Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Feature Extraction”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,027 records · Page 57
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Sub-Bandgap Features in CdSeTe Solar Cells: Parsing the Roles of Material Properties and Cell Optics

In this contribution, we investigate why different dopant species and back-contact architectures lead to different sub-bandgap behaviors in CdSeTe solar cells. Through extraction of the absorptance from photoluminescence spectra, we parse the contributions from material properties and from cell optics. We show that, as expected, arsenic doping leads to an increase in sub-bandgap features over traditional copper doping, and that this is a material property of arsenic-doped CdSeTe. Conversely, the increase in sub-bandgap absorption and emission using alternative back contact architectures can be attributed to the cell optics, and more specifically to the increased reflectance of the back interface, leading to at least a doubling of the pathlength for sub-bandgap photons.

arsenic doping↗
DOE OSTI2014

Bag of Lines (BoL) for Improved Aerial Scene Representation

Feature representation is a key step in automated visual content interpretation. In this letter, we present a robust feature representation technique, referred to as bag of lines (BoL), for high-resolution aerial scenes. The proposed technique involves extracting and compactly representing low-level line primitives from the scene. The compact scene representation is generated by counting the different types of lines representing various linear structures in the scene. Through extensive experiments, we show that the proposed scene representation is invariant to scale changes and scene conditions and can discriminate urban scene categories accurately. We compare the BoL representation with the popular scale invariant feature transform (SIFT) and Gabor wavelets for their classification and clustering performance on an aerial scene database consisting of images acquired by sensors with different spatial resolutions. The proposed BoL representation outperforms the SIFT- and Gabor-based representations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

MaTableGPT: GPT‐Based Table Data Extractor from Materials Science Literature

Abstract Efficiently extracting data from tables in the scientific literature is pivotal for building large‐scale databases. However, the tables reported in materials science papers exist in highly diverse forms; thus, rule‐based extractions are an ineffective approach. To overcome this challenge, the study presents MaTableGPT, which is a GPT‐based table data extractor from the materials science literature. MaTableGPT features key strategies of table data representation and table splitting for better GPT comprehension and filtering hallucinated information through follow‐up questions. When applied to a vast volume of water splitting catalysis literature, MaTableGPT achieves an extraction accuracy (total F1 score) of up to 96.8%. Through comprehensive evaluations of the GPT usage cost, labeling cost, and extraction accuracy for the learning methods of zero‐shot, few‐shot, and fine‐tuning, the study presents a Pareto‐front mapping where the few‐shot learning method is found to be the most balanced solution owing to both its high extraction accuracy (total F1 score >95%) and low cost (GPT usage cost of 5.97 US dollars and labeling cost of 10 I/O paired examples). The statistical analyses conducted on the database generated by MaTableGPT revealed valuable insights into the distribution of the overpotential and elemental utilization across the reported catalysts in the water splitting literature.

Yi, Gyeong Hoon [Computational Science Research Ce↗
DOE OSTI2021

Open Data and Deep Semantic Segmentation for Automated Extraction of Building Footprints

Advances in machine learning and computer vision, combined with increased access to unstructured data (e.g., images and text), have created an opportunity for automated extraction of building characteristics, cost-effectively, and at scale. These characteristics are relevant to a variety of urban and energy applications, yet are time consuming and costly to acquire with today’s manual methods. Several recent research studies have shown that in comparison to more traditional methods that are based on features engineering approach, an end-to-end learning approach based on deep learning algorithms significantly improved the accuracy of automatic building footprint extraction from remote sensing images. However, these studies used limited benchmark datasets that have been carefully curated and labeled. How the accuracy of these deep learning-based approach holds when using less curated training data has not received enough attention. The aim of this work is to leverage the openly available data to automatically generate a larger training dataset with more variability in term of regions and type of cities, which can be used to build more accurate deep learning models. In contrast to most benchmark datasets, the gathered data have not been manually curated. Thus, the training dataset is not perfectly clean in terms of remote sensing images exactly matching the ground truth building’s foot-print. A workflow that includes data pre-processing, deep learning semantic segmentation modeling, and results post-processing is introduced and applied to a dataset that include remote sensing images from 15 cities and five counties from various region of the USA, which include 8,607,677 buildings. The accuracy of the proposed approach was measured on an out of sample testing dataset corresponding to 364,000 buildings from three USA cities. The results favorably compared to those obtained from Microsoft’s recently released US building footprint dataset.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Machine Learning in Infectious Disease for Risk Factor Identification and Hypothesis Generation: Proof of Concept Using Invasive Candidiasis

