Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Automatic Data Processing”

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 487 records · Page 27
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

A database of thermally activated delayed fluorescent molecules auto-generated from scientific literature with ChemDataExtractor

A database of thermally activated delayed fluorescent (TADF) molecules was automatically generated from the scientific literature. It consists of 25,482 data records with an overall precision of 82%. Among these, 5,349 records have chemical names in the form of SMILES strings which are represented with 91% accuracy; these are grouped in a subsidiary database. Each data record contains one of the following four properties: maximum emission wavelength (λ EM ), photoluminescence quantum yield (PLQY), singlet-triplet energy splitting (ΔE ST ), and delayed lifetime (τ D ). The databases were created through text mining using ChemDataExtractor, a chemistry-aware natural-language-processing toolkit, which has been adapted for TADF research. The text-mined corpus consisted of 2,733 papers from the Royal Society of Chemistry and Elsevier. To the best of our knowledge, these databases are the first databases that have been auto-generated for TADF molecules from existing publications. The databases have been publicly released for experimental and computational applications in the TADF research field.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Adaptive Learning for Concept Drift in Application Performance Modeling

Supervised learning is a promising approach for modeling the performance of applications running on large HPC systems. A key assumption in supervised learning is that the training and testing data are obtained under the same conditions. However, in production HPC systems these conditions might not hold because the conditions of the platform can change over time as a result of hardware degradation, hardware replacement, software upgrade, and configuration updates. These changes could alter the data distribution in a way that affects the accuracy of the predictive performance models and render them less useful; this phenomenon is referred to as concept drift. Ignoring concept drift can lead to suboptimal resource usage and decreased efficiency when those performance models are deployed for tuning and job scheduling in production systems. To address this issue, we propose a concept-drift-aware predictive modeling approach that comprises two components: (1) an online Bayesian changepoint detection method that can automatically identify the location of events that lead to concept drift in near-real time and (2) a moment-matching transformation inspired by transfer learning that converts the training data collected before the drift to be useful for retraining. We use application input/output performance data collected on Cori, a production supercomputing system at the National Energy Research Scientific Computing Center, to demonstrate the effectiveness of our approach. The results show that concept-drift-aware models obtain significant improvement in accuracy; the median absolute error of the best-performing Gaussian process regression improved by 58.8% when the proposed approaches were used.

adaptive learning↗
DOE OSTI2015

MO-FG-303-01: FEATURED PRESENTATION and BEST IN PHYSICS (THERAPY): Automating LINAC QA: Design and Testing of An Image Acquisition and Processing System Utilizing a Combination of Radioluminescent Phosphors, Embedded X-Ray Markers and Optical Measurements

Purpose: The recent development of phosphors to visualize radiation beams from linear accelerators (LINAC) offers a unique opportunity for evaluating radiation fields within the context of the treatment space. The purpose of this study was to establish an automated, self-calibrating prototype system for performing quality assurance (QA) measurements. Methods: A thin layer of Gd{sub 2}O{sub 2}S:Tb phosphor and fiducial markers were embedded on several planar faces of a custom-designed phantom. The phantom was arbitrarily placed near iso-center on the couch of a LINAC equipped with on-board megavoltage (MV) and kilovoltage (kV) imagers. A plan consisting of several beams and integrated image acquisitions was delivered. Images of the phantom were collected throughout the delivery. Salient features, such as fiducials, crosshairs and beam edges were then extracted from these images used to calibrate the system, adjust for variations in phantom placement, and perform measurements. Beam edges were visualized by imaging the light generated by the phosphor on the phantom enabling direct comparison with the light field and laser locations. Registration of MV, kV and optical image data was performed using the embedded fiducial markers, enabling comparison of imaging center locations. Measurements specified by TG-142 were calculated and compared with those obtained from a commercially available QA system. Results: The system was able to automatically extract the location of the fiducials, lasers, light field and radiation field from the acquired images regardless of phantom positioning. It was also able to automatically identify the locations of fiducial markers on kV and MV images. All collected measurements were within TG-142 guidelines. The difference between the prototype and commercially available system were less than 0.2 mm. Conclusion: The prototype system demonstrated the capability of accurately and autonomously evaluating various TG-142 parameters independent of operator input and phantom setup.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

