Results for “randomized sampling” — Page 15

DOE OSTI2024

Composite Qdrift-product formulas for quantum and classical simulations in real and imaginary time

Recent study has shown that it can be advantageous to implement a composite channel that partitions the Hamiltonian H for a given simulation problem into subsets A and B such that H = A + B , where the terms in A are simulated with a Trotter-Suzuki channel and the B terms are randomly sampled via the Qdrift algorithm. Here we extend Qdrift and composite product formulas to imaginary time, formulating candidate classical algorithms for quantum Monte Carlo calculations. We upper bound the induced Schatten- 1 → 1 norm on both imaginary-time Qdrift and composite channels. Another recent result demonstrated that simulations of lattice Hamiltonians containing geometrically local interactions can be improved using a Lieb-Robinson argument to decompose H into subsets that contain only terms supported on that subset of the lattice. Here, we provide a quantum algorithm by unifying this result with the composite approach into “local composite channels” and we upper bound the diamond distance. We provide exact numerical simulations of algorithmic cost by counting the number of gates of the form e − i H j t and e − H j β to meet a certain error tolerance ε . In doing so, we optimize the partitioning into sets A and B using gradient boosted tree models from machine learning. These numerical studies are important given that product formulas have been historically known to outperform analytic upper bounds. We show constant factor advantages for a variety of interesting Hamiltonians, the maximum of which is a ≈ 20 -fold speedup that occurs in the simulation of Jellium. Published by the American Physical Society 2024

Pocrnic, Matthew (ORCID:0000000203089376)↗

DOE OSTI2025

Aspects of propagator sparsening in lattice QCD

In lattice field theory, field sparsening aims to replace quantum fields, or objects constructed from them, with approximations that preserve the appropriate symmetries and maintain many aspects of the physics that the fields determine. For example, an effective sparsening of a quark propagator provides an efficient map from a quark propagator on a fine lattice geometry to a quark propagator defined on a coarser geometry in order to reduce storage and computational costs of subsequent calculational stages while maintaining long-distance correlations and corresponding low-energy physical information. Previous studies have focused on decimating lattice sites or randomly sampling lattice sites to reduce the size of the propagator and subsequent costs of Wick contractions. Here, we extend the study of sparsening to incorporate covariant averaging of spatial sites and examine the effects on two-point and three-point correlation functions involving various hadrons. We find that sparsening is most effective in reproducing the unsparsened versions of these correlation functions when weighted covariant-averaging is sequentially applied many times.

Lattice QCD↗

DOE OSTI2021

Mitigating Catastrophic Forgetting in Deep Learning in a Streaming Setting Using Historical Summary

Recent advancements in scientific equipment and the adaptation of electronics and the Internet of Things (IoT) in our everyday lives resulted in large and complex data production at a high rate. Making meaningful and timely knowledge discovery at a modest cost from this big data is difficult for computing power and storage limitations. Training deep learning models incrementally in a streaming setting can help us with overcoming these limitations. However, in a well-known phenomenon named catastrophic forgetting, incrementally trained models increasingly perform poorly on the past data. To mitigate catastrophic forgetting in training in a streaming setting, we propose constructing a historical summary over time and use the summary with newly arrived data during incremental training. We propose various data summarization techniques such as random sampling, micro clustering, coreset computation, and Auto Encoders to counteract catastrophic forgetting. We built a pipeline for incremental training with a historical summary for training deep learning models for streaming data. We demonstrate the effectiveness of historical summary in mitigating catastrophic forgetting using three case studies involving three different deep learning applications: an Artificial Neural Network (ANN) for classification task on MNIST dataset, a language model (RNN-LM) on the WikiText2 dataset, and a Convolutional Neural Network (CNN), ResNet50 to classify the ImageNet dataset. Through the training of the models, we observe that catastrophic forgetting is evident in ANN and CNN but not in an RNN. For the first task, our method recovers up to 47.9% lost accuracy due to catastrophic forgetting. For the third task, the historical summary recovers classification accuracy by up to 25%. For the second task, though there is not proof of catastrophic forgetting, the training performance (PPL) improves by up to 26% with historical summary.

