Search NASA⌕ Search

SEARCH · Search NASA

Results for “interpretable AI”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS1986

Use of digital Munsell color space to assist interretation of imaging spectrometer data: Geologic examples from the northern Grapevine Mountains, California and Nevada

Techniques using Munsell color transformations were developed for reducing 128 channels (or less) of Airborne Imaging Spectrometer (AIS) data to a single color-composite-image suitable for both visual interpretation and digital analysis. Using AIS data acquired in 1984 and 1985, limestone and dolomite roof pendants and sericite-illite and other clay minerals related to alteration were mapped in a quartz monzonite stock in the northern Grapevine Mountains of California and Nevada. Field studies and laboratory spectral measurements verify the mineralogical distributions mapped from the AIS data.

Kruse, F. A.↗
DOE OSTI2024

Teaching Freight Mode Choice Models New Tricks Using Interpretable Machine Learning Methods

Understanding and forecasting the intricate freight mode choice behavior under various industry, policy, and technology contexts is essential in freight planning and policymaking. Numerous models have been developed in prior studies to provide insights into freight mode selection, the majority of which use discrete choice models such as multinomial logit (MNL) models. However, logit models often rely on linear specifications of independent variables, despite potential nonlinear relationships in the data. Moreover, there often lacks a heuristic and efficient approach to identify such complex relationships to define the logit model specifications. To fill this gap, we developed an MNL model for freight mode choice using the insights from state-of-the- art machine learning (ML) models. ML models can capture the nonlinear nature of the complex decision-making process, and recent advances in 'explainable AI' have greatly improved their interpretability. The interpretable ML methods help enhance the performance of MNL models and advance knowledge of freight mode choice. Specifically, the influential factors and their relationship with individual modes are identified using SHapley Additive exPlanations (SHAP) to improve the MNL's performance. The workflow is demonstrated in a case study of Austin, Texas, and the SHAP results reveal multiple nonlinear relationships predicted by ML models. Incorporating those relationships into MNL model specifications improves the interpretability and accuracy of the MNL model compared to a conventional MNL model. Findings from this study can be used to guide freight planning and inform policymakers and practitioners on how key factors affect freight decision-making.

ADVANCED PROPULSION SYSTEMS,MATHEMATICS AND COMPUT↗
NASA NTRS1985

STAR (Simple Tool for Automated Reasoning): Tutorial guide and reference manual

STAR is an interactive, interpreted programming language for the development and operation of Artificial Intelligence application systems. The language is intended for use primarily in the development of software application systems which rely on a combination of symbolic processing, central to the vast majority of AI algorithms, with routines and data structures defined in compiled languages such as C, FORTRAN and PASCAL. References to routines and data structures defined in compiled languages are intermixed with symbolic structures in STAR, resulting in a hybrid operating environment in which symbolic and non-symbolic processing and organization of data may interact to a high degree within the execution of particular application systems. The STAR language was developed in the course of a project involving AI techniques in the interpretation of imaging spectrometer data and is derived in part from a previous language called CLIP. The interpreter for STAR is implemented as a program defined in the language C and has been made available for distribution in source code form through NASA's Computer Software Management and Information Center (COSMIC). Contained within this report are the STAR Tutorial Guide, which introduces the language in a step-by-step manner, and the STAR Reference Manual, which provides a detailed summary of the features of STAR.

Borchardt, G. C.↗
NASA NTRS1986

Patterns of vegetation in the Owens Valley, California

Spectral characteristics of semi-arid shrub communities were examined using Airborne Imaging Spectrometer (AIS) data collected in the tree mode on 23 May 1985. Mesic sites with relatively high vegetation density and distinct zonation patterns exhibited greater spectral signature variations than sites with more xeric shrub communities. Spectral signature patterns were not directly related to vegetation density or physiognomy, although spatial maps derived from an 8-channel maximum likelihood classification were supported by photo-interpreted surface features. In AIS data, the principal detected effect of shrub vegetation on the alluvial fans is to lower reflectance across the spectrum. These results are similar to those reported during a period of minimal physiological activity in autumn, indicating that shadows cast by vegetation canopies are an important element of soil-vegetation interaction under conditions of relatively low canopy cover.