A non-cooperative meta-modeling game for automated third-party calibrating, validating and falsifying constitutive laws with parallelized adversarial attacks

The evaluation of constitutive models, especially for high-risk and high-regret engineering applications, requires efficient and rigorous third-party calibration, validation and falsification. While there are numerous efforts to develop paradigms and standard procedures to validate models, difficulties may arise due to the sequential, manual, and often biased nature of the commonly adopted calibration and validation processes, thus slowing down data collections, hampering the progress towards discovering new physics, increasing expenses and possibly leading to misinterpretations of the credibility and application ranges of proposed models. This work attempts to introduce concepts from game theory and machine learning techniques to overcome many of these existing difficulties. Here, we introduce an automated meta-modeling game where two competing AI agents systematically generate experimental data to calibrate a given constitutive model and to explore its weakness such that the experiment design and model robustness can be improved through competitions. The two agents automatically search for the Nash equilibrium of the meta-modeling game in an adversarial reinforcement learning framework without human intervention. In particular, a protagonist agent seeks to find the more effective ways to generate data for model calibrations, while an adversary agent tries to find the most devastating test scenarios that expose the weaknesses of the constitutive model calibrated by the protagonist. By capturing all possible design options of the laboratory experiments into a single decision tree, we recast the design of experiments as a game of combinatorial moves that can be resolved through deep reinforcement learning by the two competing players. Our adversarial framework emulates idealized scientific collaborations and competitions among researchers to achieve a better understanding of the application range of the learned material laws and prevent misinterpretations caused by conventional AI-based third-party validation. Numerical examples are given to demonstrate the wide applicability of the proposed meta-modeling game with adversarial attacks on both human-crafted constitutive models and machine learning models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

KBKit: A Python Toolkit for Kirkwood–Buff Theory from Molecular Dynamics

Thermodynamic properties of liquid mixtures govern processes that range from drug delivery to energy storage, yet extracting these properties from molecular simulations remains challenging. Kirkwood–Buff (KB) theory offers a rigorous route by linking microscopic pair distribution functions to macroscopic free energies, but practical use of the theory has been hindered by two obstacles: (i) the long simulations needed to obtain well-converged Kirkwood-Buff integrals (KBIs) and (ii) the specialized corrections required to translate finite-size data to the thermodynamic limit. $\texttt{KBKit}$ is an open-source Python package that removes these barriers. It automatically computes KBIs and derived thermodynamic quantities from GROMACS input files, applies state-of-the-art finite-size corrections, and provides built-in diagnostic tools to quantify statistical uncertainty. Written with modern software-engineering practices—continuous integration, extensive unit testing, and thorough documentation—$\texttt{KBKit}$ is both reliable and easy to extend. By condensing complex KBI analysis into a few intuitive commands, $\texttt{KBKit}$ enables researchers to incorporate KB theory into routine simulation workflows and accelerate the discovery of solution-phase thermodynamics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Adaptive Dynamic Digital Twin for Test Scenario Generation

Vehicle testing has been an important part in the development of both highly automated vehicles (HAV) and advanced driving assistant systems (ADAS). Obtaining a good representation of the Vehicle Under Test (VUT) is crucial for test scenario library generation (TSLG). Current vehicle testing methods often involve calibrating car-following models using vehicle trajectory data to create static representations that cannot be dynamically updated. For instance, when multiple vehicle trajectories are collected, it is difficult to automatically determine whether a new trajectory improves the model's representativeness or degrades its accuracy. In this paper, we introduce a dynamically updated digital twin modeling framework featuring an adaptive mechanism that evaluates new trajectory data. This mechanism can decide whether to incorporate newly collected data into the current model or create a separate digital twin model when the trajectory significantly differs from prior data. Vehicle location, speed, and acceleration extracted from the newly collected trajectory data are used to support the dynamic update decision. By integrating this digital twin model into the test library generation process, we demonstrate its ability to assist in generating test libraries while effectively handling newly collected data.