Dash, Sajal↗

DOE OSTI2018

Rapid determination of soil classes in soil profiles using vis–NIR spectroscopy and multiple objectives mixed support vector classification

Summary Visible‐near infrared (vis–NIR) spectroscopy can reveal various soil properties and facilitate soil classification. However, few studies have attempted to classify vertical soil profiles that contain several genetic horizons. Here, we propose the ‘multiple objectives mixed support vector classification’ (MOM–SVC) method to classify soil profiles. A total of 130 soil profiles were collected from genetic horizons in Zhejiang Province, China. After laboratory analysis, soil profiles were classified according to the Chinese Soil Taxonomy system. Vis–NIR spectra were recorded from each genetic horizon of each soil profile and were then pre‐processed. We performed the MOM–SVC method as follows: (i) created a support vector machine (SVM) model (one‐versus‐one approach) using spectral data from all soil horizons in calibration profiles, (ii) applied the SVM model on each horizon of the profile to be predicted, (iii) extracted ‘votes’ from each horizon and mixed (or summarized) them into the votes of each profile to be predicted and (iv) classified each profile by the majority‐voting method. We also investigated whether the additional input of auxiliary soil information (e.g. moist soil colour, soil organic matter and soil texture), which could be measured easily or be well predicted by vis–NIR spectroscopy, could improve the accuracy of soil classification when combined with it. Independent validation results showed that the MOM–SVC method performed better at the soil order level than at the suborder level. Adding auxiliary soil information to the classification model improved the overall accuracy of classification at the soil order level. The proposed MOM–SVC method provides a fast objective diagnostic of soil classes for use in soil surveys and can help to update soil databases when a more objective soil classification system is developed. Highlights The MOM–SVC method can be used to classify soil profiles objectively with a variety of soil horizons. Stratified random sampling was used to quantify prediction uncertainty in classification MOM–SVC can predict soil orders with greater accuracy than suborders. Adding auxiliary soil information into the classification model improved prediction accuracy.

Chen, S.↗

DOE OSTI2018

Quantifying climate–growth relationships at the stand level in a mature mixed‐species conifer forest

Abstract A range of environmental factors regulate tree growth; however, climate is generally thought to most strongly influence year‐to‐year variability in growth. Numerous dendrochronological (tree‐ring) studies have identified climate factors that influence year‐to‐year variability in growth for given tree species and location. However, traditional dendrochronology methods have limitations that prevent them from adequately assessing stand‐level (as opposed to species‐level) growth. We argue that stand‐level growth analyses provide a more meaningful assessment of forest response to climate fluctuations, as well as the management options that may be employed to sustain forest productivity. Working in a mature, mixed‐species stand at the Howland Research Forest of central Maine, USA , we used two alternatives to traditional dendrochronological analyses by (1) selecting trees for coring using a stratified (by size and species), random sampling method that ensures a representative sample of the stand, and (2) converting ring widths to biomass increments, which once summed, produced a representation of stand‐level growth, while maintaining species identities or canopy position if needed. We then tested the relative influence of seasonal climate variables on year‐to‐year variability in the biomass increment using generalized least squares regression, while accounting for temporal autocorrelation. Our results indicate that stand‐level growth responded most strongly to previous summer and current spring climate variables, resulting from a combination of individualistic climate responses occurring at the species‐ and canopy‐position level. Our climate models were better fit to stand‐level biomass increment than to species‐level or canopy‐position summaries. The relative growth responses (i.e., percent change) predicted from the most influential climate variables indicate stand‐level growth varies less from to year‐to‐year than species‐level or canopy‐position growth responses. By assessing stand‐level growth response to climate, we provide an alternative perspective on climate–growth relationships of forests, improving our understanding of forest growth dynamics under a fluctuating climate.

