Search NASA⌕ Search

SEARCH · Search NASA

Results for “correspondence learning”

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

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

At least 37 records · Page 2

International Space Station Passive Thermal Control System Top Ten Lessons-Learned

Final document not an Abstract attached. The International Space Station (ISS) has been on-orbit for nearly 20 years, and there have been numerous technical challenges along the way from design to assembly to on-orbit anomalies and repairs. The Passive Thermal Control System (PTCS) management team has been a key player in successfully dealing with these challenges. The PTCS team performs thermal analysis in support of design and verification, launch and assembly constraints, integration, sustaining engineering, failure response, and model validation. This analysis is a significant body of work and provides a unique opportunity to compile a wealth of real world engineering and analysis knowledge and the corresponding lessons-learned. The PTCS lessons encompass the full life cycle of flight hardware from design to on-orbit performance and sustaining engineering. These lessons can provide significant insight for new projects and programs. Key areas to be presented include thermal model fidelity, verification methods, analysis uncertainty, and operations support.

Iovine, John V.↗

Optical Control of Adaptive Nanoscale Domain Networks

Adaptive networks can sense and adjust to dynamic environments to optimize their performance. Understanding their nanoscale responses to external stimuli is essential for applications in nanodevices and neuromorphic computing. However, it is challenging to image such responses on the nanoscale with crystallographic sensitivity. Here, the evolution of nanodomain networks in (PbTiO 3 ) n /(SrTiO 3 ) n superlattices (SLs) is directly visualized in real space as the system adapts to ultrafast repetitive optical excitations that emulate controlled neural inputs. The adaptive response allows the system to explore a wealth of metastable states that are previously inaccessible. Their reconfiguration and competition are quantitatively measured by scanning x-ray nanodiffraction as a function of the number of applied pulses, in which crystallographic characteristics are quantitatively assessed by assorted diffraction patterns using unsupervised machine-learning methods. The corresponding domain boundaries and their connectivity are drastically altered by light, holding promise for light-programable nanocircuits in analogy to neuroplasticity. Phase-field simulations elucidate that the reconfiguration of the domain networks is a result of the interplay between photocarriers and transient lattice temperature. The demonstrated optical control scheme and the uncovered nanoscopic insights open opportunities for the remote control of adaptive nanoscale domain networks.

36 MATERIALS SCIENCE↗

Bridging material models across scales: An integrated approach to equation of state and molecular dynamics modeling of copper

New uncertainty-aware equation of state (EOS) and electrical conductivity (EC) models for copper have been developed. The multiphase EOS/EC models are fit to experimental solid/liquid EC isobar measurements as well as density-functional theory molecular dynamics (DFT-MD) EC calculations in both expanded and compressed regimes (0.1–16 g/ cm 3 ⁠). The liquid and solid EOS phases were fit to available experimental data along with additional DFT-MD data over the same range as the EC. Leveraging the DFT-MD data, a corresponding machine-learned interatomic potential (MLIAP) for copper was trained using genetic-algorithm optimization. The copper MLIAP was constrained by EOS shock points at high compressions. The final EOS bounded MLIAP proves to be stable over a large density range (approximately 0.1–20 g/ cm 3 ) with good agreement to an isothermal compression curve, shock Hugoniot, and liquid speed of sound measurements at high pressures (100s of GPa).

Acoustic measurements and instrumentation↗

Curved-line search algorithm for ab initio atomic structure relaxation

Ab initio atomic relaxations often take large numbers of steps and long times to converge, especially when the initial atomic configurations are far from the local minimum or there are curved and narrow valleys in the multidimensional potentials. An atomic relaxation method based on on-the-flight force learning and a corresponding curved-line search algorithm is presented to accelerate this process. In this work, results demonstrate the superior performance of this method for metal and magnetic clusters when compared with the conventional conjugate-gradient method.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Literature mining of protein-residue associations with graph rules learned through distant supervision