Ustin, S. L.↗
DOE OSTI2024

APACE: AlphaFold2 and advanced computing as a service for accelerated discovery in biophysics

The prediction of protein 3D structure from amino acid sequence is a computational grand challenge in biophysics and plays a key role in robust protein structure prediction algorithms, from drug discovery to genome interpretation. The advent of AI models, such as AlphaFold, is revolutionizing applications that depend on robust protein structure prediction algorithms. To maximize the impact, and ease the usability, of these AI tools we introduce APACE, AlphaFold2 and advanced computing as a service, a computational framework that effectively handles this AI model and its TB-size database to conduct accelerated protein structure prediction analyses in modern supercomputing environments. We deployed APACE in the Delta and Polaris supercomputers and quantified its performance for accurate protein structure predictions using four exemplar proteins: 6AWO, 6OAN, 7MEZ, and 6D6U. Using up to 300 ensembles, distributed across 200 NVIDIA A100 GPUs, we found that APACE is up to two orders of magnitude faster than off-the-self AlphaFold2 implementations, reducing time-to-solution from weeks to minutes. This computational approach may be readily linked with robotics laboratories to automate and accelerate scientific discovery.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

The Future of a Myriad of Accelerated Biodiscoveries Lies in AI‐Powered Mass Spectrometry and Multiomics Integration

The intersection of modern artificial intelligence (AI) and mass spectrometry (MS) is set to transform the MS‐based “omics” research fields, particularly proteomics, metabolomics, lipidomics, and glycomics, enabling advancements across a wide range of domains, from health to environment and industrial biotechnology. Beginning with an overview of key challenges inherent in MS software pipelines, this personal perspective explores how AI‐driven solutions can address them to enhance data processing, integration and interpretation. It proposes a paradigm shift in molecular identification and quantitation algorithms, leveraging AI to enable holistic interpretation of MS‐based multiomics data. While centered on MS‐based omics, this holistic AI‐driven paradigm is also critical for connecting dynamic biochemical changes to genomics and transcriptomics contexts, reinforcing the integrative value of MS in multiomics research. Ultimately, this AI‐driven approach could enhance efficiency, accuracy, and molecular breadth of coverage, deepening our systems‐level understanding of biological processes and accelerating a myriad of biodiscoveries.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

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↗
NASA NTRS1987

Discriminating semiarid vegetation using airborne imaging spectrometer data - A preliminary assessment

A preliminary assessment was made of Airborne Imaging Spectrometer (AIS) data for discriminating and characterizing vegetation in a semiarid environment. May and October AIS data sets were acquired over a large alluvial fan in eastern California, on which were found Great Basin desert shrub communities. Maximum likelihood classification of a principal components representation of the May AIS data enabled discrimination of subtle spatial detail in images relating to vegetation and soil characteristics. The spatial patterns in the May AIS classification were, however, too detailed for complete interpretation with existing ground data. A similar analysis of the October AIS data yielded poor results. Comparison of AIS results with a similar analysis of May Landsat Thematic Mapper data showed that the May AIS data contained approximately three to four times as much spectrally coherent information. When only two shortwave infrared TM bands were used, results were similar to those from AIS data acquired in October.

Thomas, Randall W.↗
DOE OSTI2025

AI-Optimized Polarization at Jefferson Lab

The AI-Optimized Polarization project seeks to develop experimental control applications for polarized targets and beams at Jefferson Lab using AI/ML. This paper will focus on two ongoing efforts involving a cryogenic polarized target and a linearly-polarized photon beam. Firstly, cryogenic targets, such as those used in Halls B and C (and approved for Hall D), are complex systems that are sensitive to a number of factors, including the temperature, beam currents, and the microwave and NMR apparatus. Secondly, the Hall D photon beam polarization depends on the optimal orientation of a diamond radiator, which produces coherent bremsstrahlung radiation from the electron beam incident upon it. Manual operation of both systems is tedious and error prone; implementing well-designed, interpretable control systems that incorporate AI is expected to lead to improved real-time polarization. AI optimization of nuclear physics experiments will lead, not just to cost-savings, but also to more efficient and higher-quality data, and this project will help to lay the foundation for future autonomous experiments.

Moran, Patrick [College of William and Mary, Willi↗
DOE OSTI2026

REFSafE: A RAG-Enabled Framework for Predictive Risk Analysis and Automated Safety Report Generation in Mission-Critical Environments

Operational safety in mission-critical environments requires AI systems that are accurate, interpretable, and resistant to hallucination. We present an agentic Retrieval-Augmented Generation (RAG) framework, REFSafe, for grounded hazard analysis and automated safety report generation. The system integrates Large Language Models (LLMs) with structured operational data, historical incident repositories, policy documents, and external authoritative sources. Through iterative agentic reasoning, the framework retrieves, verifies, and synthesizes evidence prior to generation, enforcing citation-backed outputs with explicit source attribution (documents, links, and prior events) to ensure traceability and trust. To mitigate hallucinations and unsupported claims, all risk assessments and forecasts are constrained to retrieved evidence, with confidence signals derived from retrieval relevance and source consistency. A transparent pipeline enables subject matter experts (SMEs) to validate predictions, and provide structured feedback, forming a continuous performance calibration loop. Preliminary deployment demonstrates improved reliability in hazard detection and safety/vulnerability report generation. This work advances trustworthy, evidence-grounded AI for predictive safety intelligence in mission-critical operations.