Teets, Aaron↗

DOE OSTI2020

Soybean ( Glycine max ) Haplotype Map (GmHapMap): a universal resource for soybean translational and functional genomics

Here, we describe a worldwide haplotype map for soybean (GmHapMap) constructed using whole-genome sequence data for 1007 Glycine max accessions and yielding 14.9 million variants as well as 4.3 M tag single-nucleotide polymorphisms (SNPs). When sampling random subsets of these accessions, the number of variants and tag SNPs plateaued beyond approximately 800 and 600 accessions, respectively. This suggests extensive coverage of diversity within the cultivated soybean. GmHapMap variants were imputed onto 21 618 previously genotyped accessions with up to 96% success for common alleles. A local association analysis was performed with the imputed data using markers located in a 1-Mb region known to contribute to seed oil content and enabled us to identify a candidate causal SNP residing in the NPC1 gene. We determined gene-centric haplotypes (407 867 GCHs) for the 55 589 genes and showed that such haplotypes can help to identify alleles that differ in the resulting phenotype. Finally, we predicted 18 031 putative loss-of-function (LOF) mutations in 10 662 genes and illustrated how such a resource can be used to explore gene function. The GmHapMap provides a unique worldwide resource for applied soybean genomics and breeding.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2016

SU-C-BRB-05: Determining the Adequacy of Auto-Contouring Via Probabilistic Assessment of Ensuing Treatment Plan Metrics in Comparison with Manual Contours

Purpose: To determine if auto-contour and manual-contour—based plans differ when evaluated with respect to probabilistic coverage metrics and biological model endpoints for prostate IMRT. Methods: Manual and auto-contours were created for 149 CT image sets acquired from 16 unique prostate patients. A single physician manually contoured all images. Auto-contouring was completed utilizing Pinnacle’s Smart Probabilistic Image Contouring Engine (SPICE). For each CT, three different 78 Gy/39 fraction 7-beam IMRT plans are created; PD with drawn ROIs, PAS with auto-contoured ROIs, and PM with auto-contoured OARs with the manually drawn target. For each plan, 1000 virtual treatment simulations with different sampled systematic errors for each simulation and a different sampled random error for each fraction were performed using our in-house GPU-accelerated robustness analyzer tool which reports the statistical probability of achieving dose-volume metrics, NTCP, TCP, and the probability of achieving the optimization criteria for both auto-contoured (AS) and manually drawn (D) ROIs. Metrics are reported for all possible cross-evaluation pairs of ROI types (AS,D) and planning scenarios (PD,PAS,PM). Bhattacharyya coefficient (BC) is calculated to measure the PDF similarities for the dose-volume metric, NTCP, TCP, and objectives with respect to the manually drawn contour evaluated on base plan (D-PD). Results: We observe high BC values (BC≥0.94) for all OAR objectives. BC values of max dose objective on CTV also signify high resemblance (BC≥0.93) between the distributions. On the other hand, BC values for CTV’s D95 and Dmin objectives are small for AS-PM, AS-PD. NTCP distributions are similar across all evaluation pairs, while TCP distributions of AS-PM, AS-PD sustain variations up to %6 compared to other evaluated pairs. Conclusion: No significant probabilistic differences are observed in the metrics when auto-contoured OARs are used. The prostate auto-contour needs improvement to achieve clinically equivalent plans.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

SU-F-T-202: An Evaluation Method of Lifetime Attributable Risk for Comparing Between Proton Beam Therapy and Intensity Modulated X-Ray Therapy for Pediatric Cancer Patients by Averaging Four Dose-Response Models for Carcinoma Induction

Purpose: To examine how much lifetime attributable risk (LAR) as an in silico surrogate marker of radiation-induced secondary cancer would be lowered by using proton beam therapy (PBT) in place of intensity modulated x-ray therapy (IMXT) in pediatric patients. Methods: From 242 pediatric patients with cancers who were treated with PBT, 26 patients were selected by random sampling after stratification into four categories: a) brain, head, and neck, b) thoracic, c) abdominal, and d) whole craniospinal (WCNS) irradiation. IMXT was re-planned using the same computed tomography and region of interest. Using dose volume histogram (DVH) of PBT and IMXT, the LAR of Schneider et al. was calculated for the same patient. The published four dose-response models for carcinoma induction: i) full model, ii) bell-shaped model, iii) plateau model, and ix) linear model were tested for organs at risk. In the case that more than one dose-response model was available, the LAR for this patient was calculated by averaging LAR for each dose-response model. Results: Calculation of the LARs of PBT and IMXT based on DVH was feasible for all patients. The mean±standard deviation of the cumulative LAR difference between PBT and IMXT for the four categories was a) 0.77±0.44% (n=7, p=0.0037), b) 23.1±17.2%,(n=8, p=0.0067), c) 16.4±19.8% (n=8, p=0.0525), and d) 49.9±21.2% (n=3, p=0.0275, one tailed t-test), respectively. The LAR was significantly lower by PBT than IMXT for the the brain, head, and neck region, thoracic region, and whole craniospinal irradiation. Conclusion: In pediatric patients who had undergone PBT, the LAR of PBT was significantly lower than the LAR of IMXT estimated by in silico modeling. This method was suggested to be useful as an in silico surrogate marker of secondary cancer induced by different radiotherapy techniques. This research was supported by the Translational Research Network Program, JSPS KAKENHI Grant No. 15H04768 and the Global Institution for Collaborative Research and Education (GI-CoRE), Hokkaido University, founded by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