Background: We propose a method for automatic extraction of protein-specific residue mentions from the biomedical literature. The method searches text for mentions of amino acids at specific sequence positions and attempts to correctly associate each mention with a protein also named in the text. The methods presented in this work will enable improved protein functional site extraction from articles, ultimately supporting protein function prediction. Our method made use of linguistic patterns for identifying the amino acid residue mentions in text. Further, we applied an automated graph-based method to learn syntactic patterns corresponding to protein-residue pairs mentioned in the text. We finally present an approach to automated construction of relevant training and test data using the distant supervision model. Results: The performance of the method was assessed by extracting protein-residue relations from a new automatically generated test set of sentences containing high confidence examples found using distant supervision. It achieved a F-measure of 0.84 on automatically created silver corpus and 0.79 on a manually annotated gold data set for this task, outperforming previous methods. Conclusions: The primary contributions of this work are to (1) demonstrate the effectiveness of distant supervision for automatic creation of training data for protein-residue relation extraction, substantially reducing the effort and time involved in manual annotation of a data set and (2) show that the graph-based relation extraction approach we used generalizes well to the problem of protein-residue association extraction. This work paves the way towards effective extraction of protein functional residues from the literature.

59 BASIC BIOLOGICAL SCIENCES↗

Probabilistic learning on manifolds constrained by nonlinear partial differential equations for small datasets

A novel extension of the Probabilistic Learning on Manifolds (PLoM) is presented. It makes it possible to synthesize solutions to a wide range of nonlinear stochastic boundary value problems described by partial differential equations (PDEs) for which a stochastic computational model (SCM) is available and which depend on a vector-valued random control parameter. The cost of a single numerical evaluation of this SCM is assumed to be such that only a limited number of points can be computed for constructing the training dataset (small data). Each point of the training dataset is made up of realizations from a vector-valued stochastic process (the stochastic solution) and the associated random control parameter on which it depends. The presented PLoM constrained by PDE allows for generating a large number of learned realizations of the stochastic process and its corresponding random control parameter. These learned realizations are generated so as to minimize the vector-valued random residual of the PDE in the mean-square sense. Appropriate novel methods are developed to solve this challenging problem. Three applications are presented. The first one is a simple uncertain nonlinear dynamical system with a nonstationary stochastic excitation. The second one concerns the 2D nonlinear unsteady Navier–Stokes equations for incompressible flows in which the Reynolds number is the random control parameter. Here, the last one deals with the nonlinear dynamics of a 3D elastic structure with uncertainties. The results obtained make it possible to validate the PLoM constrained by stochastic PDE but also provide further validation of the PLoM without constraint.

Machine learning↗

Cross Inference of Throughput Profiles Using Micro Kernel Network Method

Dedicated network connections are being increasingly deployed in cloud, centralized and edge computing and data infrastructures, whose throughput profiles are critical indicators of the underlying data transfer performance. Due to the cost and disruptions to physical infrastructures, network emulators, such as Mininet, are often used to generate measurements needed to estimate throughput profiles, typically expressed as a function of the connection round trip time. The profiles estimated using measurements from such emulated networks are usually inaccurate for high bandwidth and high latency connections, since they do not accurately reflect the critical network transport dynamics mainly due to computing and memory constraints of the host. We present a machine learning (ML) method to estimate the throughput profiles using emulation measurements to closely match the testbed and production network profiles. In particular, we propose a micro Kernel Network (mKN) that provides baseline throughput measurements on the host running Mininet emulations, which are used to learn a regression map that converts them to the corresponding testbed measurement estimates. Once initially learned, this map is applied to measurements from subsequent network emulations on the same host. We present experimental measurements to illustrate this approach, and derive generalization equations for the proposed mKN-ML method. Using a four-site scenario emulation, we show the effectiveness of this method in providing accurate concave throughput profiles from inaccurate convex or non-smooth ones indicated by Mininet emulation.

Rao, Nageswara↗

Improving ADAM through an implicit-explicit (IMEX) time-stepping approach

The ADAM optimizer, often used in machine learning for neural network training, corresponds to an underlying ordinary differential equation (ODE) in the limit of very small learning rates. Here, this work shows that the classical ADAM algorithm is a first-order implicit-explicit (IMEX) Euler discretization of the underlying ODE. Employing the time discretization point of view, we propose new extensions of the ADAM scheme obtained by using higher-order IMEX methods to solve the ODE. Based on this approach, we derive a new optimization algorithm for neural network training that performs better than classical ADAM on several regression and classification problems.

