Search NASA⌕ Search

SEARCH · Search NASA

Results for “Utility Codes”

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

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

At least 1,459 records · Page 81

Roofline Analysis in the Intel® Advisor to Deliver Optimized Performance for applications on Intel® Xeon Phi™ Processor

In this session we show, in two case studies, how the roofline feature of Intel Advisor has been utilized to optimize the performance of kernels of the XGC1 and PICSAR codes in preparation for Intel Knights Landing architecture. The impact of the implemented optimizations and the benefits of using the automatic roofline feature of Intel Advisor to study performance of large applications will be presented. This demonstrates an effective optimization strategy that has enabled these science applications to achieve up to 4.6 times speed-up and prepare for future exascale architectures. # Goal/Relevance of Session The roofline model [1,2] is a powerful tool for analyzing the performance of applications with respect to the theoretical peak achievable on a given computer architecture. It allows one to graphically represent the performance of an application in terms of operational intensity, i.e. the ratio of flops performed and bytes moved from memory in order to guide optimization efforts. Given the scale and complexity of modern science applications, it can often be a tedious task for the user to perform the analysis on the level of functions or loops to identify where performance gains can be made. With new Intel tools, it is now possible to automate this task, as well as base the estimates of peak performance on measurements rather than vendor specifications. The goal of this session is to demonstrate how the roofline feature of Intel Advisor can be used to balance memory vs. computation related optimization efforts and effectively identify performance bottlenecks. A series of typical optimization techniques: cache blocking, structure refactoring, data alignment, and vectorization illustrated by the kernel cases will be addressed. # Description of the codes ## XGC1 The XGC1 code [3] is a magnetic fusion Particle-In-Cell code that uses an unstructured mesh for its Poisson solver that allows it to accurately resolve the edge plasma of a magnetic fusion device. After recent optimizations to its collision kernel [4], most of the computing time is spent in the electron push (pushe) kernel, where these optimization efforts have been focused. The kernel code scaled well with MPI+OpenMP but had almost no automatic compiler vectorization, in part due to indirect memory addresses and in part due to low trip counts of low-level loops that would be candidates for vectorization. Particle blocking and sorting have been implemented to increase trip counts of low-level loops and improve memory locality, and OpenMP directives have been added to vectorize compute-intensive loops that were identified by Advisor. The optimizations have improved the performance of the pushe kernel 2x on Haswell processors and 1.7x on KNL. The KNL node-for-node performance has been brought to within 30% of a NERSC Cori phase I Haswell node and we expect to bridge this gap by reducing the memory footprint of compute intensive routines to improve cache reuse. ## PICSAR is a Fortran/Python high-performance Particle-In-Cell library targeting at MIC architectures first designed to be coupled with the PIC code WARP for the simulation of laser-matter interaction and particle accelerators. PICSAR also contains a FORTRAN stand-alone kernel for performance studies and benchmarks. A MPI domain decomposition is used between NUMA domains and a tile decomposition (cache-blocking) handled by OpenMP has been added for shared-memory parallelism and better cache management. The so-called current deposition and field gathering steps that compose the PIC time loop constitute major hotspots that have been rewritten to enable more efficient vectorization. Particle communications between tiles and MPI domain has been merged and parallelized. All considered, these improvements provide speedups of 3.1 for order 1 and 4.6 for order 3 interpolation shape factors on KNL configured in SNC4 quadrant flat mode. Performance is similar between a node of cori phase 1 and KNL at order 1 and better on KNL by a factor 1.6 at order 3 with the considered test case (homogeneous thermal plasma).