Machine learning (ML) models can handle large data sets without assuming underlying relationships and can be useful for evaluating disease characteristics, yet they are more commonly used for predicting individual disease risk than for identifying factors at the population level. We offer a proof of concept applying random forest (RF) algorithms to Candida-positive hospital encounters in an electronic health record database of patients in the United States. Candida-positive encounters were extracted from the Cerner HealthFacts database; invasive infections were laboratory-positive sterile site Candida infections. Features included demographics, admission source, care setting, physician specialty, diagnostic and procedure codes, and medications received before the first positive Candida culture. We used RF to assess risk factors for 3 outcomes: any invasive candidiasis (IC) vs non-IC, within-species IC vs non-IC (eg, invasive C. glabrata vs noninvasive C. glabrata), and between-species IC (eg, invasive C. glabrata vs all other IC). Fourteen of 169 (8%) variables were consistently identified as important features in the ML models. When evaluating within-species IC, for example, invasive C. glabrata vs non-invasive C. glabrata, we identified known features like central venous catheters, intensive care unit stay, and gastrointestinal operations. In contrast, important variables for invasive C. glabrata vs all other IC included renal disease and medications like diabetes therapeutics, cholesterol medications, and antiarrhythmics. Known and novel risk factors for IC were identified using ML, demonstrating the hypothesis-generating utility of this approach for infectious disease conditions about which less is known, specifically at the species level or for rarer diseases.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Deblurring for nuclei: 3D characteristics of heavy-ion collisions

Observables from nuclear and high-energy experiments can be degraded by detector performance and/or methodology in extracting the observables, such as of the final-state characteristics of heavy-ion collisions in relation to a coarsely estimated reaction-plane direction. We propose the use of deblurring methods, such as in optics, to correct for observable degradation. Our main focus is the restoration of triple-differential particle distributions in heavy-ion collisions. We demonstrate that these could be extracted from collision measurements following the Richardson-Lucy deblurring method from optics. We illustrate basic features of the restoration methodology in a schematic model assuming either ideal or more realistic particle detection. Here, the inferred three-dimensional (3D) distributions for collisions may be easier to interpret in terms of collision dynamics and sought properties of bulk matter than the currently employed Fourier coefficients, that combine information from different azimuthal angles relative to the reaction plane.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2016

SU-F-R-13: Decoding 18F-FDG Uptake Heterogeneity for Primary and Lymphoma Tumors by Using Texture Analysis in PET Images