WE-AB-207B-07: Dose Cloud: Generating “Big Data” for Radiation Therapy Treatment Plan Optimization Research

Purpose: To provide a tool to generate large sets of realistic virtual patient geometries and beamlet doses for treatment optimization research. This tool enables countless studies exploring the fundamental interplay between patient geometry, objective functions, weight selections, and achievable dose distributions for various algorithms and modalities. Methods: Generating realistic virtual patient geometries requires a small set of real patient data. We developed a normalized patient shape model (PSM) which captures organ and target contours in a correspondence-preserving manner. Using PSM-processed data, we perform principal component analysis (PCA) to extract major modes of variation from the population. These PCA modes can be shared without exposing patient information. The modes are re-combined with different weights to produce sets of realistic virtual patient contours. Because virtual patients lack imaging information, we developed a shape-based dose calculation (SBD) relying on the assumption that the region inside the body contour is water. SBD utilizes a 2D fluence-convolved scatter kernel, derived from Monte Carlo simulations, and can compute both full dose for a given set of fluence maps, or produce a dose matrix (dose per fluence pixel) for many modalities. Combining the shape model with SBD provides the data needed for treatment plan optimization research. Results: We used PSM to capture organ and target contours for 96 prostate cases, extracted the first 20 PCA modes, and generated 2048 virtual patient shapes by randomly sampling mode scores. Nearly half of the shapes were thrown out for failing anatomical checks, the remaining 1124 were used in computing dose matrices via SBD and a standard 7-beam protocol. As a proof of concept, and to generate data for later study, we performed fluence map optimization emphasizing PTV coverage. Conclusions: We successfully developed and tested a tool for creating customizable sets of virtual patients suitable for large-scale radiation therapy optimization research.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