97 MATHEMATICS AND COMPUTING↗

R “SHINY” GUI DEVELOPMENT FOR URANIUM ISOTOPIC ANALYSIS WITH MATRIX-ASSISTED IONIZATION MASS SPECTROMETRY

The international nuclear safeguards community continues to seek rapid, accurate, and precise characterization capabilities for the in-field measurement of uranium isotopic compositions in nuclear facilities. Mass spectrometry (MS) is considered the “gold standard” for analysis of relatively long-lived actinides such as uranium (U) and plutonium; however, conventional MS analysis often requires time consuming sample preparation and complex analytical methodologies that are difficult to perform in-field or in-facility. Matrix assisted ionization (MAI) is a novel ambient ionization MS technique (i.e., MAI-MS) that potentially addresses these challenges due to the relative simplicity of the ionization phenomenon and ruggedness of ambient MS instrumentation. Savannah River National Laboratory (SRNL, USA) has demonstrated this technique for nanogram-level 235U/238U isotope ratio measurements within seconds, with percent-level analytical uncertainties capable of discriminating depleted, natural, and low-enriched uranium. Current experimental work on developing MAI methods for uranium isotopic analysis has been enabled by parallel development of a comprehensive MAI-MS data analysis suite at SRNL. Development of this bespoke data analysis software was necessary because commercially available ambient MS software is poorly suited for uranium isotope ratio measurement. The effort leverages the power of R, a popular open-source programming language, and Shiny, an R package providing tools for graphical user interface (GUI) and web interface coding. This software allows researchers without any programming experience to harness and utilize R’s considerable data analysis/visualization power.

LaBone, Elizabeth D.↗

Comparison of XSUSA and 'two-step' approaches for full-core uncertainty quantification

While there are multiple sources of error that are introduced into the standard computational regime for simulating reactor cores, rigorous uncertainty analysis methods are available primarily for quantifying the effects of cross section uncertainties. Two methods for propagating cross section uncertainties through core simulators are the XSUSA statistical approach and the 'Two-Step' method. The XSUSA approach, which is based on the SUSA code package, is fundamentally a stochastic sampling method. Alternatively, the Two-Step method utilizes generalized perturbation theory in the first step and stochastic sampling in the second step. The consistency of these two methods in quantifying uncertainties in the multiplication factor and in the core power distribution will be examined in the framework of phase 1-3 of the UAM Benchmark. Using the TMI core as a base model for analysis, the XSUSA and Two-Step methods are applied with certain limitations and the results are compared to those produced by other stochastic sampling-based codes. Based on the uncertainty analysis results, conclusions are made for which method is currently a more viable option for computing uncertainties in burnup and transient calculations. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Thermal Analysis of a Fuel Channel

To better study and understand the efficiency of fuel channels in reactors, a thermal analysis method was developed to conduct a study on any fuel channel to gain a better understanding of the temperature, heat rate, heat flux, and power profiles of such fuel channels. This study is conducted on a typical PWR (Pressurized Water Reactor) fuel assembly. The justification for specifically choosing a PWR fuel channel is because PWRs are the most widely used type of reactors around the world. This study was performed using NEWT code in SCALE 6, ANSYS, and SolidWorks. MATLAB and Excel were also utilized for some of the calculations. The power and heat rate of the fuel rod were calculated, alongside a 2D mapping of temperature, heat flux, directional heat flux, and thermal error on a 3D model of the fuel channel by analyzing the fuel rod and the surrounding water moderator independently.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Additive Manufactured Composite Phase-Change Material for Thermal Energy Storage Applications

Phase-change materials play a critical role in industrial energy storage applications to drive efficiency improvements, thermal energy management, and carbon emissions reductions. Recently, it has been shown that rapid solidification of alloys with metastable immiscibility in the liquid phase has the potential to form unique microstructures in which a low-melting phase is uniformly distributed in a high-melting matrix. This feature can be exploited using additive manufacturing to produce components with complex geometries containing such unique phase-change microstructures. Phase-field simulations utilizing high-performance computing were used to provide a detailed description of the evolution of the active phase during service in terms of their morphology and composition in different polycrystalline matrix grain morphologies that are typically produced during additive manufacturing. Phase field simulations were performed using, MEUMAPPS-SL (Microstructure Evolution Using Massively Parallel Phase-field Simulations – Solid Liquid) code that was developed in-house by the Oak Ridge National Laboratory. The simulations utilized the capabilities of the Kestrel supercomputer at the National Renewable Energy Laboratory. The simulation results were compared with experimental results generated at Siemens Energy, Inc. The results indicate that the kinetics of liquid spreading along grain boundaries is largely determined by the mobility of the triple line along the intersection of the grain boundary liquid and the grain boundary plane.

