rely on density map estimation (DME) of the discrete grid representation by decoding extracted image semantic features from designed convolutional neural networks (CNNs). Relying on discrete density maps not only leads to information loss dependent on the original image resolution, but also has a scalability issue when analyzing high-resolution images with cubically increasing memory complexity. Furthermore, none of the existing methods can offer reliable uncertainty quantification (UQ) for the derived count estimates. To overcome these limitations, we design UNcertainty-aware, hypernetwork-based Implicit neural representations for Counting (UNIC) to assign probabilities and the corresponding counting confidence over continuous spatial coordinates. We derive a sampling-based Bayesian counting loss function and develop the corresponding model training algorithm. UNIC outperforms existing methods on the Remote Sensing Object Counting (RSOC) dataset with reliable UQ and improved interpretability of the derived count estimates. Our code is available at https://github.com/SiyuanXu-tamu/UNIC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Excluded volume and ion-ion correlation effects on the ionic atmosphere around B-DNA: Theory, simulations, and experiments

The ionic atmosphere around a nucleic acid regulates its stability in aqueous salt solutions. One major source of complexity in biological activities involving nucleic acids arises from the strong influence of the surrounding ions and water molecules on their structural and thermodynamic properties. Here, we implement a classical density functional theory for cylindrical polyelectrolytes embedded in aqueous electrolytes containing explicit (neutral hard sphere) water molecules at experimental solvent concentrations. Our approach allows us to include ion correlations as well as solvent and ion excluded volume effects for studying the structural and thermodynamic properties of highly charged cylindrical polyelectrolytes. Several models of size and charge asymmetric mixtures of aqueous electrolytes at physiological concentrations are studied. Our results are in good agreement with Monte Carlo simulations. Our numerical calculations display significant differences in the ion density profiles for the different aqueous electrolyte models studied. However, similar results regarding the excess number of ions adsorbed to the B-DNA molecule are predicted by our theoretical approach for different aqueous electrolyte models. These findings suggest that ion counting experimental data should not be used alone to validate the performance of aqueous DNA-electrolyte models.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRSDate not supplied

Aero-Engines AI - A Machine-Learning App for Aircraft Engine Concepts Assessment

Effective deployment of trained machine-learning models could drive a high level of efficiency in aircraft engine conceptual design. Aero-Engines AI is a Windows app that has been created to deploy trained machine-learning models to assess aircraft engine concepts. It was created using tkinter, a GUI (graphical user interface) module that is built into the standard Python library. Employing tkinter greatly facilitates the sharing of machine-learning application as an executable file which can be run on Windows machines (without the need to have Python or any library installed). Current version of the app focuses on the performance prediction of conventional turbofans. The app gets user input for a turbofan design, preprocesses the input data, and deploys trained machine-learning models to predict turbofan thrust specific fuel consumption (TSFC), engine weight, core size, and turbomachinery stage-counts. The machine-learning predictive models were built by employing supervised deep-learning algorithm to study patterns in an existing open-source database of production and research turbofan engines. They were trained, cross-validated, and tested in Keras, an open-source neural networks API (application programming interface) written in Python, with TensorFlow (Google open-source artificial intelligence library) serving as the backend engine. The smooth deployment of these machine-learning models using the app shows that Aero-Engines AI is an easy-to-use and a time-saving tool for aircraft engine design-space exploration during the conceptual design stage.