97 MATHEMATICS AND COMPUTING↗

Heterogeneous Machine Learning on High Performance Computing for End to End Driving of Autonomous Vehicles

Current artificial intelligence techniques for end to end driving of autonomous vehicles typically rely on a single form of learning or training processes along with a corresponding dataset or simulation environment. Relatively speaking, success has been shown for a variety of learning modalities in which it can be shown that the machine can successfully “drive” a vehicle. However, the realm of real-world driving extends significantly beyond the realm of limited test environments for machine training. This creates an enormous gap in capability between these two realms. With their superior neural network structures and learning capabilities, humans can be easily trained within a short period of time to proceed from limited test environments to real world driving. For machines though, this gap is guarded by at least two challenges: 1) machine learning techniques remain brittle and unable to generalize to a wide range of scenarios, and 2) effective training data that enhances generalization and generates the desired driving behavior. Further, each challenge can be computationally intensive on its own thereby exasperating the gap. Moreover, is has not yet been shown that a single form of learning or training is capable of addressing a large range of scenarios. As a result, solving the first challenge does not inherently solve the second and vice versa. The work described here discusses an approach to address the first challenge that would also provide a foundation for solving the second. Our approach utilizes a combination of conditional imitation learning with a static dataset, reinforcement learning with a simulation environment, and high-performance computing to train a neural network. As a result, this reduces the “time to solution” from to the existing techniques for autonomous driving and provides an extensible framework to address the second key challenge.