25 ENERGY STORAGE↗

Materials for Advanced Ultra-Supercritical (A-USC) Steam Turbines --- A-USC Component Demonstration

The U.S. Advanced Ultra-Supercritical (A-USC) Consortium was formed in 2001 as a government/industry program, sponsored by the U.S. Department of Energy (DOE) and the Ohio Coal Development Office (OCDO) and cost shared by industrial and not-for-profit partners. The purpose of the consortium was to advance the state of the art for power generation by evaluating and developing materials that allow the use of advanced steam cycles in coal-based power plants. These advanced cycles, with steam temperatures up to 1400°F (760°C), can increase the efficiency of coal-fired boilers from an average of 35% (current U.S. fleet) to more than 45% higher heating value (HHV) (>49% lower heating value [LHV]). The increase in a plant’s efficiency is limited unless new materials able to withstand these higher operating temperatures and pressures are identified and approved for use. The A-USC Consortium identified these needed materials during earlier phases of the program. It developed the welding and joining techniques along with manufacturing processes for casting and wrought products made from these new high-nickel alloys. It subjected these materials to extensive laboratory and steam loop testing. It then obtained ASME code approval for their use in U.S. boiler systems. The program’s successes leave this last remaining activity (ComTest Phase 2) that the U.S. utility industry has recommended to be accomplished prior to commercialization. The focus of the activity is the evaluation and demonstration of commercial readiness for “full scale” components to be made from these nickel-based alloy materials and provided by a U.S. domestic supply chain that is new to working with these alloys. According to studies completed by the Electric Power Research Institute (EPRI), the cost of an A-USC plant is approximately 20% higher than a non-A-USC plant because of its use of nickel-based alloys needed for the high temperature operating conditions. However, CO 2 reductions of approximately 30% from the current fleet average provide a strong incentive for its consideration. The actual costs and perceived value for CO 2 abatement will determine whether new or retrofitted plants are undertaken, although decisions to build A-USC plants in India would indicate its economic feasibility while also being part of a global carbon emissions strategy. The work by the A-USC Consortium, prior to the start of the ComTest project, has included lab scale and pilot scale materials testing, both in air and oxy-combustion. This testing has included air-cooled and steam-cooled “loops” that were installed into existing operating utility boilers to gain exposure of these materials to realistic conditions of high temperature and corrosion caused by the constituents in the coal ash. The A-USC Consortium also gained ASME Code approval of the Inconel 740 material, has cast and extruded the largest high nickel precipitation hardened alloys, and developed unique welding techniques to avoid problems identified by the competing European program. However, as valuable as these material test loops and accomplishments have been for obtaining information, their scale is below that required to minimize the risk associated for a U.S. utility to build a multibillion-dollar A-USC power plant. To reduce the final identified risk barrier to full-scale commercialization of these advanced materials and systems, the A-USC Consortium (guided by a utility industry advisory committee) has identified the key areas of the technology they desire to see as being capable of full-scale manufacturing and/or fabrication from an identified, capable U.S. domestic supplier base. A significant amount of work was accomplished during Phase 1 to identity the components, as well as the component size, that would be manufactured from advanced alloys such as Inconel 740H or Haynes 282 alloys. Pathways to supply these components for ComTest have been identified, as well as any further development that would be required. The Phase 2 effort used Phase 1 findings for designing these key full-scale components for A-USC boilers and turbines to include large castings; extrusions, forgings, fabrication of water walls and steam loops with headers from advanced materials, raw material (such as pipe extrusion billets) are at the commercial readiness level to permit advancement to a demonstration project. The Phase 2 work scope was addressed by a diverse team, including government, industry, and not-for-profit partners. The work scope under Phase 2 addressed fabrication of components identified as being outside of the proven capabilities of the existing supply chain, including the following: Steam turbine rotor forging and Haynes 282 nozzle carrier casting Superheater and reheater header and tube assemblies Large-diameter pipe extrusions and forgings Test valve articles to support ASME Code approval. In addition, key fabrication steps were completed, including boiler weld overlays and simulated field repairs. Throughout, extensive inspection and quality assurance testing of the components were performed. The team worked to advance ASME Code approval for key components and processes. Although much of the focus of ComTest Phase 2 was the high-temperature nickel-based alloy materials, a broader range of materials were incorporated, which would be representative of the materials used in full-scale A-USC power plant applications and have cross-cutting applicability on other high-temperature power generation options, such as advanced nuclear, supercritical CO 2 cycles, and central solar receivers. This report that has been submitted is organized in the following manner: Section 1 contains an Executive Summary. Section 2 discusses the ComTest project background and organization. Section 3 discusses project management and reporting. Section 4 discusses the procurement of nickel-based alloy and other A-USC materials and components. Section 5 discusses the fabrication of procurement of nickel-based alloy and other A-USC materials and components. Section 6 discusses the fabrication of cast nickel-based A-USC steam turbine components. Section 7 discusses the fabrication of forged nickel-based A-USC steam turbine piping and steam pipe components. Section 8 discusses the qualification of pressure relieve valves (PRVs) for A-USC power plants. Section 9 discusses proposed plans for future evaluation of A-USC components. Section 10 contains the summary and conclusion.