machine learning↗
DOE OSTI2019

Waste Characterization Methods Based on Gamma-ray Imaging Spectrometers - 19454

Characterization of LLW, ILW, U, NORM and TENORM is critical to the nuclear industry's bottom line. Accurate measurements enable the disposal of waste at the lowest possible cost. In the past the most accurate technology available involved using HPGe detectors and extensive system modeling to estimate the total source term given what the user knows about the geometry of the source and any shielding between the source and the detectors. Now there are a variety of imaging detectors that are available commercially, including CZT and HPGe detectors that achieve energy resolution ≤1% FWHM at 662 keV. These devices couple the traditional high resolution spectroscopic analysis with next generation gamma-ray imaging information to more accurately characterize the radiation source term. Uncertainties in the source location will result in significant error in the characterization of an object's radioactivity. It is helpful if the object can be spun, such as a barrel, so that any asymmetry in the source distribution is averaged out from the detector's point of view. Even in the case of a spinning barrel it is important to determine if the radiation is concentrated near the center of the barrel or distributed near the edge. The closer one measures to a barrel, which usually dictates the time required to acquire data, the more these spatial variations in the source distribution will affect the calculated activity of each isotope. This paper will focus on showing the use of CZT imaging spectrometers to characterize complicated radiation source terms. The radiation images are used to determine where the source term should be defined in a radiation transport model that calculates the expected number of photopeak counts for each isotope detected. This is used to calculate the activity of the source. The transport model is used to determine the attenuation of the radiation source term through distance and shielding material, but the detection efficiency is based solely on radiation measurement data. These measurements are taken with a robotic source positioning device that allows for detailed directional efficiency calibration. It is a fully automated system where a detector is loaded into the device and all of the efficiency curves are captured as a function of direction by creating new measurements for each location the robot positions the source. These measurements are tagged with the true source location and the efficiency curve is calculated for each measurement. This is used to create a response function matrix as a function of energy and incident direction which is used to convert the direction dependent flux on the detector head determined from the simulation model to the expected number of counts per unit activity of each isotope of interest. The results of these source term reconstructions in both the spatial and spectral domain will be shown. The impact of improper specification of the source term will be shown for a few standard waste characterization scenarios to demonstrate the importance of proper source location determination. Finally, the underlying calibration method and some sample efficiency curves as a function of direction will be shown to demonstrate the importance of calibrating out directional response functions in imaging devices (which often have a strong forward sensitivity bias). (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2019

Total Measurement Uncertainty in Neutron Coincidence Multiplicity Analysis

Neutron multiplicity analysis is a widely used technique in international safeguards and is most commonly applied to the assay of product and scrap materials in the production of mixed oxide fuel. The typical neutron multiplicity analysis applies only a crude, almost cursory, uncertainty analysis limited to propagation of statistical errors in the derived singles, doubles, and triples count rates through the point source model. While these analyses do not account for uncertainties in basic detector parameters (e.g., efficiency or gate utilization factors), nuclear data, or covariances between any of these values, measurement uncertainties are often quoted in the range of 1–2% of the resulting mass values. In this work, we examined the impact of the nuclear data and instrumentation uncertainties on the multiplicity assay result for various counting regimes using the point-model equations as the basis for discussion. Potential biases introduced by the current nuclear data set on the measurement and the need for a more complete error treatment when we progress beyond empirical calibrations based on highly representative items will be discussed.

Mcelroy Jr, Robert↗
DOE OSTI2024

Crowd-based spatial risk assessment of urban flooding: Results from a municipal flood hotline in Detroit, MI

Climate change is increasing the frequency and intensity of extreme precipitation events, raising the risk of urban flood disasters. This study uses a crowd-sourced municipal call database to characterize the spatial distribution of flood risk in Detroit, MI. Call data including dates and addresses were obtained from the City of Detroit Department of Public Works for 2021. Calls were mapped and aggregated to census tract counts and merged with neighborhood-level data. Associations of predictors with flood calls were tested using spatial regression models. Flooding calls were located throughout the city but were concentrated in specific areas. Multivariate models of census tract level call counts indicated that increased poverty and Black, immigrant, and older residents were positively associated with flood calls, while increased elevation was associated with protective effects. Longer distances from waste water interceptors were associated with higher risk for calls. Crowd-sourced flood hotline call data can be used for effective spatial flood risk assessment. Though flooding occurs throughout the city of Detroit, infrastructural, neighborhood, and household factors influence flooding extent. Limitations included the self-reported nature of calls. Future modeling efforts might include input from local stakeholders to improve spatial risk assessment.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1994

The Galactic far-ultraviolet sky as seen by Faust: Modeling and observations

We test the model of the Galactic point source sky recently extended to the far-ultraviolet by Cohen (1994) against the set of Galactic far-ultraviolet point sources detected by the Far Ultraviolet Space Telescope (FAUST) experiment. We find excellent agreement between the predicted (log N, log S) and the observed source counts for a range of Galactic directions observed by Faust. At low galactic latitudes, we can directly infer the amount of interstellar extinction in a given direction associated with local molecular complexes, based on comparison of model calculations and FAUST star counts. The model can also be used to predict the statistical stellar content of any field. We compare the model predcitions with the observations of Brosch et al. (1994), who have spectroscopically studied an optically identified complete sample of FAUST objects in a region near the north Galactic pole. We have used the model to determine that the contribution of far-ultraviolet stars fainter than 6 x 10(exp -14) ergs/s/sq cm/A (the FAUST detection limit) is less than 1% of the diffuse sky brightness, and less than about 4% of the total point source flux.

