Search NASA⌕ Search

SEARCH · Search NASA

Results for “differentiable machine learning”

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

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

At least 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Toward Muon Neutrino Charged-Current Neutral Pion Cross Section Measurements at ICARUS

At ICARUS and the broader Short-Baseline Neutrino Program, few neutrino interaction channels offer as much utility as those producing neutral pions, which are an important background to electron neutrino searches and provide a standard candle for calibrating the electromagnetic shower energy scale. Beyond this, measurements of neutrino-induced neutral pions offer a probe of resonant interactions that are relevant to future accelerator neutrino experiments like the Deep Underground Neutrino Experiment. In this document, a measurement of charged-current muon neutrino interactions in liquid argon with a single neutral pion in the final state is presented, making use of neutrino interaction events recorded with the ICARUS T600 detector at Fermilab while exposed to the Booster Neutrino Beam. Event selection is carried out with a fully-automated machine learning reconstruction framework, allowing for the extraction of single-differential cross section measurements as a function of muon and neutral pion kinematic observables.

Carber, Dan [Colorado State U.]↗
DOE OSTI2025

Predictive Modeling and Uncertainty Quantification in Condition Monitoring of Active Components: A Reactor Coolant Pump Use Case

This work develops data-driven models for onset of thermal barrier leakage in reactor coolant pumps. It incorporates uncertainty quantification to enhance the reliability and robustness of pre- dictions. Using synthetic data generated by the Generic Pressurized Water Reactor simulator, realistic degradation scenarios were simulated across lifecycle stages—beginning, middle, and end of life. Key variables, including differential pressure, flow rate, vibration, and temperatures, were analyzed using machine learning framework. The fully connected neural network models demonstrated exceptional performance, achieving R2 scores exceeding 0.99 and root mean square errors as low as around 8.23 × 10-2 gallon per minute (gpm) for the three stages of the lifecy- cle. UQ analysis further validated the model’s robustness, with narrow uncertainty bounds during steady-state operations and appropriately wider bounds during transitional phases, reflecting the physical behavior of the system. This work addresses important gaps in real-time condition moni- toring and regulatory compliance by integrating advanced condition monitoring technologies with UQ into IST programs. The ability to detect thermal barrier leakage early and quantify prediction reliability supports optimizing maintenance strategies while ensuring nuclear power plants’ safe and reliable operation.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2025

Application of physics-informed neural networks (PINNs) solution to coupled thermal and hydraulic processes in silty sands

Abstract The accurate modeling of water and heat transport in soils is crucial for both geo-environmental and geothermal engineering. Traditional modeling methods are problematic because they require well-defined boundaries and initial conditions. Recently, physics-informed neural networks (PINNs), which incorporate partial differential equations (PDEs) to solve forward and inverse problems, have attracted increasing attention in machine learning research. In this study, we applied PINNs to tackle hydraulic and thermal transport coupling forward problems in silty sands. A fully connected deep neural network was utilized for training. This neural network model leverages automatic differentiation to apply the governing equations as constraints, based on the mathematical approximations established by the neural network itself. We conducted forward problems and compared the solutions derived from PINNs with those from Finite Element Method (FEM) simulations. The forward problem results demonstrate the PINNs model’s capability in predicting hydraulic transport, heat transport, and thermal–hydraulic coupling in silty sands under various boundary conditions. The PINNs exhibited great performance in simulating the thermal–hydraulic coupling problem. The accuracy of the PINNs solutions shows its potential for simulation in geotechnical engineering.

Feng, Yuan↗
DOE OSTI2022

Machine Learning Pattern Recognition Algorithm With Applications to Coherent Laser Combination

Herein we analyze a new kind of machine learning algorithm designed to feedback stabilize coherently combined lasers. This algorithm learns differential, rather than absolute, values of action in phase space, in order to facilitate learning on initially unstable systems. Experiments have shown that this approach can control small-scale spatial beam combination with high stability. In this paper we analyze the algorithm's performance and limitations in depth, showing that it can continuously learn during operation in order to track changes. Using simulation, we extend the application to temporal combination, and show that it scales to more complex instances by combining 81 beams.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Advanced Facility Design and AI/ML Enabled Safeguards to Establish Secure, Economical Recycling of Fast Reactor Fuels (Final Scientific/Technical Report)