TH-A-207B-00: Shear-Wave Imaging and a QIBA US Biomarker Update

Imaging of tissue elastic properties is a relatively new and powerful approach to one of the oldest and most important diagnostic tools. Imaging of shear wave speed with ultrasound is has been added to most high-end ultrasound systems. Understanding this exciting imaging mode aiding its most effective use in medicine can be a rewarding effort for medical physicists and other medical imaging and treatment professionals. Assuring consistent, quantitative measurements across the many ultrasound systems in a typical imaging department will constitute a major step toward realizing the great potential of this technique and other quantitative imaging. This session will target these two goals with two presentations. A. Basics and Current Implementations of Ultrasound Imaging of Shear Wave Speed and Elasticity - Shigao Chen, Ph.D. Learning objectives-To understand: Introduction: Importance of tissue elasticity measurement Strain vs. shear wave elastography (SWE), beneficial features of SWE The link between shear wave speed and material properties, influence of viscosity Generation of shear waves External vibration (Fibroscan) ultrasound radiation force Point push Supersonic push (Aixplorer) Comb push (GE Logiq E9) Detection of shear waves Motion detection from pulse-echo ultrasound Importance of frame rate for shear wave imaging Plane wave imaging detection How to achieve high effective frame rate using line-by-line scanners Shear wave speed calculation Time to peak Random sample consensus (RANSAC) Cross correlation Sources of bias and variation in SWE Tissue viscosity Transducer compression or internal pressure of organ Reflection of shear waves at boundaries B. Elasticity Imaging System Biomarker Qualification and User Testing of Systems – Brian Garra, M.D. Learning objectives-To understand: Goals Review the need for quantitative medical imaging Provide examples of quantitative imaging biomarkers Acquaint the participant with the purpose of the RSNA Quantitative Imaging Biomarker Alliance and the need for such an organization Review the QIBA process for creating a quantitative biomarker Summarize steps needed to verify adherence of site, operators, and imaging systems to a QIBA profile Underlying Premise and Assumptions Objective, quantifiable results are needed to enhance the value of diagnostic imaging in clinical practice Reasons for quantification Evidence based medicine requires objective, not subjective observer data Computerized decision support tools (eg CAD) generally require quantitative input. Quantitative, reproducible measures are more easily used to develop personalized molecular medical diagnostic and treatment systems What is quantitative imaging? Definition from Imaging Metrology Workshop The Quantitative Imaging Biomarker Alliance Formation 2008 Mission Structure Example Imaging Biomarkers Being Explored Biomarker Selection Groundwork Draft Protocol for imaging and data evaluation QIBA Profile Drafting Equipment and Site Validation Technical Clinical Site and Equipment QA and Compliance Checking Ultrasound Elasticity Estimation Biomarker US Elasticity Estimation Background Current Status and Problems Biomarker Selection-process and outcome US SWS for Liver Fibrosis Biomarker Work Groundwork Literature search and analysis results Phase I phantom testing-Elastic phantoms Phase II phantom testing-Viscoelastic phantoms Digital Simulated Data Protocol and Profile Drafting Protocol: based on UPICT and existing literature and standards bodies protocols Profile-Current claims, Manufacturer specific appendices What comes after the profile Profile Validation Technical validation Clinical validation QA and Compliance Possible approaches Site Operator testing Site protocol re-evaluation Imaging system Manufacturer testing and attestation User acceptance testing and periodic QA Phantom Tests Digital Phantom Based Testing Standard QA Testing Remediation Schemes Profile Evolution Towards additional applications Towards higher accuracy and precision Supported in part by NIH contract HHSN268201300071C from NIBIB. Collaboration with GE Global Research, no personal support.; S. Chen, Some technologies described in this presentation have been licensed. Mayo Clinic and Dr. Chen have financial interests these technologies.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

TH-A-207B-01: Basics and Current Implementations of Ultrasound Imaging of Shear Wave Speed and Elasticity

Imaging of tissue elastic properties is a relatively new and powerful approach to one of the oldest and most important diagnostic tools. Imaging of shear wave speed with ultrasound is has been added to most high-end ultrasound systems. Understanding this exciting imaging mode aiding its most effective use in medicine can be a rewarding effort for medical physicists and other medical imaging and treatment professionals. Assuring consistent, quantitative measurements across the many ultrasound systems in a typical imaging department will constitute a major step toward realizing the great potential of this technique and other quantitative imaging. This session will target these two goals with two presentations. A. Basics and Current Implementations of Ultrasound Imaging of Shear Wave Speed and Elasticity - Shigao Chen, Ph.D. Learning objectives-To understand: Introduction: Importance of tissue elasticity measurement Strain vs. shear wave elastography (SWE), beneficial features of SWE The link between shear wave speed and material properties, influence of viscosity Generation of shear waves External vibration (Fibroscan) ultrasound radiation force Point push Supersonic push (Aixplorer) Comb push (GE Logiq E9) Detection of shear waves Motion detection from pulse-echo ultrasound Importance of frame rate for shear wave imaging Plane wave imaging detection How to achieve high effective frame rate using line-by-line scanners Shear wave speed calculation Time to peak Random sample consensus (RANSAC) Cross correlation Sources of bias and variation in SWE Tissue viscosity Transducer compression or internal pressure of organ Reflection of shear waves at boundaries B. Elasticity Imaging System Biomarker Qualification and User Testing of Systems – Brian Garra, M.D. Learning objectives-To understand: Goals Review the need for quantitative medical imaging Provide examples of quantitative imaging biomarkers Acquaint the participant with the purpose of the RSNA Quantitative Imaging Biomarker Alliance and the need for such an organization Review the QIBA process for creating a quantitative biomarker Summarize steps needed to verify adherence of site, operators, and imaging systems to a QIBA profile Underlying Premise and Assumptions Objective, quantifiable results are needed to enhance the value of diagnostic imaging in clinical practice Reasons for quantification Evidence based medicine requires objective, not subjective observer data Computerized decision support tools (eg CAD) generally require quantitative input. Quantitative, reproducible measures are more easily used to develop personalized molecular medical diagnostic and treatment systems What is quantitative imaging? Definition from Imaging Metrology Workshop The Quantitative Imaging Biomarker Alliance Formation 2008 Mission Structure Example Imaging Biomarkers Being Explored Biomarker Selection Groundwork Draft Protocol for imaging and data evaluation QIBA Profile Drafting Equipment and Site Validation Technical Clinical Site and Equipment QA and Compliance Checking Ultrasound Elasticity Estimation Biomarker US Elasticity Estimation Background Current Status and Problems Biomarker Selection-process and outcome US SWS for Liver Fibrosis Biomarker Work Groundwork Literature search and analysis results Phase I phantom testing-Elastic phantoms Phase II phantom testing-Viscoelastic phantoms Digital Simulated Data Protocol and Profile Drafting Protocol: based on UPICT and existing literature and standards bodies protocols Profile-Current claims, Manufacturer specific appendices What comes after the profile Profile Validation Technical validation Clinical validation QA and Compliance Possible approaches Site Operator testing Site protocol re-evaluation Imaging system Manufacturer testing and attestation User acceptance testing and periodic QA Phantom Tests Digital Phantom Based Testing Standard QA Testing Remediation Schemes Profile Evolution Towards additional applications Towards higher accuracy and precision Supported in part by NIH contract HHSN268201300071C from NIBIB. Collaboration with GE Global Research, no personal support.; S. Chen, Some technologies described in this presentation have been licensed. Mayo Clinic and Dr. Chen have financial interests these technologies.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