autonomous vehicles↗

TU-EF-BRA-00: MR Basics I

NMR, and Proton Density MRI of the 1D Patient - Anthony Wolbarst Net Voxel Magnetization, m(x,t). T1-MRI; The MRI Device - Lisa Lemen ‘Classical’ NMR; FID Imaging in 1D via k-Space - Nathan Yanasak Spin-Echo; S-E/Spin Warp in a 2D Slice - Ronald Price Magnetic resonance imaging not only reveals the structural, anatomic details of the body, as does CT, but also it can provide information on the physiological status and pathologies of its tissues, like nuclear medicine. It can display high-quality slice and 3D images of organs and vessels viewed from any perspective, with resolution better than 1 mm. MRI is perhaps most extraordinary and notable for the plethora of ways in which it can create unique forms of image contrast, reflective of fundamentally different biophysical phenomena. As with ultrasound, there is no risk from ionizing radiation to the patient or staff, since no X-rays or radioactive nuclei are involved. Instead, MRI harnesses magnetic fields and radio waves to probe the stable nuclei of the ordinary hydrogen atoms (isolated protons) occurring in water and lipid molecules within and around cells. MRI consists, in essence, of creating spatial maps of the electromagnetic environments around these hydrogen nuclei. Spatial variations in the proton milieus can be related to clinical differences in the biochemical and physiological properties and conditions of the associated tissues. Imaging of proton density (PD), and of the tissue proton spin relaxation times known as T1 and T2, all can reveal important clinical information, but they do so with approaches so dissimilar from one another that each is chosen for only certain clinical situations. T1 and T2 in a voxel are determined by different aspects of the rotations and other motions of the water and lipid molecules involved, as constrained by the local biophysical surroundings within and between its cells – and they, in turn, depend on the type of tissue and its state of health. Three other common applications of MRI exploit its capability to detect and image distinct movements of fluids: MR angiography (MRA), which rivals CT angiography but often requires no contrast medium, monitors the bulk flow of blood; functional MRI (f MRI), distinguishes the perfusion of oxygenated blood from that of de-oxygenated, and lights up parts of the brain that are activated by a stimulus, rather like PET; and diffusion tensor imaging (DTI) indicates the diffusion of free water along tracts of axons, thereby bringing nerve trunks into view. There are variants on all of these themes, and on others as well. Magnetic Resonance Spectroscopy (MRS), for example, can perform non-invasive ‘virtual biopsies’ that allow identification of certain cancers and other lesions. And an MRI-guided needle biopsy can sample brain tissue from a region only millimeters in dimensions. MRI, however, involves deeper and more complex aspects of physics, technology, and biology than do most other imaging modalities, and it is widely considered to be correspondingly more difficult to learn. We could probably cover all of this rather comprehensively if we had 50 hours available rather than 2 ̶ but, to paraphrase a former Secretary of Defense, you tell your story in the time you have allotted. The four presenters and another physicist, Kevin King from GE, have combined their efforts to co-author a single slide show that describes essentials of MRI as simply as possible. It is obviously far from thorough, but hopefully it will succeed in explaining some of the basics in a simplified but still valid fashion; in providing a taste of the numerous capabilities and complexities of the modality; and in whetting your appetite to learn more. Part I. NMR, and Proton Density MRI of the 1D Patient (Wolbarst), begins with an introductory case study that illustrates a half dozen ways in which MRI provides valuable clinical information. It then explores the nuclear magnetic resonance (NMR) phenomenon, which underlies MRI. NMR can be introduced with either of two approaches. In the first, one thinks (loosely) of the nuclei of hydrogen atoms as (rotating and charged and therefore) magnetic objects, whose spin-axes tend to align in a strong external magnetic field, much like a compass needle. As with the Bohr atom, this spin-up/spin-down picture is a highly abridged version of the full quantum mechanical treatment, but still it leads to some useful, legitimate pictures of the NMR process occurring within a voxel: When RF photons of the correct (Larmor) frequency elevate protons in a fixed magnetic field out of their lower-energy spin state into the upper, the NMR phenomenon is indicated by the detectable absorption of RF power. With the addition of a linear gradient field along a multi-voxel, one-dimensional patient/phantom, as well, we can determine the water content of each compartment – an example of a real MRI study, albeit in 1D. Part I concludes with a discussion of the net magnetization at position x, m0(x), under conditions of dynamic thermal equilibrium, which leads into: Part II. Net Voxel Magnetization, m(x,t); T1-MRI; The MRI Device (Lemen), investigates the biophysics of the form of proton spin relaxation process characterized by the time T1. It then moves on to the creation of an MR image that displays the spatial variation in the values of this clinically relevant parameter, again in 1D. Finally, the design and workings of a clinical MRI machine are sketched, in preparation for: Part III. ‘Classical’ NMR; FID Imaging in 1D via k-Space (Yanasak) presents the second standard approach to NMR and MRI, the classical model. It focuses on the time dependence of the net nuclear magnetization, m(x,t), the overall magnetic field generated by the cohort of protons in the voxel at position x. Quite remarkably, this nuclear net magnetization itself acts in a strong magnetic field like a gyroscope in a gravitational field. This tack is better for explaining Free Induction Decay (FID), which involves a brief introduction to the Fourier transform and k-space. This leads to conventional Spin-Echo (S-E) reconstruction techniques for creating clinical images from raw data, and sets the stage for: Part IV. Spin-Echo; S-E / Spin Warp in a 2D Slice (Price) discusses application of the S- E sequence of radiofrequency pulses and gradient magnetic fields to the 1D patient. T2 is introduced but not explained. This Part also considers how to manipulate the image acquisition parameters so as to generate clinical pictures with contrast dominated by spatial variations in PD, T1, or T2. We conclude by demonstrating the spin-warp approach to imaging in 2D with a simple 2×2, 4-voxel example. Much of this material is presented in more detail in the chapter “MRI of the One-dimensional Patient, Part I”, in Advances in Medical Physics, Vol 5 (2014). Copies are on display at the Medical Physics Publishing booth. Learning Objectives: The participant will learn about the processes of NMR and T1 spin relaxation in a tissue voxel in a uniform magnetic field. The participant will learn about combining spin-up/spin-down NMR and T1 processes with a linear gradient to effect frequency-encoding of voxel spatial position. This approach can be used to create proton density and T1 MRI maps, respectively, of the contents of multi-voxel 1D phantoms. The participant will learn about how the ‘classical’ model of NMR it can generate Free Induction Decay (FID) images of 1D phantoms, which involves the use of the Fourier transform in k-space. This can lead simply into standard Spin-Echo images. The participant will learn about extending Spin-Echo imaging into 2 and more dimensions.