The project, "Advanced Facility Design and AI/ML Enabled Safeguards to Establish Secure, Economical Recycling of Fast Reactor Fuels," represents a significant advancement in nuclear fuel recycling technology. It integrates cutting-edge multimodal sensor fusion, machine learning (ML), and digital twin (DT) technologies to address challenges in material safeguarding, process optimization, and regulatory compliance for pyroprocessing facilities. This research has significantly enhanced the understanding of pyrochemical fuel recycling processes by developing innovative tools and methodologies. The Multimodal Safeguards Monitoring Unit (MSMU) combines electroanalytical techniques, Raman spectroscopy, and differential thermal analysis (DTA) to enable high-fidelity, near-real-time material accountancy measurements. Machine learning techniques, such as Long Short-Term Memory (LSTM) autoencoders, are utilized to detect anomalies in material balances and sensor data, improving the reliability of safeguards monitoring. Additionally, digital twin technology has been established to provide real-time system-level monitoring and diagnostics, integrating physics-based models with sensor data to optimize process safety and efficiency.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Improving generative model-based unfolding with Schrödinger bridges

Machine learning-based unfolding has enabled unbinned and high-dimensional differential cross section measurements. Two main approaches have emerged in this research area; one based on discriminative models and one based on generative models. The main advantage of discriminative models is that they learn a small correction to a starting simulation while generative models scale better to regions of phase space with little data. We propose to use Schrödinger bridges and diffusion models to create , an unfolding approach that combines the strengths of both discriminative and generative models. The key feature of is that its generative model maps one set of events into another without having to go through a known probability density as is the case for normalizing flows and standard diffusion models. We show that achieves excellent performance compared to state of the art methods on a synthetic Z + jets dataset. Published by the American Physical Society 2024

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Handling Iterative Solvers in an Algorithmic Differentiation Framework Using Implicit Methods

Differentiable programming is a powerful concept as it enables the seemly propagation of gradients through functions, algorithms, and/or whole physics simulations. These gradients are useful for a wide variety of applications, including sensitivity studies and machine learning, but one of particular interest is optimization. Gradient-based optimization, enabled through automatic/algorithmic differentiation (AD), can be used on predictive physical models to efficiently optimize a set of design variables. AD methods are a particularly promising approach to complex physics simulations because they can be shown to scale well with an increasing number of design variables; however, care must be taken when coupling between different models or different states of a single model.

algorithmic differentiation↗
DOE OSTI2019

Subtask 1.1 – Advanced Characterization of Unconventional Oil and Gas Reservoirs to Enhance CO2 Storage Resource Estimates