Cohen, Martin↗
DOE OSTI2019

PRIMA-X - Performance Retargeting of Instrumentation, Measurement, and Analysis Technologies for Exascale Computing

The PRIMA-X (Performance Retargeting of Instrumentation, Measurement, and Analysis Technologies for Exascale Computing) project is the successor of the DOE PRIMA (Performance Refactoring of Instrumentation, Measurement, and Analysis Technologies for Petascale Computing) project, which addressed the challenge of creating a core measurement infrastructure that would serve as a common platform for both integrating leading parallel performance systems (notably TAU and Scalasca) and developing next-generation scalable performance tools. The PRIMA-X project shifts the focus away from refactorization of robust performance tools towards a re-targeting of the parallel performance measurement and analysis architecture for extreme scales. The massive concurrency, asynchronous execution dynamics, hardware heterogeneity, and multi-objective prerequisites (performance, power, resilience) that identify exascale systems introduce fundamental constraints on the ability to carry forward existing performance methodologies. In particular, there must be a deemphasis of per-thread observation techniques to significantly reduce the otherwise unsustainable flood of redundant performance data. Instead, it will be necessary to assimilate multi-level resource observations into macroscopic performance views, from which resilient performance metrics can be attributed to the computational features of the application. This requires a scalable framework for node-level and system-wide monitoring and runtime analyses of dynamic performance information. Also, the interest in optimizing parallelism parameters with respect to performance and energy drives the integration of tool capabilities in the exascale environment further. Initially, PRIMA-X was a collaborative project between the University of Oregon (lead institution) and the German Research School for Simulation Sciences (GRS). Because Prof. Wolf, the PI at GRS, accepted a position as full professor at Technische Universitaet Darmstadt (TU Darmstadt) starting February 1st, 2015, the work of GRS was continued at TU Darmstadt. This report covers the work at TU Darmstadt after the transition. The first main accomplishment of TU Darmstadt is the development of different techniques to aggregate performance data on the level of threads. TU Darmstadt evaluated different schemes designed at GRS during the first phase of this project and integrated them into Score-P, a widely used performance measurement framework for HPC applications. The second main accomplishment is a substantial increase of Score-P’s scalability, achieved by improving its internal representation of the underlying system. Third, we extended our performance-modeling tool Extra-P to automatically create models with more than one input parameter (e.g., process count and input size), which was previously impossible. These models allow the performance behavior of an application to be extrapolated to larger machines. Specific applications of the new multi-parameter capability realized in this project include modeling the isoefficiency function of task-based applications and a novel co-design methodology. Further extensions of Extra-P comprise online modeling functionality, automatic refinement of the model search space, and segmented performance modeling. Further contributions include new locality metrics for both shared- and distributed-memory applications, a new way of measuring the impact of inter-application inference on the execution time of an application, and finally an automatic method that selects appropriate OpenMP constructs for the parallelization of sequential code and classifies the affected variables based on their sharing semantics.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2017

Mapping Application for Penguin Populations and Projected Dynamics (MAPPPD): Data and Tools for Dynamic Management and Decision Support

The Mapping Application for Penguin Populations and Projected Dynamics (MAPPPD) is a web-based, open access, decision-support tool designed to assist scientists, non-governmental organizations and policy-makers working to meet the management objectives as set forth by the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) and other components of the Antarctic Treaty System (ATS) (that is, Consultative Meetings and the ATS Committee on Environmental Protection). MAPPPD was designed specifically to complement existing efforts such as the CCAMLR Ecosystem Monitoring Program (CEMP) and the ATS site guidelines for visitors. The database underlying MAPPPD includes all publicly available (published and unpublished) count data on emperor, gentoo, Adelie) and chinstrap penguins in Antarctica. Penguin population models are used to assimilate available data into estimates of abundance for each site and year.Results are easily aggregated across multiple sites to obtain abundance estimates over any user-defined area of interest. A front end web interface located at www.penguinmap.com provides free and ready access to the most recent count and modelled data, and can act as a facilitator for data transfer between scientists and Antarctic stakeholders to help inform management decisions for the continent.

Humphries, G. R. W.↗
DOE OSTI2023

Outbreak analysis from case count and survey data

This code enables fitting a model of an infectious disease outbreak, customized for when available data is a combination of case counts and symptom survey responses. This is just a small bit of code being used only for a publication; we need to release it to fulfill a journal's requirement that our work be reproducible.