Purpose: To explore 18F-FDG uptake heterogeneity of primary tumor and lymphoma tumor by texture features of PET image and quantify the heterogeneity difference between primary tumor and lymphoma tumor. Methods: 18 patients with primary tumor and lymphoma tumor in lung cancer were enrolled. All patients underwent whole-body 18F-FDG PET/CT scans before treatment. Texture features, based on Gray-level Co-occurrence Matrix, second and high order matrices are extracted from code using MATLAB software to quantify 18F-FDG uptake heterogeneity. The relationships of volume between energy, entropy, correlation, homogeneity and contrast were analyzed. Results: For different cases, tumor heterogeneity was not the same. Texture parameters (contrast, entropy, and correlation) of lymphoma were lower than primary tumor. On the contrast, the texture parameters (energy, homogeneity and inverse different moment) of lymphoma were higher than primary tumor. Significantly, correlations were observed between volume and energy (primary, r=−0.194, p=0.441; lymphoma, r=−0.339, p=0.582), homogeneity (primary, r=−0.146, p=0.382; lymphoma, r=−0.193, p=0.44), inverse difference moment (primary, r=−0.14, p=0.374; lymphoma, r=−0.172, p=0.414) and a positive correlation between volume and entropy (primary, r=0.233, p=0.483; lymphoma, r=0.462, p=0.680), contrast (primary, r=0.159, p=0.399; lymphoma, r=0.341, p=0.584), correlation (primary, r=0.027, p=0.165; lymphoma, r=0.046, p=0.215). For the same patient, energy for primary and lymphoma tumor is equal. The volume of lymphoma is smaller than primary tumor, but the homogeneity were higher than primary tumor. Conclusion: This study showed that there were effective heterogeneity differences between primary and lymphoma tumor by FDG-PET image texture analysis.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

maslov-group/FUN-PROSE

mRNA levels of all genes in a genome is a critical piece of information defining the overall state of the cell in a given environmental condition. Being able to reconstruct such condition-specific expression in fungal genomes is particularly important to metabolically engineer these organisms to produce desired chemicals in industrially scalable conditions. Most previous deep learning approaches focused on predicting the average expression levels of a gene based on its promoter sequence, ignoring its variation across different conditions. Here we present FUN-PROSE—a deep learning model trained to predict differential expression of individual genes across various conditions using their promoter sequences and expression levels of all transcription factors. We train and test our model on three fungal species and get the correlation between predicted and observed condition-specific gene expression as high as 0.85. We then interpret our model to extract promoter sequence motifs responsible for variable expression of individual genes. We also carried out input feature importance analysis to connect individual transcription factors to their gene targets. A sizeable fraction of both sequence motifs and TF-gene interactions learned by our model agree with previously known biological information, while the rest corresponds to either novel biological facts or indirect correlations.

Liu, Simon↗
DOE OSTI2017

Modeling Condensation in Deep Convection

Cloud-scale models apply two drastically different methods to represent condensation of water vapor to form and grow cloud droplets. Maintenance of water saturation inside liquid clouds is assumed in the computationally efficient saturation adjustment approach used in most bulk microphysics schemes. When super- or subsaturations are allowed, condensation/evaporation can be calculated using the predicted saturation ratio and (either predicted or prescribed) mean droplet radius and concentration. Here, we investigate differences between simulations of deep unorganized convection applying a saturation adjustment condensation scheme (SADJ) and a scheme with supersaturation prediction (SPRE). A double-moment microphysics scheme with CCN activation parameterized as a function of the local vertical velocity is applied to compare cloud fields simulated applying SPRE and SADJ. Clean CCN conditions are assumed to demonstrate upper limits of the SPRE and SADJ difference. Microphysical piggybacking is used to extract the impacts with confidence. Results show a significant impact on deep convection dynamics, with SADJ featuring more cloud buoyancy and thus stronger updrafts. This leads to around a 3% increase of the surface rain accumulation in SADJ. Upper-tropospheric anvil cloud fractions are much larger in SPRE than in SADJ because of the higher ice concentrations and thus longer residence times of anvil particles in SPRE, as demonstrated by sensitivity tests. Higher ice concentrations in SPRE come from significantly larger ice supersaturations in strong convective updrafts that feature water supersaturations of several percent.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

FUN-PROSE: A deep learning approach to predict condition-specific gene expression in fungi

mRNA levels of all genes in a genome is a critical piece of information defining the overall state of the cell in a given environmental condition. Being able to reconstruct such condition-specific expression in fungal genomes is particularly important to metabolically engineer these organisms to produce desired chemicals in industrially scalable conditions. Most previous deep learning approaches focused on predicting the average expression levels of a gene based on its promoter sequence, ignoring its variation across different conditions. Here we present FUN-PROSE—a deep learning model trained to predict differential expression of individual genes across various conditions using their promoter sequences and expression levels of all transcription factors. We train and test our model on three fungal species and get the correlation between predicted and observed condition-specific gene expression as high as 0.85. We then interpret our model to extract promoter sequence motifs responsible for variable expression of individual genes. We also carried out input feature importance analysis to connect individual transcription factors to their gene targets. A sizeable fraction of both sequence motifs and TF-gene interactions learned by our model agree with previously known biological information, while the rest corresponds to either novel biological facts or indirect correlations.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Prediction of Individual Social-Demographic Role Based on Travel Behavior Variability Using Long-Term GPS Data