The Energy & Environmental Research Center (EERC) developed methods and approaches for advanced characterization of unconventional oil and gas reservoirs to enhance CO2 storage resource estimates using samples collected from organic-rich shales of the Bakken Formation. The project was funded under U.S. Department of Energy (DOE) Cooperative Agreement No. DE-FE0024233, with nonfederal cost share provided by Hitachi High Technologies America. The project participants included the EERC, Hitachi High Technologies America, the team at the National Energy Technology Laboratory’s (NETL’s) computed tomography (CT) scanning laboratory in Morgantown, West Virginia, and several NETL staff who have worked on CO2 storage resource estimation. Investigations on the use of subsurface geologic formations for CO2 storage have been ongoing for many years. In the last decade, improvements in directional drilling technology and hydraulic fracturing of organic-rich shale and other tight rock formations have opened new opportunities for CO2 storage, with the potential cobenefit of enhancing gas and oil recovery. Gas-bearing, organic-rich shales have been of particular interest as a carbon storage target because their ability to store large volumes of CO2 is implied by their ability to store methane; however, estimating the CO2 storage resource of organic-rich shales and other unconventional reservoirs is challenging because of the lack of detailed geologic and petrophysical data that are needed to develop improved volumetric equations. According to NETL’s method for estimating regional-scale CO2 storage resources, the key factors that control CO2 storage in geologic formations include void space within induced fractures, natural fractures and matrix pores, as well as the occurrence and distribution of organic matter (OM) and clay minerals. A key goal of this effort was to use field emission scanning electron microscopy (FESEM) analysis of fine-grain unconventional reservoir rocks coupled with advanced image analysis and machine learning techniques to quantify the occurrence and distribution of porosity, OM, and clay minerals. By better understanding the potential migration pathways of CO2 within tight rocks and quantifying the occurrence of sorptive components (OM and clays) along potential flow pathways, methods to estimate the CO2 storage resource of such rocks can be refined. The Bakken petroleum system (Bakken) was used as the source of rock samples for this effort because it includes organic-rich shales with contrasting degrees of thermal maturity as well as tight, nonshale clastic rocks with varying clay mineral content. To better understand the key components of interest in tight rock samples, the EERC worked closely with Hitachi to optimize the FESEM settings necessary to resolve the sample mineralogy, porosity and OM content at micro- to nanoscale resolution. The EERC also worked closely with Hitachi to apply and improve its advanced mineral identification and characterization system (AMICS) software so that it is better suited for mineral identification at the high resolution necessary to resolve features of interest in fine-grained rocks such as organic-rich shales. The EERC also applied advanced image analysis software packages, including Ilastik, to classify and quantify the key components of interest within the samples using FESEM imagery. These advanced techniques allowed for quantification and differentiation of the void space within fractures, pores and OM, as well as a technique to determine and quantify the key mineral and OM components surrounding pore spaces. Being able to better quantify the void space and the potential accessibility of CO2 to sorptive minerals or OM within tight rocks allows for improved estimation of the CO2 storage resource potential of organic-rich shales and other unconventional formations. The information derived from the image analysis techniques was used to refine NETL’s method for estimating regional-scale CO2 storage resources in collaboration with staff from NETL. An additional goal of this work was to better understand rates of CO2 permeation, flow pathways, and sorption in organic-rich shales and nonshale reservoir rocks and to evaluate potential changes in CO2 flow over time. The EERC conducted long-duration (several weeks) CO2 flow-through testing at reservoir conditions using Bakken shale and nonshale samples to assess CO2 permeation rates and retention/sorption within core plugs to represent bulk matrix behavior. Conventional isothermal adsorption testing over a range of pressure conditions was also performed on pulverized Bakken samples to better assess potential maximum sorptive capacity of the samples. The results of this testing illustrated the higher sorptive capacity of the Bakken shales when compared to the sorptive capacity of the nonshale reservoir rock, which contains very little OM when compared to the Bakken shales. The results also illustrate the differences in permeation behavior as a function of fracture flow versus matrix flow. In an attempt to visualize CO2 permeation and flow behavior in Bakken shale samples over time, the EERC collaborated with NETL’s CT Lab in Morgantown, West Virginia. A variety of tests were performed at the NETL lab on unfractured and fractured samples of the Bakken shale. The unfractured Bakken shale samples did not result in CO2 migration within the samples over the testing duration; thus imagery of CO2 migration pathways within the matrix of Bakken samples was not obtained. The results of the flow-through testing within fractured samples illustrate changes in CO2 permeation rates over time, with a trend of decreasing CO2 flow over time. The results of this effort suggest that FESEM analysis coupled with advanced image analysis and machine learning techniques are effective in characterizing and quantifying the mineralogy, porosity, fractures, and OM content of 2-D imagery at the high resolution needed to identify features of interest in tight rocks. In addition, the image analysis techniques allow for identification and quantification of different porosity types as well as the key mineral components and/or OM that CO2 would contact in the reservoir as it migrates through pores and fracture networks. This is key information needed to better estimate the CO2 storage resource potential of organic-rich shales and other tight reservoirs. Additional work is needed to determine the sorptive capacity of the individual sample components (OM and clay minerals) at reservoir conditions so that this approach and the refined CO2 storage resource equation can be applied to other shale and unconventional reservoirs. In addition, subsequent work should be performed to better understand the number of images needed to use stereological approaches to estimate the volume of sample components using 2-D imagery, to explore machine learning techniques to differentiate between induced vs. natural fractures, and to estimate the connectivity of pores and fracture networks using 2-D imagery. Being able to derive this information at the microscale would facilitate scaling up of the sample characterization data used for reservoir modeling and simulation to estimate the CO2 storage resource potential of tight formations.

03 NATURAL GAS↗
DOE OSTI2023

A bacterial sensor taxonomy across earth ecosystems for machine learning applications