Chen, Hanlin [ORNL] (ORCID:0000000165087715)↗
DOE OSTI2014

Innovative Ballasted Flat Roof Solar PV Racking System

The objective of this project was to reduce the cost of racking for PV solar on flat commercial rooftops. Cost reductions would come from both labor savings and material savings related to the installation process. The rack would need to accommodate the majority of modules available on the market. Cascade Engineering has a long history of converting traditional metal type applications over to plastic. Injection molding of plastics have numerous advantages including selection of resin for the application, placing the material exactly where it is needed, designing in features that will speed up the installation process, and weight reduction of the array. A plastic rack would need to meet the requirements of UL2703, Mounting systems, mounting devices, clamping/retention devices, and ground lugs for use with flat-plate photovoltaic modules and panels. Comparing original data to the end of project racking design, racking material costs were reduced 50% and labor costs reduced 64%. The racking product accommodates all 60 and 72 cell panels on the market, meets UL2703 requirements, contributes only 1.3 pounds per square foot of weight to the array, requires little ballast to secure the array, automatically grounds the module when the module is secured, stacks/nests well for shipping/fewer lifts to the roof, provides integrated wire routing, allows water to drain on the roof, and accommodates various seismic roof connections. Project goals were achieved as noted in the original funding application.

14 SOLAR ENERGY↗
DOE OSTI2023

AutoReP: Automatic ReLU Replacement for Fast Private Network Inference

The proliferation of the Machine-Learning-As-A-Service (MLaaS) market has brought to light a number of clients’ data privacy and security concerns. One promising solution is private inference (PI) techniques using cryptographic primitives. These techniques often come with high computation and communication overhead associated with the non-linear operator such as ReLU. Several approaches have been developed in reducing the number of ReLU operations, however, they either require a heuristic threshold selection or introduce significant accuracy drop. This work presents AutoReP, a gradient-based framework for non-linear operators reduction that aims to mitigate these concerns from a systematic perspective. AutoReP automates the process of discrete selection of ReLU and polynomial functions on neurons to accelerate PI applications. We also introduce distribution-aware polynomial approximation (DaPa) to accurately approximate ReLUs under given distribution, preserving model expressivity. Our experimental results demonstrate significant accuracy improvements of 6.12% (94.31%, 12.9K ReLU budget, CIFAR-10), 8.39% (74.92%, 12.9K ReLU budget, CIFAR-100), and 9.45% (63.69%, 55K ReLU budget, Tiny-ImageNet) over current state-of-the-art methods, e.g., SNL. Morever, AutoReP is applied to EfficientNet-B2 on ImageNet dataset, and achieved 75.55% accuracy with 176.1 × ReLU budget reduction.

Peng, Hongwu↗
DOE OSTI2021

Cylinder Occupancy and Measurement Suite

This software automatically records data from the Unattended Cylinder Verification Station, determines the state of the instrument, captures images of uranium hexafluoride shipment cylinders, and determines their weight. The code has been developed primarily for the International Atomic Energy Agency, but this type of instrument could be of interest to any facility operator in the uranium supply chain that handles uranium hexafluoride shipment cylinders (e.g. the 30B and 48Y cylinder). Operators may be interested in having this instrument confirm shipper declared measurement values upon receipt of a cylinder. They may also use it to aid in process monitoring if the quantity of material in the cylinder changes due to a processing activity.

Stewart, Scott L↗
DOE OSTI2016

Assessment of a Hybrid Continuous/Discontinuous Galerkin Finite Element Code for Geothermal Reservoir Simulations

FALCON (“Fracturing And Liquid CONvection”) is a hybrid continuous / discontinuous Galerkin finite element geothermal reservoir simulation code based on the MOOSE (“Multiphysics Object-Oriented Simulation Environment”) framework being developed and used for multiphysics applications. In the present work, a suite of verification and validation (“V&V”) test problems for FALCON was defined to meet the design requirements, and solved to the interests of enhanced geothermal system (“EGS”) design. Furthermore, the intent for this test problem suite is to provide baseline comparison data that demonstrates the performance of the FALCON solution methods. The simulation problems vary in complexity from singly mechanical or thermo process, to coupled thermo-hydro-mechanical processes in geological porous media. Numerical results obtained by FALCON agreed well with either the available analytical solution or experimental data, indicating the verified and validated implementation of these capabilities in FALCON. Some form of solution verification has been attempted to identify sensitivities in the solution methods, where possible, and suggest best practices when using the FALCON code.