Das, Sanjay [ORNL] (ORCID:0009000542591915)↗
DOE OSTI2025

Increasing the Scale of the Mass Spectrometry Query Language Compendium with Explainable AI

A significant bottleneck in metabolomics data interpretation is the effective use of domain knowledge to assign structural information based on fragmentation patterns. The mass spectrometry query language (MassQL) aims to make this process accessible and applicable across multiple analysis platforms. While advanced computational methods are capable of predicting compound structures from fragmentation data, AI/ML approaches often rely on complex, opaque criteria that are difficult to interpret or modify. As a result, their predictive patterns cannot be readily translated into human-readable rules, such as those used in MassQL. Here, in this study, we introduce ChemEcho, a machine learning embedding method that converts tandem mass spectrometry data into sparse feature vectors containing peak and neutral mass subformulae to enhance explainable AI/ML-based methods. An advantage of this approach is that decision trees trained using these feature vectors can be directly translated to MassQL. Using a battery of decision trees trained using ChemEcho embeddings to predict molecular attributes, we generated over 1500 MassQL queries for 765 molecular features and evaluated their precision and recall. From these queries, the 50 highest-performing queries were integrated into the MassQL compendium. This set of generated MassQL queries included environmentally and biologically relevant classes such as PFAS and molecules containing phosphate or sulfate substructures. To illustrate the impact these queries would have on a typical metabolomics experiment, these MassQL queries were applied to a public metabolomics data set─resulting in a marked increase in the structural information derived from tandem mass spectra. Access and reuse of these queries is expected to enhance structural annotation in untargeted experiments, leading to more specific claims and advancing many applications in metabolomics.

Harwood, Thomas V. [USDOE Joint Genome Institute (↗
DOE OSTI2022

Decoding the protein–ligand interactions using parallel graph neural networks

Abstract Protein–ligand interactions (PLIs) are essential for biochemical functionality and their identification is crucial for estimating biophysical properties for rational therapeutic design. Currently, experimental characterization of these properties is the most accurate method, however, this is very time-consuming and labor-intensive. A number of computational methods have been developed in this context but most of the existing PLI prediction heavily depends on 2D protein sequence data. Here, we present a novel parallel graph neural network (GNN) to integrate knowledge representation and reasoning for PLI prediction to perform deep learning guided by expert knowledge and informed by 3D structural data. We develop two distinct GNN architectures: $$\hbox {GNN}_{\mathrm{F}}$$ GNN F is the base implementation that employs distinct featurization to enhance domain-awareness, while $$\hbox {GNN}_{\mathrm{P}}$$ GNN P is a novel implementation that can predict with no prior knowledge of the intermolecular interactions. The comprehensive evaluation demonstrated that GNN can successfully capture the binary interactions between ligand and protein’s 3D structure with 0.979 test accuracy for $$\hbox {GNN}_{\mathrm{F}}$$ GNN F and 0.958 for $$\hbox {GNN}_{\mathrm{P}}$$ GNN P for predicting activity of a protein–ligand complex. These models are further adapted for regression tasks to predict experimental binding affinities and $$\hbox {pIC}_{\mathrm{50}}$$ pIC 50 crucial for compound’s potency and efficacy. We achieve a Pearson correlation coefficient of 0.66 and 0.65 on experimental affinity and 0.50 and 0.51 on $$\hbox {pIC}_{\mathrm{50}}$$ pIC 50 with $$\hbox {GNN}_{\mathrm{F}}$$ GNN F and $$\hbox {GNN}_{\mathrm{P}}$$ GNN P , respectively, outperforming similar 2D sequence based models. Our method can serve as an interpretable and explainable artificial intelligence (AI) tool for predicted activity, potency, and biophysical properties of lead candidates. To this end, we show the utility of $$\hbox {GNN}_{\mathrm{P}}$$ GNN P on SARS-Cov-2 protein targets by screening a large compound library and comparing the prediction with the experimentally measured data.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1986

Analysis of AIS data of the Bonanza Creek Experimental Forest, Alaska

Airborne Imaging Spectrometer (AIS) data were acquired in 1985 over the Bonanza Creek Experimental Forest, Alaska for the analysis of canopy characteristics including biochemistry. Concurrent with AIS overflights, foliage from fifteen coniferous and deciduous forest stands were analyzed for a variety of biochemical constituents including nitrogen, lignin, protein, and chlorophyll. Preliminary analysis of AIS spectra indicates that the wavelength region between 1450 to 1800 namometers (nm) displays distinct differences in spectral response for some of the forest stands. A flat field subtraction (forest stand spectra - flat field spectra) of the AIS spectra assisted in the interpretation of features of the spectra that are related to biology.