Microbial communities have evolved to colonize all ecosystems of the planet, from the deep sea to the human gut. Microbes survive by sensing, responding, and adapting to immediate environmental cues. This process is driven by signal transduction proteins such as histidine kinases, which use their sensing domains to bind or otherwise detect environmental cues and “transduce” signals to adjust internal processes. We hypothesized that an ecosystem’s unique stimuli leave a sensor “fingerprint,” able to identify and shed insight on ecosystem conditions. To test this, we collected 20,712 publicly available metagenomes from Host-associated, Environmental, and Engineered ecosystems across the globe. We extracted and clustered the collection’s nearly 18M unique sensory domains into 113,712 similar groupings with MMseqs2. We built gradient-boosted decision tree machine learning models and found we could classify the ecosystem type (accuracy: 87%) and predict the levels of different physical parameters (R2 score: 83%) using the sensor cluster abundance as features. Feature importance enables identification of the most predictive sensors to differentiate between ecosystems which can lead to mechanistic interpretations if the sensor domains are well annotated. To demonstrate this, a machine learning model was trained to predict patient’s disease state and used to identify domains related to oxygen sensing present in a healthy gut but missing in patients with abnormal conditions. Moreover, since 98.7% of identified sensor domains are uncharacterized, importance ranking can be used to prioritize sensors to determine what ecosystem function they may be sensing. Furthermore, these new predictive sensors can function as targets for novel sensor engineering with applications in biotechnology, ecosystem maintenance, and medicine.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Speeding up and reducing memory usage for scientific machine learning via mixed precision

Scientific machine learning (SciML) has emerged as a versatile approach to address complex computational science and engineering problems. Within this field, physics-informed neural networks (PINNs) and deep operator networks (DeepONets) stand out as the leading techniques for solving partial differential equations by incorporating both physical equations and experimental data. However, training PINNs and DeepONets require significant computational resources, including long computational times and large amounts of memory. In search of computational efficiency, training neural networks using half precision (float16) rather than the conventional single (float32) or double (float64) precision has gained substantial interest, given the inherent benefits of reduced computational time and memory consumed. However, we find that float16 cannot be applied to SciML methods, because of gradient divergence at the start of training, weight updates going to zero, and the inability to converge to a local minima. To overcome these limitations, we explore mixed precision, which is an approach that combines the float16 and float32 numerical formats to reduce memory usage and increase computational speed. Our experiments showcase that mixed precision training not only substantially decreases training times and memory demands but also maintains model accuracy. Here, we also reinforce our empirical observations with a theoretical analysis. The research has broad implications for SciML in various computational applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Evaluation of normalization strategies for mass spectrometry-based multi-omics datasets

Introduction Data normalization is crucial for multi-omics integration, reducing systematic errors and maximizing the likelihood of discovering true biological variation. Most studies assess normalization for a single omics type or use datasets from separate experiments. Few address time-course data, where normalization might bias temporal differentiation. In this study, we compared common normalization methods and a machine learning approach, Systematical Error Removal using Random Forest (SERRF), using multi-omics datasets generated from the same experiment—even from the same cell lysate. Objectives To develop a straightforward process to assess normalization effects and identify the most robust methods across multi-omics datasets. Methods We analyzed metabolomics, lipidomics, and proteomics datasets from primary human cardiomyocytes and motor neurons exposed to acetylcholine-active compounds over time. Normalization effectiveness was evaluated based on improvement in QC features consistency and observing the change in treatment and time-related variance. Results Probabilistic Quotient Normalization (PQN) and Locally Estimated Scatterplot Smoothing (LOESS) QC were identified as optimal for metabolomics and lipidomics, while PQN, Median, and LOESS normalization excelled for proteomics. These methods consistently enhanced QC feature consistency in metabolomics and lipidomics, and preserved time-related variance or treatment-related variance in proteomics, demonstrating their effectiveness and robustness. SERRF normalization, applied only to metabolomics in this study, outperformed other methods in some datasets but inadvertently masked treatment-related variance in others. Conclusion Our evaluation identified PQN and LoessQC as the top methods for metabolomics and lipidomics, and PQN, Median, and Loess normalization for proteomics, in multi-omics integration in a temporal study.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Multidimensional quantitative phenotypic and molecular analysis reveals neomorphic behaviors of p53 missense mutants

Abstract Mutations in the TP53 tumor suppressor gene occur in >80% of the triple-negative or basal-like breast cancer. To test whether neomorphic functions of specific TP53 missense mutations contribute to phenotypic heterogeneity, we characterized phenotypes of non-transformed MCF10A-derived cell lines expressing the ten most common missense mutant p53 proteins and observed a wide spectrum of phenotypic changes in cell survival, resistance to apoptosis and anoikis, cell migration, invasion and 3D mammosphere architecture. The p53 mutants R248W, R273C, R248Q, and Y220C are the most aggressive while G245S and Y234C are the least, which correlates with survival rates of basal-like breast cancer patients. Interestingly, a crucial amino acid difference at one position—R273C vs. R273H—has drastic changes on cellular phenotype. RNA-Seq and ChIP-Seq analyses show distinct DNA binding properties of different p53 mutants, yielding heterogeneous transcriptomics profiles, and MD simulation provided structural basis of differential DNA binding of different p53 mutants. Integrative statistical and machine-learning-based pathway analysis on gene expression profiles with phenotype vectors across the mutant cell lines identifies quantitative association of multiple pathways including the Hippo/YAP/TAZ pathway with phenotypic aggressiveness. Further, comparative analyses of large transcriptomics datasets on breast cancer cell lines and tumors suggest that dysregulation of the Hippo/YAP/TAZ pathway plays a key role in driving the cellular phenotypes towards basal-like in the presence of more aggressive p53 mutants. Overall, our study describes distinct gain-of-function impacts on protein functions, transcriptional profiles, and cellular behaviors of different p53 missense mutants, which contribute to clinical phenotypic heterogeneity of triple-negative breast tumors.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Field Work Proposal ERKJ358: Black-box training for scientific machine learning models (Final Report)