15 GEOTHERMAL ENERGY↗
DOE OSTI2013

Weighted simultaneous algebraic reconstruction technique for tomosynthesis imaging of objects with high-attenuation features

Purpose: This paper introduces a nonlinear weighting scheme into the backprojection operation within the simultaneous algebraic reconstruction technique (SART). It is designed for tomosynthesis imaging of objects with high-attenuation features in order to reduce limited angle artifacts. Methods: The algorithm estimates which projections potentially produce artifacts in a voxel. The contribution of those projections into the updating term is reduced. In order to identify those projections automatically, a four-dimensional backprojected space representation is used. Weighting coefficients are calculated based on a dissimilarity measure, evaluated in this space. For each combination of an angular view direction and a voxel position an individual weighting coefficient for the updating term is calculated. Results: The feasibility of the proposed approach is shown based on reconstructions of the following real three-dimensional tomosynthesis datasets: a mammography quality phantom, an apple with metal needles, a dried finger bone in water, and a human hand. Datasets have been acquired with a Siemens Mammomat Inspiration tomosynthesis device and reconstructed using SART with and without suggested weighting. Out-of-focus artifacts are described using line profiles and measured using standard deviation (STD) in the plane and below the plane which contains artifact-causing features. Artifacts distribution in axial direction is measured using an artifact spread function (ASF). The volumes reconstructed with the weighting scheme demonstrate the reduction of out-of-focus artifacts, lower STD (meaning reduction of artifacts), and narrower ASF compared to nonweighted SART reconstruction. It is achieved successfully for different kinds of structures: point-like structures such as phantom features, long structures such as metal needles, and fine structures such as trabecular bone structures. Conclusions: Results indicate the feasibility of the proposed algorithm to reduce typical tomosynthesis artifacts produced by high-attenuation features. The proposed algorithm assigns weighting coefficients automatically and no segmentation or tissue-classification steps are required. The algorithm can be included into various iterative reconstruction algorithms with an additive updating strategy. It can also be extended to computed tomography case with the complete set of angular data.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

The Liquid Argon Software Toolkit (LArSoft): Goals, Status and Plan

LArSoft is a toolkit that provides a software infrastructure and algorithms for the simulation, reconstruction and analysis of events in Liquid Argon Time Projection Chambers (LArTPCs). It is used by the ArgoNeuT, LArIAT, MicroBooNE, DUNE (including 35ton prototype and ProtoDUNE) and SBND experiments. The LArSoft collaboration provides an environment for the development, use, and sharing of code across experiments. The ultimate goal is to develop fully automatic processes for reconstruction and analysis of LArTPC events. The toolkit is based on the art framework and has a well-defined architecture to interface to other packages, including to GEANT4 and GENIE simulation software and the Pandora software development kit for pattern recognition. It is designed to facilitate and support the evolution of algorithms including their transition to new computing platforms. The development of the toolkit is driven by the scientific stakeholders involved. The core infrastructure includes standard definitions of types and constants, means to input experiment geometries as well as meta and event- data in several formats, and relevant general utilities. Examples of algorithms experiments have contributed to date are: photon-propagation; particle identification; hit finding, track finding and fitting; electromagnetic shower identification and reconstruction. We report on the status of the toolkit and plans for future work.

Pordes, Rush↗
DOE OSTI2019

Parallel beamlet dose calculation via beamlet contexts in a distributed multi‐GPU framework