Spanner, M. A.↗
NASA NTRS1987

Optical and digital pattern recognition; Proceedings of the Meeting, Los Angeles, CA, Jan. 13-15, 1987

The papers presented in this volume provide an overview of current research in both optical and digital pattern recognition, with a theme of identifying overlapping research problems and methodologies. Topics discussed include image analysis and low-level vision, optical system design, object analysis and recognition, real-time hybrid architectures and algorithms, high-level image understanding, and optical matched filter design. Papers are presented on synthetic estimation filters for a control system; white-light correlator character recognition; optical AI architectures for intelligent sensors; interpreting aerial photographs by segmentation and search; and optical information processing using a new photopolymer.

Liu, Hua-Kuang↗
DOE OSTI2021

FAIR data infrastructure and tools for AI-assisted streamflow prediction

Focal Area(s) Areas: We discuss how the integration of AI into Earth Science models can impact streamflow predictions at both the science and data levels. Doing so, we address cross-cutting needs related to the goal of making data FAIR (Findable, Accessible, Interoperable, and Re-usable [1]) for seamless use with Artificial Intelligence/Machine Learning (AI/ML) in Earth System Science at DOE. A novel idea is that AI/ML itself can help with the FAIR data goal and address issues in targeted areas e.g. missing data, data quality and reduction. In addition, the interpretability of results obtained with new AI methods is poised to impact broader scientific challenges in hydrology..

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Refining PeakDecoder Version 2

Novel computational tools for processing multidimensional mass spectrometry (MS) data are necessary to enable deeper and automated detection and quantification of metabolites in complex backgrounds. Multidimensional MS data includes measurements from liquid chromatography (LC) and ion mobility spectrometry (IM) separations, and precursor and fragment ion spectra collected in data-independent acquisition (DIA) mode. PeakDecoder is an artificial intelligence (AI)-based software that enables automated interpretation of this kind of data to identify and quantify individual metabolites in complex mixtures. The goal of this project was to improve and re-implement PeakDecoder in a better suited programming language to enable its commercialization.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1985

Expert systems for Space Station automation

The expert systems required for automating key functions of the Manned Space Station (MSS) are explored. It is necessary that the expert systems developed be flexible, degrade gracefully in the case of a failure, and be able to work with incomplete data. The AI systems will have to perform interpretation and diagnosis, design, prediction and induction, and monitoring and control functions. Both quantitative and qualitative reasoning capabilities need improvements, as do automatic verification techniques, explanation and learning capabilities, and the use of metaknowledge, i. e., knowledge about the knowledge contained in the knowledge base. Information retrieval, fault isolation and manufacturing process control demonstrations are needed to validate expert systems for the MSS.

Georgeff, M. P.↗
DOE OSTI2024

Using AI Segmentation Models to Improve Foreign Body Detection and Triage from Ultrasound Images

Medical imaging can be a critical tool for triaging casualties in trauma situations. In remote or military medicine scenarios, triage is essential for identifying how to use limited resources or prioritize evacuation for the most serious cases. Ultrasound imaging, while portable and often available near the point of injury, can only be used for triage if images are properly acquired, interpreted, and objectively triage scored. Here, we detail how AI segmentation models can be used for improving image interpretation and objective triage evaluation for a medical application focused on foreign bodies embedded in tissues at variable distances from critical neurovascular features. Ultrasound images previously collected in a tissue phantom with or without neurovascular features were labeled with ground truth masks. These image sets were used to train two different segmentation AI frameworks: YOLOv7 and U-Net segmentation models. Overall, both approaches were successful in identifying shrapnel in the image set, with U-Net outperforming YOLOv7 for single-class segmentation. Both segmentation models were also evaluated with a more complex image set containing shrapnel, artery, vein, and nerve features. YOLOv7 obtained higher precision scores across multiple classes whereas U-Net achieved higher recall scores. Using each AI model, a triage distance metric was adapted to measure the proximity of shrapnel to the nearest neurovascular feature, with U-Net more closely mirroring the triage distances measured from ground truth labels. Overall, the segmentation AI models were successful in detecting shrapnel in ultrasound images and could allow for improved injury triage in emergency medicine scenarios.

47 OTHER INSTRUMENTATION↗