60 APPLIED LIFE SCIENCES↗

TU-EF-BRA-01: NMR and Proton Density MRI of the 1D Patient

NMR, and Proton Density MRI of the 1D Patient - Anthony Wolbarst Net Voxel Magnetization, m(x,t). T1-MRI; The MRI Device - Lisa Lemen ‘Classical’ NMR; FID Imaging in 1D via k-Space - Nathan Yanasak Spin-Echo; S-E/Spin Warp in a 2D Slice - Ronald Price Magnetic resonance imaging not only reveals the structural, anatomic details of the body, as does CT, but also it can provide information on the physiological status and pathologies of its tissues, like nuclear medicine. It can display high-quality slice and 3D images of organs and vessels viewed from any perspective, with resolution better than 1 mm. MRI is perhaps most extraordinary and notable for the plethora of ways in which it can create unique forms of image contrast, reflective of fundamentally different biophysical phenomena. As with ultrasound, there is no risk from ionizing radiation to the patient or staff, since no X-rays or radioactive nuclei are involved. Instead, MRI harnesses magnetic fields and radio waves to probe the stable nuclei of the ordinary hydrogen atoms (isolated protons) occurring in water and lipid molecules within and around cells. MRI consists, in essence, of creating spatial maps of the electromagnetic environments around these hydrogen nuclei. Spatial variations in the proton milieus can be related to clinical differences in the biochemical and physiological properties and conditions of the associated tissues. Imaging of proton density (PD), and of the tissue proton spin relaxation times known as T1 and T2, all can reveal important clinical information, but they do so with approaches so dissimilar from one another that each is chosen for only certain clinical situations. T1 and T2 in a voxel are determined by different aspects of the rotations and other motions of the water and lipid molecules involved, as constrained by the local biophysical surroundings within and between its cells – and they, in turn, depend on the type of tissue and its state of health. Three other common applications of MRI exploit its capability to detect and image distinct movements of fluids: MR angiography (MRA), which rivals CT angiography but often requires no contrast medium, monitors the bulk flow of blood; functional MRI ( f MRI), distinguishes the perfusion of oxygenated blood from that of de-oxygenated, and lights up parts of the brain that are activated by a stimulus, rather like PET; and diffusion tensor imaging (DTI) indicates the diffusion of free water along tracts of axons, thereby bringing nerve trunks into view. There are variants on all of these themes, and on others as well. Magnetic Resonance Spectroscopy (MRS), for example, can perform non-invasive ‘virtual biopsies’ that allow identification of certain cancers and other lesions. And an MRI-guided needle biopsy can sample brain tissue from a region only millimeters in dimensions. MRI, however, involves deeper and more complex aspects of physics, technology, and biology than do most other imaging modalities, and it is widely considered to be correspondingly more difficult to learn. We could probably cover all of this rather comprehensively if we had 50 hours available rather than 2 ̶ but, to paraphrase a former Secretary of Defense, you tell your story in the time you have allotted. The four presenters and another physicist, Kevin King from GE, have combined their efforts to co-author a single slide show that describes essentials of MRI as simply as possible. It is obviously far from thorough, but hopefully it will succeed in explaining some of the basics in a simplified but still valid fashion; in providing a taste of the numerous capabilities and complexities of the modality; and in whetting your appetite to learn more. Part I. NMR, and Proton Density MRI of the 1D Patient (Wolbarst), begins with an introductory case study that illustrates a half dozen ways in which MRI provides valuable clinical information. It then explores the nuclear magnetic resonance (NMR) phenomenon, which underlies MRI. NMR can be introduced with either of two approaches. In the first, one thinks (loosely) of the nuclei of hydrogen atoms as (rotating and charged and therefore) magnetic objects, whose spin-axes tend to align in a strong external magnetic field, much like a compass needle. As with the Bohr atom, this spin-up/spin-down picture is a highly abridged version of the full quantum mechanical treatment, but still it leads to some useful, legitimate pictures of the NMR process occurring within a voxel: When RF photons of the correct (Larmor) frequency elevate protons in a fixed magnetic field out of their lower-energy spin state into the upper, the NMR phenomenon is indicated by the detectable absorption of RF power. With the addition of a linear gradient field along a multi-voxel, one-dimensional patient/phantom, as well, we can determine the water content of each compartment – an example of a real MRI study, albeit in 1D. Part I concludes with a discussion of the net magnetization at position x, m0(x), under conditions of dynamic thermal equilibrium, which leads into: Part II. Net Voxel Magnetization, m(x,t); T1-MRI; The MRI Device (Lemen), investigates the biophysics of the form of proton spin relaxation process characterized by the time T1. It then moves on to the creation of an MR image that displays the spatial variation in the values of this clinically relevant parameter, again in 1D. Finally, the design and workings of a clinical MRI machine are sketched, in preparation for: Part III. ‘Classical’ NMR; FID Imaging in 1D via k-Space (Yanasak) presents the second standard approach to NMR and MRI, the classical model. It focuses on the time dependence of the net nuclear magnetization, m(x,t), the overall magnetic field generated by the cohort of protons in the voxel at position x. Quite remarkably, this nuclear net magnetization itself acts in a strong magnetic field like a gyroscope in a gravitational field. This tack is better for explaining Free Induction Decay (FID), which involves a brief introduction to the Fourier transform and k-space. This leads to conventional Spin-Echo (S-E) reconstruction techniques for creating clinical images from raw data, and sets the stage for: Part IV. Spin-Echo; S-E / Spin Warp in a 2D Slice (Price) discusses application of the S- E sequence of radiofrequency pulses and gradient magnetic fields to the 1D patient. T2 is introduced but not explained. This Part also considers how to manipulate the image acquisition parameters so as to generate clinical pictures with contrast dominated by spatial variations in PD, T1, or T2. We conclude by demonstrating the spin-warp approach to imaging in 2D with a simple 2×2, 4-voxel example. Much of this material is presented in more detail in the chapter “MRI of the One-dimensional Patient, Part I”, in Advances in Medical Physics, Vol 5 (2014). Copies are on display at the Medical Physics Publishing booth. Learning Objectives: The participant will learn about the processes of NMR and T1 spin relaxation in a tissue voxel in a uniform magnetic field. The participant will learn about combining spin-up/spin-down NMR and T1 processes with a linear gradient to effect frequency-encoding of voxel spatial position. This approach can be used to create proton density and T1 MRI maps, respectively, of the contents of multi-voxel 1D phantoms. The participant will learn about how the ‘classical’ model of NMR it can generate Free Induction Decay (FID) images of 1D phantoms, which involves the use of the Fourier transform in k-space. This can lead simply into standard Spin-Echo images. The participant will learn about extending Spin-Echo imaging into 2 and more dimensions.