01 COAL, LIGNITE, AND PEAT↗

MPIRUN: A Portable Loader for Multidisciplinary and Multi-Zonal Applications

Multidisciplinary and multi-zonal applications are an important class of applications in the area of Computational Aerosciences. In these codes, two or more distinct parallel programs or copies of a single program are utilized to model a single problem. To support such applications, it is common to use a programming model where a program is divided into several single program multiple data stream (SPMD) applications, each of which solves the equations for a single physical discipline or grid zone. These SPMD applications are then bound together to form a single multidisciplinary or multi-zonal program in which the constituent parts communicate via point-to-point message passing routines. One method for implementing the message passing portion of these codes is with the new Message Passing Interface (MPI) standard. Unfortunately, this standard only specifies the message passing portion of an application, but does not specify any portable mechanisms for loading an application. MPIRUN was developed to provide a portable means for loading MPI programs, and was specifically targeted at multidisciplinary and multi-zonal applications. Programs using MPIRUN for loading and MPI for message passing are then portable between all machines supported by MPIRUN. MPIRUN is currently implemented for the Intel iPSC/860, TMC CM5, IBM SP-1 and SP-2, Intel Paragon, and workstation clusters. Further, MPIRUN is designed to be simple enough to port easily to any system supporting MPI.

Fineberg, Samuel A.↗

Characterization of Filters Loaded With Reactor Strontium Carbonate - 13203

A collection of three highly radioactive filters containing reactor strontium carbonate were being prepared for disposal. All three filters were approximately characterized at the time of manufacture by gravimetric methods. The first filter had been partially emptied, and the quantity of residual activity was uncertain. Dose rate to activity modeling using the Monte-Carlo N Particle (MCNP) code was selected to confirm the gravimetric characterization of the full filters, and to fully characterize the partially emptied filter. Although dose rate to activity modeling using MCNP is a common technique, it is not often used for Bremsstrahlung-dominant materials such as reactor strontium. As a result, different MCNP modeling options were compared to determine the optimum approach. This comparison indicated that the accuracy of the results were heavily dependent on the MCNP modeling details and the location of the dose rate measurement point. The optimum model utilized a photon spectrum generated by the Oak Ridge Isotope Generation and Depletion (ORIGEN) code and dose rates measured at 30 cm. Results from the optimum model agreed with the gravimetric estimates within 15%. It was demonstrated that dose rate to activity modeling can be successful for Bremsstrahlung-dominant radioactive materials. However, the degree of success is heavily dependent on the choice of modeling techniques. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

PIO Tools

This is a set of utilities for reading and manipulating PIO files written in python and C++. These files are intended to be building blocks for other scripts / programs that will use them to do great things. The PIO format itself is described in LA-UR-05-7425 at page 102 and embodied in code LA-CC-05-052. In addition to reading the raw PIO files, the bundled utilities will expand variable that use a compressed sparse row notation

Swaminarayan, Sriram↗

Plug-in Electric Vehicle Infrastructure: A Foundation for Electrified Transportation: Preprint