Purpose Dose calculation is one of the most computationally intensive, yet essential tasks in the treatment planning process. With the recent interest in automatic beam orientation and arc trajectory optimization techniques, there is a great need for more efficient model‐based dose calculation algorithms that can accommodate hundreds to thousands of beam candidates at once. Foundational work has shown the translation of dose calculation algorithms to graphical processing units ( GPU s), lending to remarkable gains in processing efficiency. But these methods provide parallelization of dose for only a single beamlet, serializing the calculation of multiple beamlets and under‐utilizing the potential of modern GPU s. In this paper, the authors propose a framework enabling parallel computation of many beamlet doses using a novel beamlet context transformation and further embed this approach in a scalable network of multi‐ GPU computational nodes. Methods The proposed context‐based transformation separates beamlet‐local density and TERMA into distinct beamlet contexts that independently provide sufficient data for beamlet dose calculation. Beamlet contexts are arranged in a composite context array with dosimetric isolation, and the context array is subjected to a GPU collapsed‐cone convolution superposition procedure, producing the set of beamlet‐specific dose distributions in a single pass. Dose from each context is converted to a sparse representation for efficient storage and retrieval during treatment plan optimization. The context radius is a new parameter permitting flexibility between the speed and fidelity of the dose calculation process. A distributed manager‐worker architecture is constructed around the context‐based GPU dose calculation approach supporting an arbitrary number of worker nodes and resident GPU s. Phantom experiments were executed to verify the accuracy of the context‐based approach compared to Monte Carlo and a reference CPU ‐ CCCS implementation for single beamlets and broad beams composed by addition of beamlets. Dose for representative 4 π beam sets was calculated in lung and prostate cases to compare its efficiency with that of an existing beamlet‐sequential GPU ‐ CCCS implementation. Code profiling was also performed to evaluate the scalability of the framework across many networked GPU s. Results The dosimetric accuracy of the context‐based method displays <1.35% and 2.35% average error from the existing serialized CPU ‐ CCCS algorithm and Monte Carlo simulation for beamlet‐specific PDD s in water and slab phantoms, respectively. The context‐based method demonstrates substantial speedup of up to two orders of magnitude over the beamlet‐sequential GPU ‐ CCCS method in the tested configurations. The context‐based framework demonstrates near linear scaling in the number of distributed compute nodes and GPU s employed, indicating that it is flexible enough to meet the performance requirements of most users by simply increasing the hardware utilization. Conclusions The context‐based approach demonstrates a new expectation of performance for beamlet‐based dose calculation methods. This approach has been successful in accelerating the dose calculation process for very large‐scale treatment planning problems ‐ such as automatic 4 π IMRT beam orientation and VMAT arc trajectory selection, with hundreds of thousands of beamlets ‐ in clinically feasible timeframes. The flexibility of this framework makes it as a strong candidate for use in a variety of other very large‐scale treatment planning tasks and clinical workflows.

Neph, Ryan↗
DOE OSTI2017

Development of Fusion Neutronics Methodology, Software, and Testing Facility in FDS Team

Fusion neutronics including methodology, software, experiments, etc. is studied concerning the generation, utilizing and protection of neutrons in fusion facilities. Achievements have been made in the past decades to solve the new problems fusion neutronics. However there is still some way to go such as following. Automatic modeling method based on CAD has been developed to avoid the heavy burden of manual geometry description. The pre-processing costs much time to simplify the CAD model for converting to Constructive Solid Geometry (CSG), 2D or 3D mesh based geometry. This problem has become a new focus. Deep penetration of radiation shielding is notable problem for fusion. Adaptive variance reduction methods such as Consistent Adjoint Driven Importance Sampling (CADIS) were developed to accelerate convergence. However the gap still exists between engineering requirements and current methodology. The geometry simplification, homogenizing based tally, error in nuclear data and error propagation in coupled calculation, etc., causes considerable uncertainty, which leads to 'safety factor' applied in simulation results to achieve required safety margin. It can significantly raise the cost and even makes it impractical to initiate the fusion facilities. Uncertainty qualification, error analysis and error propagation problem should be studied. New methodology and software should be validated. For example the (n,p), (n,a), (n,t) cross section of Cu, Fe, W, Ni which is important for evaluation of material radiation damage is absent. The validation of neutronics properties of special components such as first wall, blanket and coil are needed. The D-T fusion neutron generator can directly represent the neutron environment in fusion reactors. However the neutron yield is several orders of magnitudes lower than requirements. To resolve obstacles in fusion neutronics, broad improvements of methodology, software and testing facility have been achieved by FDS Team in recent years and the main progress was introduced in this paper.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

EcoSpec: Highly Equipped Tower-Based Hyperspectral and Thermal Infrared Automatic Remote Sensing System for Investigating Plant Responses to Environmental Changes