Goldberg, Emma↗
DOE OSTI2022

TULIP: An RNA-seq-based Primary Tumor Type Prediction Tool Using Convolutional Neural Networks

Background: With cancer as one of the leading causes of death worldwide, accurate primary tumor type prediction is critical in identifying genetic factors that can inhibit or slow tumor progression. There have been efforts to categorize primary tumor types with gene expression data using machine learning, and more recently with deep learning, in the last several years. Methods In this paper, we developed four 1-dimensional (1D) Convolutional Neural Network (CNN) models to classify RNA-seq count data as one of 17 highly represented primary tumor types or 32 primary tumor types regardless of imbalanced representation. Additionally, we adapted the models to take as input either all Ensembl genes (60,483) or protein coding genes only (19,758). Unlike previous work, we avoided selection bias by not filtering genes based on expression values. RNA-seq count data expressed as FPKM-UQ of 9,025 and 10,940 samples from The Cancer Genome Atlas (TCGA) were downloaded from the Genomic Data Commons (GDC) corresponding to 17 and 32 primary tumor types respectively for training and validating the models. Results: All 4 1D-CNN models had an overall accuracy of 94.7% to 97.6% on the test dataset. Further evaluation indicates that the models with protein coding genes only as features performed with better accuracy compared to the models with all Ensembl genes for both 17 and 32 primary tumor types. For all models, the accuracy by primary tumor type was above 80% for most primary tumor types. Conclusions: We packaged all 4 models as a Python-based deep learning classification tool called TULIP (TUmor CLassIfication Predictor) for performing quality control on primary tumor samples and characterizing cancer samples of unknown tumor type. Further optimization of the models is needed to improve the accuracy of certain primary tumor types.

Jones, Sara↗
DOE OSTI2013

HerMES: COSMIC INFRARED BACKGROUND ANISOTROPIES AND THE CLUSTERING OF DUSTY STAR-FORMING GALAXIES

We present measurements of the auto- and cross-frequency power spectra of the cosmic infrared background (CIB) at 250, 350, and 500 {mu}m (1200, 860, and 600 GHz) from observations totaling {approx}70 deg{sup 2} made with the SPIRE instrument aboard the Herschel Space Observatory. We measure a fractional anisotropy {delta}I/I = 14% {+-} 4%, detecting signatures arising from the clustering of dusty star-forming galaxies in both the linear (2-halo) and nonlinear (1-halo) regimes; and that the transition from the 2- to 1-halo terms, below which power originates predominantly from multiple galaxies within dark matter halos, occurs at k{sub {theta}} {approx} 0.10-0.12 arcmin{sup -1} (l {approx} 2160-2380), from 250 to 500 {mu}m. New to this paper is clear evidence of a dependence of the Poisson and 1-halo power on the flux-cut level of masked sources-suggesting that some fraction of the more luminous sources occupy more massive halos as satellites, or are possibly close pairs. We measure the cross-correlation power spectra between bands, finding that bands which are farthest apart are the least correlated, as well as hints of a reduction in the correlation between bands when resolved sources are more aggressively masked. In the second part of the paper, we attempt to interpret the measurements in the framework of the halo model. With the aim of fitting simultaneously with one model the power spectra, number counts, and absolute CIB level in all bands, we find that this is achievable by invoking a luminosity-mass relationship, such that the luminosity-to-mass ratio peaks at a particular halo mass scale and declines toward lower and higher mass halos. Our best-fit model finds that the halo mass which is most efficient at hosting star formation in the redshift range of peak star-forming activity, z {approx} 1-3, is log(M{sub peak}/M{sub Sun }) {approx} 12.1 {+-} 0.5, and that the minimum halo mass to host infrared galaxies is log(M{sub min}/M{sub Sun }) {approx} 10.1 {+-} 0.6.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

A time-varying vulnerability index for COVID-19 in New Mexico, USA using generalized propensity scores

The coronavirus disease (COVID-19) pandemic has highlighted systemic inequities in the United States and resulted in a larger burden of negative social outcomes for marginalized communities. New Mexico, a state in the southwestern US, has a unique population with a large racial minority population and a high rate of poverty that may make communities more vulnerable to negative social outcomes from COVID-19. To identify which communities may be at the highest relative risk, we created a county-level vulnerability index. After the first COVID-19 case was reported in New Mexico on March 11, 2020, we fit a generalized propensity score model that incorporates sociodemographic factors to predict county-level viral exposure and thus, the generic risk to negative social outcomes such as unemployment or mental health impacts. We used four static sociodemographic covariates important for the state of New Mexico—population, poverty, household size, and minority population—and weekly cumulative case counts to iteratively run our model each week and normalize the exposure score to create a time-varying vulnerability index. We found the relative vulnerability between counties varied in the first eight weeks from the initial COVID-19 case before stabilizing. This framework for creating a location-specific vulnerability index in response to an ongoing disaster may be used as a quick, deployable metric to inform health policy decisions such as allocating state resources to the county level.

59 BASIC BIOLOGICAL SCIENCES↗