Plug-in electric vehicles (PEVs)--which include all-electric vehicles and plug-in hybrid electric vehicles--provide a new opportunity for reducing oil consumption by drawing power from the electric grid. To maximize the benefits of PEVs, the emerging PEV infrastructure--from battery manufacturing to communication and control between the vehicle and the grid--must provide access to cleanelectricity, satisfy stakeholder expectations, and ensure safety. Currently, codes and standards organizations are collaborating on a PEV infrastructure plan. Establishing a PEV infrastructure framework will create new opportunities for business and job development initiating the move toward electrified transportation. This paper summarizes the components of the PEV infrastructure, challengesand opportunities related to the design and deployment of the infrastructure, and the potential benefits.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Calculation of the dynamic first electronic hyperpolarizability β(−ω{sub σ}; ω{sub 1}, ω{sub 2}) of periodic systems. Theory, validation, and application to multi-layer MoS{sub 2}

We describe our implementation of a fully analytical scheme, based on the 2n + 1 rule, for computing the coupled perturbed Hartree Fock and Kohn-Sham dynamic first hyperpolarizability tensor β(−ω{sub σ}; ω{sub 1}, ω{sub 2}) of periodic 1D (polymer), 2D (slab), and 3D (crystal) systems in the CRYSTAL code [R. Dovesi et al., Int. J. Quantum Chem. 114, 1287 (2014)], which utilizes local Gaussian type basis sets. The dc-Pockels (dc-P) and second harmonic generation (SHG) tensors are included as special cases. It is verified that (i) symmetry requirements are satisfied; (ii) using LiF as an example, the infinite periodic polymer result agrees with extrapolated finite oligomer calculations and, likewise, for the build-up to a 2D slab and a 3D crystal; (iii) the values converge to the static case for low frequencies; and (iv) the Bishop-deKee dispersion formulas relating dc-P, SHG, and general processes are reproduced through quartic terms. Preliminary SHG calculations on multi-layer MoS{sub 2} satisfactorily reproduce experimental data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Single Primary Heat Extraction and Removal Emulator (SPHERE) Power Transient Testing

For heat pipe cooled microreactor development, it is essential to understand the characteristics of heat pipes and how they function under a wide range of operating conditions. Modeling the startup and shutdown of heat pipe cooled microreactors has been a major challenge for the modeling code. Validating these parameters within the code is key for the development and licensing of heat pipe cooled microreactor designs. Idaho National Laboratory (INL) has completed testing on power transients of a heat pipe with a range of operating conditions. The resulting temperature profiles from this testing can be utilized to aid in the validation of the startup and shutdown portions of the heat pipe modeling code.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Verification and Validation of CFD Simulations involving Twin Rectangular Jets using Steady RANS in Star-CCM+

Analyses such as component design and safety analysis can be conducted through careful application of computational fluid dynamics (CFD) codes. Before these codes can be considered ready for specific analysis, benchmarking (validation) against applicable datasets is needed. The twin jet water facility (TJWF) developed originally at the University of Tennessee, Knoxville has been purposed for aiding in this type of work. The TJWF allows for large data sets to be created using state of the art non-invasive measurement techniques within a mostly transparent testing volume. Utilizing velocity data collected in the TJWF by Wang et. al., the Star-CCM+ code was benchmarked a using steady Reynolds averaged Navier Stokes (RANS) framework for a combining two jet case. Careful efforts to properly conduct verification of the numerical work was undertaken with a sample reported in this summary. The computational results were validated against the experimental work and found to be within acceptable limits. The parameters of interest were the merging and combining points (MP and CP) of the twin jets, and the velocity profiles at increasing heights after injection. Fig.1 provides a graphical depiction of the twin jets combining within a free shear volume with key geometric factors included. The MP and velocity profiles were predicted with reasonable agreement. Whereas, the combining point of the twin jets was found to be over-predicted by {approx}24% difference. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

AXI4MLIR: User-Driven Automatic Host Code Generation for Custom AXI-Based Accelerators