The overarching goal of this project is to develop a scalable black-box training capability for scientific machine learning (SciML) problems that are non-trainable with existing automatic differentiation (AD)-based algorithms. AD assumes that a loss function can be decomposed into a sequence of elementary operations whose derivatives are known. This assumption is violated when the loss function includes a black-box physical model (e.g., a legacy simulator). The current strategy, converting a black-box simulator to an AD-enabled code via differential programming, is inflexible and time-, labor-consuming. Thus, black-box optimization is a main workhorse for training SciML models, e.g., in scientific reinforcement learning, hyper-parameter fine tuning, designing SciML models with adversarial robustness, etc.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Scientific Discovery with Physics-Informed System Identification (Abbreviated Report)

My fellowship research focused on making physics-based simulations faster and more useful through machine learning. Many problems in science and engineering are governed by partial differential equations, but high-fidelity simulations are often too expensive to run repeatedly. I worked on improving Latent Space Dynamics Identification (LaSDI), a reduced-order modeling framework that compresses large simulation data sets into a smaller representation and then learns how that representation evolves over time. The motivation was to develop reduced models that remain accurate for more challenging systems, especially when predictions must remain reliable over long time intervals or when the underlying dynamics are more complicated than standard methods can easily handle. I also contributed to related work on Quandary, a high-performance software effort for simulation and control of open quantum systems, before focusing primarily on Latent Space Dynamics Identification methods. The main outcomes of the fellowship were two new algorithms (both of which were published), Rollout-LaSDI and Higher-Order LaSDI, together with supporting work on multi-stage Latent Space Dynamics Identification. Rollout-LaSDI improved long-term prediction by training the model to stay accurate over extended time horizons, and Higher-Order LaSDI broadened the method so it could model systems with higher-order time dynamics. My contributions to multistage Latent Space Dynamics Identification also helped show that its later training stages could be simplified without losing effectiveness, and that this behavior held across different model architectures and training strategies. Taken together, these advances improved the accuracy, flexibility, and practical value of reduced-order modeling tools for computational science.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

LIPA: Lunar Ice Perception Algorithm

The highest concentration of Lunar water-ice stores exists within the Permanently Shadowed Regions (PSRs) of the Lunar South Pole. As such, the ability to locate in situ water-ice stores in an accurate, systematic, and safe manner will prove vital for future Lunar activities which rely on hydrogen-based resources. Here we show how the strong absorptive properties of ice can be exploited (by coupling robotics, infrared imaging techniques, and machine learning) so that surface frost located in PSRs can be easily differentiated from the surrounding frozen regolith. Testbeds which simulate an icy lunar landscape were created and then imaged using a mid-wave infrared (MWIR) camera system mounted to a robotic arm (UR5e). Testbeds were imaged under two filter modes (1) wide band mode: whereby imagery captured filled a spectral range of 3.0 - 5.0 μm and (2) narrow band mode: whereby imagery captured were confined to a single central wavelength (CWL) of 3.15 ± 0.03 μm. A CWL of 3.15 μm was chosen due to the highly absorptive nature of ice at that specific wavelength. Images produced under both camera modes were processed in MATLAB. Narrow band images (NB) were subtracted from their wide band (WB) counterparts to produce differenced images (DI) which clearly demonstrated the spatial extent of ice (e.g., WB – NB = DI). Differenced images were used to train a Microsoft Azure model to discriminate between frozen regolith which did and did not contain ice. These works prove promising for future in situ resource utilization (ISRU) missions which employ robotics in combination with camera systems to advance science objectives (e.g., locate water-ice in frozen regolith) on the lunar surface.