60 APPLIED LIFE SCIENCES↗

TU-EF-BRA-02: Longitudinal Proton Spin Relaxation and T1-Imaging

NMR, and Proton Density MRI of the 1D Patient - Anthony Wolbarst Net Voxel Magnetization, m(x,t). T1-MRI; The MRI Device - Lisa Lemen ‘Classical’ NMR; FID Imaging in 1D via k-Space - Nathan Yanasak Spin-Echo; S-E/Spin Warp in a 2D Slice - Ronald Price Magnetic resonance imaging not only reveals the structural, anatomic details of the body, as does CT, but also it can provide information on the physiological status and pathologies of its tissues, like nuclear medicine. It can display high-quality slice and 3D images of organs and vessels viewed from any perspective, with resolution better than 1 mm. MRI is perhaps most extraordinary and notable for the plethora of ways in which it can create unique forms of image contrast, reflective of fundamentally different biophysical phenomena. As with ultrasound, there is no risk from ionizing radiation to the patient or staff, since no X-rays or radioactive nuclei are involved. Instead, MRI harnesses magnetic fields and radio waves to probe the stable nuclei of the ordinary hydrogen atoms (isolated protons) occurring in water and lipid molecules within and around cells. MRI consists, in essence, of creating spatial maps of the electromagnetic environments around these hydrogen nuclei. Spatial variations in the proton milieus can be related to clinical differences in the biochemical and physiological properties and conditions of the associated tissues. Imaging of proton density (PD), and of the tissue proton spin relaxation times known as T1 and T2, all can reveal important clinical information, but they do so with approaches so dissimilar from one another that each is chosen for only certain clinical situations. T1 and T2 in a voxel are determined by different aspects of the rotations and other motions of the water and lipid molecules involved, as constrained by the local biophysical surroundings within and between its cells – and they, in turn, depend on the type of tissue and its state of health. Three other common applications of MRI exploit its capability to detect and image distinct movements of fluids: MR angiography (MRA), which rivals CT angiography but often requires no contrast medium, monitors the bulk flow of blood; functional MRI ( f MRI), distinguishes the perfusion of oxygenated blood from that of de-oxygenated, and lights up parts of the brain that are activated by a stimulus, rather like PET; and diffusion tensor imaging (DTI) indicates the diffusion of free water along tracts of axons, thereby bringing nerve trunks into view. There are variants on all of these themes, and on others as well. Magnetic Resonance Spectroscopy (MRS), for example, can perform non-invasive ‘virtual biopsies’ that allow identification of certain cancers and other lesions. And an MRI-guided needle biopsy can sample brain tissue from a region only millimeters in dimensions. MRI, however, involves deeper and more complex aspects of physics, technology, and biology than do most other imaging modalities, and it is widely considered to be correspondingly more difficult to learn. We could probably cover all of this rather comprehensively if we had 50 hours available rather than 2 ̶ but, to paraphrase a former Secretary of Defense, you tell your story in the time you have allotted. The four presenters and another physicist, Kevin King from GE, have combined their efforts to co-author a single slide show that describes essentials of MRI as simply as possible. It is obviously far from thorough, but hopefully it will succeed in explaining some of the basics in a simplified but still valid fashion; in providing a taste of the numerous capabilities and complexities of the modality; and in whetting your appetite to learn more. Part I. NMR, and Proton Density MRI of the 1D Patient (Wolbarst), begins with an introductory case study that illustrates a half dozen ways in which MRI provides valuable clinical