Tensor algebra operations represent an important class of algorithms used across many applications, including machine learning, scientific computing, and data analytics. As a result, the efficient generation of custom accelerators for tensor operations has received increased attention. Previous efforts have produced automated tools enabling users to prototype and explore optimized accelerators. However, little effort has been focused on the host-accelerator interaction in these tools. Efficient use of hardware accelerators requires knowledge about the accelerator's capabilities (operations, data formats, and opcode support), the host CPU microarchitecture (e.g., memory hierarchy), the host-accelerator interface, and the application's features (which code regions should be mapped onto an accelerator). Manually rewriting the original applications to facilitate improved custom accelerator mapping is an error-prone and time-consuming endeavor. To cope with this, we propose AXI4MLIR, a new framework to automatically generate and optimize the communication between the host CPU and arbitrary accelerators that implement linear algebra algorithms. AXI4MLIR extends the MLIR compiler framework to automatically generate efficient host-accelerator driver code for accelerators with AXI-based interfaces. Our compiler extensions enable automatic driver code generation while carefully considering the host's memory hierarchy and target accelerator features. To demonstrate the flexibility and utility of AXI4MLIR, we test it with diverse use cases that include different types of accelerators, tiling scenarios, and dataflow schemes. We compare our experimental results to manual implementations of host-accelerator driver code and find that our approach can reduce CPU cache references by 56% and deliver up to a 1.65x speedup.

Bohm Agostini, Nicolas↗

Optimizing the FEDVR-TDCC Code for Exploring the Quantum Dynamics of Two-Electron Systems in Intense Laser Pulses

To efficiently solve the three-dimensional (3D) time-dependent linear and nonlinear Schrödinger equation, we have developed a large-scale parallel code RSP-FEDVR [B. I. Schneider, L. A. Collins, and S. X. Hu, Phys. Rev. E 73, 036708 (2006)], which combines the finite-element discrete variable representation (FEDVR) with the real-space product algorithm. Using the similar algorithm, we have derived an accurate approach to solve the time-dependent close-coupling (TDCC) equation for exploring two-electron dynamics in linearly polarized intense laser pulses. However, when the number N of partial waves used for the TDCC expansion increases, the FEDVR-TDCC code unfortunately slows down, because the potential-matrix operation scales as ~O(N^2). In this paper, we show that the full potential-matrix operation can be decomposed into a series of small-matrix operations utilizing the sparse property of the N N potential matrix. Such optimization speeds up the FEDVR-TDCC code by an order of magnitude for N=256. This may facilitate the ultimate solution to the 3D two-electron quantum dynamics in ultrashort intense optical laser pulses, where a large number of partial waves are required.

Hu, S X↗

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↗

TH-CD-207A-08: Simulated Real-Time Image Guidance for Lung SBRT Patients Using Scatter Imaging

Purpose: To develop a comprehensive Monte Carlo-based model for the acquisition of scatter images of patient anatomy in real-time, during lung SBRT treatment. Methods: During SBRT treatment, images of patient anatomy can be acquired from scattered radiation. To rigorously examine the utility of scatter images for image guidance, a model is developed using MCNP code to simulate scatter images of phantoms and lung cancer patients. The model is validated by comparing experimental and simulated images of phantoms of different complexity. The differentiation between tissue types is investigated by imaging objects of known compositions (water, lung, and bone equivalent). A lung tumor phantom, simulating materials and geometry encountered during lung SBRT treatments, is used to investigate image noise properties for various quantities of delivered radiation (monitor units(MU)). Patient scatter images are simulated using the validated simulation model. 4DCT patient data is converted to an MCNP input geometry accounting for different tissue composition and densities. Lung tumor phantom images acquired with decreasing imaging time (decreasing MU) are used to model the expected noise amplitude in patient scatter images, producing realistic simulated patient scatter images with varying temporal resolution. Results: Image intensity in simulated and experimental scatter images of tissue equivalent objects (water, lung, bone) match within the uncertainty (∼3%). Lung tumor phantom images agree as well. Specifically, tumor-to-lung contrast matches within the uncertainty. The addition of random noise approximating quantum noise in experimental images to simulated patient images shows that scatter images of lung tumors can provide images in as fast as 0.5 seconds with CNR∼2.7. Conclusions: A scatter imaging simulation model is developed and validated using experimental phantom scatter images. Following validation, lung cancer patient scatter images are simulated. These simulated patient images demonstrate the clinical utility of scatter imaging for real-time tumor tracking during lung SBRT.

60 APPLIED LIFE SCIENCES↗