TH-A-207B-02: QIBA Ultrasound Elasticity Imaging System Biomarker Qualification and User Testing of Systems

Imaging of tissue elastic properties is a relatively new and powerful approach to one of the oldest and most important diagnostic tools. Imaging of shear wave speed with ultrasound is has been added to most high-end ultrasound systems. Understanding this exciting imaging mode aiding its most effective use in medicine can be a rewarding effort for medical physicists and other medical imaging and treatment professionals. Assuring consistent, quantitative measurements across the many ultrasound systems in a typical imaging department will constitute a major step toward realizing the great potential of this technique and other quantitative imaging. This session will target these two goals with two presentations. A. Basics and Current Implementations of Ultrasound Imaging of Shear Wave Speed and Elasticity - Shigao Chen, Ph.D. Learning objectives-To understand: Introduction: Importance of tissue elasticity measurement Strain vs. shear wave elastography (SWE), beneficial features of SWE The link between shear wave speed and material properties, influence of viscosity Generation of shear waves External vibration (Fibroscan) ultrasound radiation force Point push Supersonic push (Aixplorer) Comb push (GE Logiq E9) Detection of shear waves Motion detection from pulse-echo ultrasound Importance of frame rate for shear wave imaging Plane wave imaging detection How to achieve high effective frame rate using line-by-line scanners Shear wave speed calculation Time to peak Random sample consensus (RANSAC) Cross correlation Sources of bias and variation in SWE Tissue viscosity Transducer compression or internal pressure of organ Reflection of shear waves at boundaries B. Elasticity Imaging System Biomarker Qualification and User Testing of Systems – Brian Garra, M.D. Learning objectives-To understand: Goals Review the need for quantitative medical imaging Provide examples of quantitative imaging biomarkers Acquaint the participant with the purpose of the RSNA Quantitative Imaging Biomarker Alliance and the need for such an organization Review the QIBA process for creating a quantitative biomarker Summarize steps needed to verify adherence of site, operators, and imaging systems to a QIBA profile Underlying Premise and Assumptions Objective, quantifiable results are needed to enhance the value of diagnostic imaging in clinical practice Reasons for quantification Evidence based medicine requires objective, not subjective observer data Computerized decision support tools (eg CAD) generally require quantitative input. Quantitative, reproducible measures are more easily used to develop personalized molecular medical diagnostic and treatment systems What is quantitative imaging? Definition from Imaging Metrology Workshop The Quantitative Imaging Biomarker Alliance Formation 2008 Mission Structure Example Imaging Biomarkers Being Explored Biomarker Selection Groundwork Draft Protocol for imaging and data evaluation QIBA Profile Drafting Equipment and Site Validation Technical Clinical Site and Equipment QA and Compliance Checking Ultrasound Elasticity Estimation Biomarker US Elasticity Estimation Background Current Status and Problems Biomarker Selection-process and outcome US SWS for Liver Fibrosis Biomarker Work Groundwork Literature search and analysis results Phase I phantom testing-Elastic phantoms Phase II phantom testing-Viscoelastic phantoms Digital Simulated Data Protocol and Profile Drafting Protocol: based on UPICT and existing literature and standards bodies protocols Profile-Current claims, Manufacturer specific appendices What comes after the profile Profile Validation Technical validation Clinical validation QA and Compliance Possible approaches Site Operator testing Site protocol re-evaluation Imaging system Manufacturer testing and attestation User acceptance testing and periodic QA Phantom Tests Digital Phantom Based Testing Standard QA Testing Remediation Schemes Profile Evolution Towards additional applications Towards higher accuracy and precision Supported in part by NIH contract HHSN268201300071C from NIBIB. Collaboration with GE Global Research, no personal support.; S. Chen, Some technologies described in this presentation have been licensed. Mayo Clinic and Dr. Chen have financial interests these technologies.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2020