With the development of and advances in smartphones and global positioning system (GPS) devices, travelers’ long-term travel behaviors are not impossible to obtain. This study investigates the pattern of individual travel behavior and its correlation with social-demographic features. For different social-demographic groups (e.g., full-time employees and students), the individual travel behavior may have specific temporal-spatial-mobile constraints. The study first extracts the home-based tours, including Home-to-Home and Home-to-Non-Home, from long-term raw GPS data. The travel behavior pattern is then delineated by home-based tour features, such as departure time, destination location entropy, travel time, and driving time ratio. The travel behavior variability describes the variances of travelers’ activity behavior features for an extended period. After that, the variability pattern of an individual’s travel behavior is used for estimating the individual’s social-demographic information, such as social-demographic role, by a supervised learning approach, support vector machine. In this study, a long-term (18-month) recorded GPS data set from Puget Sound Regional Council is used. The experiment’s result is very promising. The sensitivity analysis shows that as the number of tours thresholds increases, the variability of most travel behavior features converges, while the prediction performance may not change for the fixed test data.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2022

Deep learning-based spatio-temporal estimate of greenhouse gas emissions using satellite data

Accurate estimation of greenhouse gases (GHGs) emissions is very important for developing mitigation strategies to climate change by controlling and reducing GHG emissions. This project aims to develop multiple deep learning approaches to estimate anthropogenic greenhouse gas emissions using multiple types of satellite data. NO2 concentration is chosen as an example of GHGs to evaluate the proposed approach. Two sentinel satellites (sentinel-2 and sentinel-5P) provide multiscale observations of GHGs from 10-60m resolution (sentinel-2) to ~kilometer scale resolution (sentinel-5P). Among multiple deep learning (DL) architectures evaluated, two best DL models demonstrate that key features of spatio-temporal satellite data and additional information (e.g., observation times and/or coordinates of ground stations) can be extracted using convolutional neural networks and feed forward neural networks, respectively. In particular, irregular time series data from different NO 2 observation stations limit the flexibility of long short-term memory architecture, requiring zero-padding to fill in missing data. However, deep neural operator (DNO) architecture can stack time-series data as input, providing the flexibility of input structure without zero-padding. As a result, the DNO outperformed other deep learning architectures to account for time-varying features. Overall, temporal patterns with smooth seasonal variations were predicted very well, while frequent fluctuation patterns were not predicted well. In addition, uncertainty quantification using conformal inference method is performed to account for prediction ranges. Overall, this research will lead to a new groundwork for estimating greenhouse gas concentrations using multiple satellite data to enhance our capability of tracking the cause of climate change and developing mitigation strategies.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Mapping Vegetation at Species Level with High-Resolution Multispectral and Lidar Data Over a Large Spatial Area: A Case Study with Kudzu