Ane Slabic↗
DOE OSTI2023

A multifidelity deep operator network approach to closure for multiscale systems

Projection-based reduced order models (PROMs) have shown promise in representing the behavior of multiscale systems using a small set of generalized (or latent) variables. Despite their success, PROMs can be susceptible to inaccuracies, even instabilities, due to the improper accounting of the interaction between the resolved and unresolved scales of the multiscale system (known as the closure problem). In the current work, we interpret closure as a multifidelity problem and use a multifidelity deep operator network (DeepONet) framework to address it. In addition, to enhance the stability and/or accuracy of the multifidelity-based closure, we employ the recently developed ``in-the-loop" training approach from the literature on coupling physics and machine learning models. The resulting approach is tested on shock advection for the one-dimensional viscous Burgers equation and vortex merging for the two-dimensional Navier-Stokes equations. Furthermore, the numerical experiments show significant improvement of the predictive ability of the closure-corrected PROM over the un-corrected one both in the interpolative and the extrapolative regimes.

42 ENGINEERING↗
NASA NTRS2019

Predictive Modeling for Differential Diagnosis and Mortality Risk Assessment

The prevalence of electronic health record (EHR) systems has brought prodigious biomedical informatics opportunity. Automated machine learning methods can effectively utilize such data and have become common tools for healthcare predictive modeling. Researches in medical informatics have explored the potential of deep learning and classical models in emergent care scenarios. In particular, predicting differential diagnoses for admissions have proven useful in decreasing unnecessary lab tests and improving inpatient triage decision-making. Moreover, identification of high-risk patients for in-hospital mortality is vitally important to maximize allocation of medical resources.The Medical Information Mart for Intensive Care (MIMIC-III) database, containing de-identified critical care inpatient was used in our study. This data set captures hospital patient laboratory measurements, pharmacologic prescriptions, diagnostic data and procedure event recordings. When considering adult patients and discounting admissions with ICU length of stay less than 24 hours, there were 37,787 unique admissions and 30,414 total patients. We examined the top 25 most prevalent ICD-9 group-level disease specificities in MIMIC-III using a multi-label classification model. In-hospital mortality was modeled as binary classification with 4,155 (13%) adult patients that expired, of which 3,138 (75.5%) were in the ICU setting. The metrics AUC, F1 score, sensitivity and specificity values calculated for each disease label measured prediction performance.The usage of ICD-9 group codes reduced feature dimension from 14,567 to 942 and greatly improved distribution of patient diagnostic categories. Disease temporal patterns were captured by considering the most frequently sampled 6 vital signs and 13 laboratory values. Missing data were imputed at each time-stamp. Time-series raw hourly average values were converted into 5 summary features (mean, standard deviation, number of observations, min & max values). Patient demographic variables such as age, gender, marital status and ethnicity were also factored into the modeling. Choi et al showed that contextual embedding of medical data, diagnostic and procedural codes alone can predict future diagnoses with sensitivity as high as 0.79. We utilized an embedding technique called word2vec which allowed sparse representations of medical history to be transformed into dense word vectors. The mappings captured contextual information by treating each admission as a sentence and learning the most likely neighboring words in a sliding window fashion. Binary and multi-label classification was achieved via collapse models, which do not consider temporal information, as well as recurrent neural networks with regularization, Softmax output layer activation together with categorical cross-entropy as the loss function.

US Army collaboration↗
DOE OSTI2026

Quantitative Analysis and Prediction of Thermal Runaway Metrics of High-Nickel Oxide Cathodes by Machine Learning Models

The pursuit of higher energy density in lithium-ion batteries has made high-nickel (Ni) layered oxides leading cathode candidates for next-generation electric vehicles. However, their poor thermal stability, particularly at Ni contents ≥ 90%, increases the risk of cathode-initiated thermal runaway. Furthermore, we present a data-driven framework combining linear and nonlinear machine learning models to predict key thermal runaway descriptors from a high-throughput differential scanning calorimetry database. With cathode composition and state of charge (SOC) as input features, the ensemble model accurately predicts peak temperature, heat release, and peak heat flow. SHAP analysis identifies Ni content and SOC as the dominant factors controlling thermal runaway temperature, while SOC primarily governs heat release and peak heat flow. Al, Mg, and Mn improve thermal stability by strengthening metal–oxygen bonding and delaying structural transformation, whereas B mainly reduces heat release through surface passivation. Validation with a new cathode composition confirms accurate prediction of SOC-dependent thermal runaway behavior and critical SOC.

25 ENERGY STORAGE↗