Stochastic Gradients for Large-Scale Tensor Decomposition

Tensor decomposition is a well-known tool for multiway data analysis. This work proposes using stochastic gradients for efficient generalized canonical polyadic (GCP) tensor decomposition of large-scale tensors. GCP tensor decomposition is a recently proposed version of tensor decomposition that allows for a variety of loss functions such as Bernoulli loss for binary data or Huber loss for robust estimation. Here, the stochastic gradient is formed from randomly sampled elements of the tensor and is efficient because it can be computed using the sparse matricized-tensor times Khatri--Rao product tensor kernel. For dense tensors, we simply use uniform sampling. For sparse tensors, we propose two types of stratified sampling that give precedence to sampling nonzeros. Numerical results demonstrate the advantages of the proposed approach and its scalability to large-scale problems.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2017

Obtaining and Managing Answer Quality for Online Data-Intensive Services

Online data-intensive (OLDI) services use anytime algorithms to compute over large amounts of data and respond quickly. Interactive response times are a priority, so OLDI services parallelize query execution across distributed software components and return best effort answers based on the data so far processed. Omitted data from slow components could lead to better answers, but tracing online how much better the answers could be is difficult. Here, we introduce Ubora, a design method to measure the effect of slow-running components on the quality of answers. Ubora randomly samples online queries and executes them a second time. The first online execution omits data from slow components and provides interactive answers. The second execution uses mature results from intermediate components completed after the online execution finishes. Ubora uses memoization to speed up mature executions by replaying network messages exchanged between components. Our systems-level implementation works for a wide range of services, including Hadoop/Yarn, Apache Lucene, the EasyRec Recommendation Engine, and the OpenEphyra question-answering system. Ubora computes answer quality with more mature executions per second than competing approaches that do not use memoization. With Ubora, we illustrate that answer quality is effective at guiding online admission control. While achieving the same answer quality on high-priority queries, our adaptive controller had 55% higher peak throughput on low-priority queries than a competing controller guided by the rate of timeouts.

answer quality↗

DOE OSTI2017

Use of a Reduced Order Model (ROM) to Simulate SOFC Performance in System Models

Numerical models for design and optimization of fuel cell-based power generation systems require an accurate representation of the SOFC stack performance over a wide range of operating conditions. Response surface techniques were used with detailed SOFC stack model results to create a computationally efficient reduced order model (ROM) of the stack that retains desirable information about its internal state, which can be important in system optimization studies to preclude operation under off-design conditions that can adversely impact overall reliability. A tool was developed to randomly sample selected input parameters, obtain stack solutions for these sampled cases, perform regression to obtain response surfaces for stack performance and other metrics of interest, and export the response surfaces as a ROM. The ROM was then successfully implemented in an Aspen Plus® system model for a natural gas fuel cell (NGFC) power system. Additional work to characterize the ROM’s approximation error is ongoing.