Mapping vegetation species is critical to facilitate related quantitative assessment, and mapping invasive plants is important to enhance monitoring and management activities. Integrating high-resolution multispectral remote-sensing (RS) images and lidar (light detection and ranging) point clouds can provide robust features for vegetation mapping. However, using multiple sources of high-resolution RS data for vegetation mapping on a large spatial scale can be both computationally and sampling intensive. Here, we designed a two-step classification workflow to potentially decrease computational cost and sampling effort and to increase classification accuracy by integrating multispectral and lidar data in order to derive spectral, textural, and structural features for mapping target vegetation species. We used this workflow to classify kudzu, an aggressive invasive vine, in the entire Knox County (1362 km2) of Tennessee (U.S.). Object-based image analysis was conducted in the workflow. The first-step classification used 320 kudzu samples and extensive, coarsely labeled samples (based on national land cover) to generate an overprediction map of kudzu using random forest (RF). For the second step, 350 samples were randomly extracted from the overpredicted kudzu and labeled manually for the final prediction using RF and support vector machine (SVM). Computationally intensive features were only used for the second-step classification. SVM had constantly better accuracy than RF, and the producer’s accuracy, user’s accuracy, and Kappa for the SVM model on kudzu were 0.94, 0.96, and 0.90, respectively. SVM predicted 1010 kudzu patches covering 1.29 km2 in Knox County. We found the sample size of kudzu used for algorithm training impacted the accuracy and number of kudzu predicted. The proposed workflow could also improve sampling efficiency and specificity. Our workflow had much higher accuracy than the traditional method conducted in this research, and could be easily implemented to map kudzu in other regions as well as map other vegetation species.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Ejected Particle Size Distributions from Shocked Metal Surfaces

Here, we present size distributions for particles ejected from features machined onto the surface of shocked Sn targets. The functional form of the size distributions is assumed to be log-normal, and the characteristic parameters of the distribution are extracted from the measured angular distribution of light scattered from a laser beam incident on the ejected particles. We also found strong evidence for a bimodal distribution of particle sizes with smaller particles evolved from features machined into the target surface and larger particles being produced at the edges of these features.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Atlas of Nuclear Isomers

We present an atlas of nuclear isomers containing the experimental data for the isomers with a half-life ≥ 10 ns together with their various properties such as excitation-energy, half-life, decay mode(s), spin-parity, energies and multipolarities of emitted gamma transitions, etc. The ENSDF database complemented by the XUNDL database has been extensively used in extracting the relevant data. Recent literature from primary nuclear physics journals, and the NSR bibliographic database have been searched to ensure that the compiled data Table is as complete and current as possible. The data from NUBASE-12 have also been checked for completeness, but as far as possible original references have been cited. Many interesting systematic features of nuclear isomers emerge, some of them new; these are discussed and presented in various graphs and figures. The cutoff date for the extraction of data from the literature is August 15, 2015.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2019

Electron radiated power in cyclotron radiation emission spectroscopy experiments

The recently developed technique of Cyclotron Radiation Emission Spectroscopy (CRES) uses frequency information from the cyclotron motion of an electron in a magnetic bottle to infer its kinetic energy. Here we derive the expected radio-frequency signal from an electron in a waveguide CRES apparatus from first principles. We demonstrate that the frequency-domain signal is rich in information about the electron's kinematic parameters and extract a set of measurables that in a suitably designed system are sufficient for disentangling the electron's kinetic energy from the rest of its kinematic features. This lays the groundwork for high-resolution energy measurements in future CRES experiments, such as the Project 8 neutrino mass measurement.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2018

Raman Spectral Imaging Based Histopathology of Human Skin Tissue

Confocal Raman microspectral imaging (CRMI) provides a versatile tool to extract biochemical information on histologically defined regions with improved spatial resolution and sensitivity. Here, we investigate spectral features and perform both uni- and multivariate analyses to obtain precise correlation between the biochemical profile and histological architecture of ex vivo human skin tissue. The skin dermis layer is visualized by integrated images of collagen peaks at 855, 940, and 1248 cm{sup −1}, while the stratum corneum layer is described by chemical images of ceramides. M elanin accumulation patterns in epidermis are obtained by integration of 1375 and 1580 cm{sup −1} Raman peaks. In addition, distinct spectral features of the melanin precursor are revealed by certain Raman vibrations, which structurally characterize the molecular composition and organization of hair follicles. These results offer a unique insight into the skin histology, biochemistry, and biophysics, which provide the possibility of interpreting the underlying mechanisms of skin malignant processes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