Despite an advanced ability to forecast ecosystem functions and climate at regional and global scales, little is known about relationships between local variations in water and carbon fluxes and large-scale phenomena. To enable data collection of local-scale ecosystem functions to support such investigations, we developed the EcoSpec system, a highly equipped remote sensing system that houses a hyperspectral radiometer (350–2500 nm) and five optical and infrared sensors in a compact tower. Its custom software controls the sequence and timing of movement of the sensors and system components and collects measurements at 12 locations around the tower. The data collected using the system was processed to remove sun-angle effects, and spectral vegetation indices computed from the data (i.e., the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Photochemical Reflectance Index (PRI), and Moisture Stress Index (MSI)) were compared with the fraction of photochemically active radiation (fPAR) and canopy temperature. The results showed that the NDVI, NDWI, and PRI were strongly correlated with fPAR; the MSI was correlated with canopy temperature at the diurnal scale. These correlations suggest that this type of near-surface remote sensing system would complement existing observatories to validate satellite remote sensing observations and link local and large-scale phenomena to improve our ability to forecast ecosystem functions and climate. The system is also relevant for precision agriculture to study crop growth, detect disease and pests, and compare traits of cultivars.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

HEPOM: Using Graph Neural Networks for the Accelerated Predictions of Hydrolysis Free Energies in Different pH Conditions

Hydrolysis is a fundamental family of chemical reactions where water facilitates the cleavage of bonds. The process is ubiquitous in biological and chemical systems, owing to water’s remarkable versatility as a solvent. However, accurately predicting the feasibility of hydrolysis through computational techniques is a difficult task, as subtle changes in reactant structure like heteroatom substitutions or neighboring functional groups can influence the reaction outcome. Furthermore, hydrolysis is sensitive to the pH of the aqueous medium, and the same reaction can have different reaction properties at different pH conditions. In this work, we have combined reaction templates and high-throughput ab initio calculations to construct a diverse data set of hydrolysis free energies. The developed framework automatically identifies reaction centers, generates hydrolysis products, and utilizes a trained graph neural network (GNN) model to predict ΔG values for all potential hydrolysis reactions in a given molecule. The long-term goal of the work is to develop a data-driven, computational tool for high-throughput screening of pH-specific hydrolytic stability and the rapid prediction of reaction products, which can then be applied in a wide array of applications including chemical recycling of polymers and ion-conducting membranes for clean energy generation and storage.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

GEDet: Detecting Erroneous Nodes with A Few Examples

Detecting nodes with erroneous values in real-world graphs re- mains challenging due to the lack of examples and various error scenarios. We demonstrate GEDet, an error detection engine that can detect erroneous nodes in graphs with a few examples. The GEDet framework tackles error detection as a few-shot node classification problem. We invite the attendees to experience the following unique features. (1) Few-shot detection. Users only need to provide a few examples of erroneous nodes to perform error detection with GEDet. GEDet achieves desirable accuracy with (a) a graph augmentation module, which automatically generates synthetic examples to learn the classifier, and (b) an adversarial detection module, which improves classifiers to better distinguish erroneous nodes from both cleaned nodes and synthetic examples. We show that GEDet significantly improves the state-of-the-art error detection methods. (2) Diverse error scenarios. GEDet profiles data errors with a built-in library of transformation functions from correct values to errors. Users can also easily “plug in” new error types or examples. (3) User-centric detection. GEDet supports (a) an active learning mode to engage users to verify detected results, and adapts the error detection process accordingly; and (b) visual interfaces to interpret and track detected errors.

Guan, Sheng↗
DOE OSTI2020

DDxNet: a deep learning model for automatic interpretation of electronic health records, electrocardiograms and electroencephalograms

Effective patient care mandates rapid, yet accurate, diagnosis. With the abundance of non-invasive diagnostic measurements and electronic health records (EHR), manual interpretation for differential diagnosis has become time-consuming and challenging. This has led to wide-spread adoption of AI-powered tools, in pursuit of improving accuracy and efficiency of this process. While the unique challenges presented by each modality and clinical task demand customized tools, the cumbersome process of making problem-specific choices has triggered the critical need for a generic solution to enable rapid development of models in practice. In this spirit, we develop DDxNet, a deep architecture for time-varying clinical data, which we demonstrate to be well-suited for diagnostic tasks involving different modalities (ECG/EEG/EHR), required level of characterization (abnormality detection/phenotyping) and data fidelity (single-lead ECG/22-channel EEG). Using multiple benchmark problems, we show that DDxNet produces high-fidelity predictive models, and sometimes even provides significant performance gains over problem-specific solutions.

60 APPLIED LIFE SCIENCES↗