Modeling & Simulation↗

DOE OSTI2020

Tidal Disruption Event Galaxy Binner

This software simulates astronomical survey detections of tidal disruptions of stars by super-massive black holes. It begins with the synthetic galaxy catalogue described in van Velzen 2008 (https://arxiv.org/abs/1707.03458). The stellar disruption rate in each galaxy is estimated based on Stone & Metzger 2016 (https://arxiv.org/abs/1410.7772). Based on these rates, and the present-day stellar mass function in the galaxy, disruptions are randomly sampled, and the properties of the resulting flares are sampled based on empirical distributions. The code also accounts for obscuration by dust in the host galaxy. Finally, the survey selection effects are applied. The detectable simulated flares are stored in a database, allowing histograms of their properties to be created.

Roth, NathanielJ.↗

DOE OSTI2023

Reliability of Open Public Electric Vehicle Direct Current Fast Chargers

The aim was to systematically evaluate the usability of all public electric vehicles (EV) direct current fast chargers (DCFC) in the San Francisco region. To achieve a rapid transition to EVs, a highly reliable and easy to use charging infrastructure is critical to building confidence among consumers. The functionality and usability of all 182 open, public DCFC charging stations with CCS connectors (combined charging system) in the 9 counties of the Bay Area were tested (655 electric vehicle service equipment (EVSE) ports). An EVSE was classified as functional if it charged an EV for 2 minutes. Overall, 73.3% of the 655 EVSEs were functional. The causes of the nonfunctioning EVSEs (23.5%) were blank or unresponsive screens or error messages; payment system failures; charge initiation failures; network failures; or broken connectors. In addition, the cable was too short to reach the EV inlet for 3.2% of the EVSEs. A random sampling of 10% of the EVSEs, approximately 8 days after the first evaluation, found no overall change in functionality. The level of functionality found with field testing conflicts with the 95–98% uptime reported by the EV service providers (EVSPs) who operate the EV charging stations. There is a need for precise and verifiable definitions of uptime, downtime, and excluded time, as applied to public EV chargers. In conclusion, the level of failure of the existing public EV DCFC charge infrastructure highlights the importance of improving the system design and maintenance to improve adoption of EVs.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2024

A graphics processing unit accelerated sparse direct solver and preconditioner with block low rank compression

We present the GPU implementation efforts and challenges of the sparse solver package STRUMPACK. The code is made publicly available on github with a permissive BSD license. STRUMPACK implements an approximate multifrontal solver, a sparse LU factorization which makes use of compression methods to accelerate time to solution and reduce memory usage. Multiple compression schemes based on rank-structured and hierarchical matrix approximations are supported, including hierarchically semi-separable, hierarchically off-diagonal butterfly, and block low rank. Here, in this paper, we present the GPU implementation of the block low rank (BLR) compression method within a multifrontal solver. Our GPU implementation relies on highly optimized vendor libraries such as cuBLAS and cuSOLVER for NVIDIA GPUs, rocBLAS and rocSOLVER for AMD GPUs and the Intel oneAPI Math Kernel Library (oneMKL) for Intel GPUs. Additionally, we rely on external open source libraries such as SLATE (Software for Linear Algebra Targeting Exascale), MAGMA (Matrix Algebra on GPU and Multi-core Architectures), and KBLAS (KAUST BLAS). SLATE is used as a GPU-capable ScaLAPACK replacement. From MAGMA we use variable sized batched dense linear algebra operations such as GEMM, TRSM and LU with partial pivoting. KBLAS provides efficient (batched) low rank matrix compression for NVIDIA GPUs using an adaptive randomized sampling scheme. The resulting sparse solver and preconditioner runs on NVIDIA, AMD and Intel GPUs. Interfaces are available from PETSc, Trilinos and MFEM, or the solver can be used directly in user code. We report results for a range of benchmark applications, using the Perlmutter system from NERSC, Frontier from ORNL, and Aurora from ALCF. For a high frequency wave equation on a regular mesh, using 32 Perlmutter compute nodes, the factorization phase of the exact GPU solver is about 6.5× faster compared to the CPU-only solver. The BLR-enabled GPU solver is about 13.8× faster than the CPU exact solver. For a collection of SuiteSparse matrices, the STRUMPACK exact factorization on a single GPU is on average 1.9× faster than NVIDIA’s cuDSS solver.

97 MATHEMATICS AND COMPUTING↗