information. It then explores the nuclear magnetic resonance (NMR) phenomenon, which underlies MRI. NMR can be introduced with either of two approaches. In the first, one thinks (loosely) of the nuclei of hydrogen atoms as (rotating and charged and therefore) magnetic objects, whose spin-axes tend to align in a strong external magnetic field, much like a compass needle. As with the Bohr atom, this spin-up/spin-down picture is a highly abridged version of the full quantum mechanical treatment, but still it leads to some useful, legitimate pictures of the NMR process occurring within a voxel: When RF photons of the correct (Larmor) frequency elevate protons in a fixed magnetic field out of their lower-energy spin state into the upper, the NMR phenomenon is indicated by the detectable absorption of RF power. With the addition of a linear gradient field along a multi-voxel, one-dimensional patient/phantom, as well, we can determine the water content of each compartment – an example of a real MRI study, albeit in 1D. Part I concludes with a discussion of the net magnetization at position x, m0(x), under conditions of dynamic thermal equilibrium, which leads into: Part II. Net Voxel Magnetization, m(x,t); T1-MRI; The MRI Device (Lemen), investigates the biophysics of the form of proton spin relaxation process characterized by the time T1. It then moves on to the creation of an MR image that displays the spatial variation in the values of this clinically relevant parameter, again in 1D. Finally, the design and workings of a clinical MRI machine are sketched, in preparation for: Part III. ‘Classical’ NMR; FID Imaging in 1D via k-Space (Yanasak) presents the second standard approach to NMR and MRI, the classical model. It focuses on the time dependence of the net nuclear magnetization, m(x,t), the overall magnetic field generated by the cohort of protons in the voxel at position x. Quite remarkably, this nuclear net magnetization itself acts in a strong magnetic field like a gyroscope in a gravitational field. This tack is better for explaining Free Induction Decay (FID), which involves a brief introduction to the Fourier transform and k-space. This leads to conventional Spin-Echo (S-E) reconstruction techniques for creating clinical images from raw data, and sets the stage for: Part IV. Spin-Echo; S-E / Spin Warp in a 2D Slice (Price) discusses application of the S- E sequence of radiofrequency pulses and gradient magnetic fields to the 1D patient. T2 is introduced but not explained. This Part also considers how to manipulate the image acquisition parameters so as to generate clinical pictures with contrast dominated by spatial variations in PD, T1, or T2. We conclude by demonstrating the spin-warp approach to imaging in 2D with a simple 2×2, 4-voxel example. Much of this material is presented in more detail in the chapter “MRI of the One-dimensional Patient, Part I”, in Advances in Medical Physics, Vol 5 (2014). Copies are on display at the Medical Physics Publishing booth. Learning Objectives: The participant will learn about the processes of NMR and T1 spin relaxation in a tissue voxel in a uniform magnetic field. The participant will learn about combining spin-up/spin-down NMR and T1 processes with a linear gradient to effect frequency-encoding of voxel spatial position. This approach can be used to create proton density and T1 MRI maps, respectively, of the contents of multi-voxel 1D phantoms. The participant will learn about how the ‘classical’ model of NMR it can generate Free Induction Decay (FID) images of 1D phantoms, which involves the use of the Fourier transform in k-space. This can lead simply into standard Spin-Echo images. The participant will learn about extending Spin-Echo imaging into 2 and more dimensions.

60 APPLIED LIFE SCIENCES↗

TU-EF-BRA-03: Free Induction Decay (without the Decay) and Spin-Echo Imaging

NMR, and Proton Density MRI of the 1D Patient - Anthony Wolbarst Net Voxel Magnetization, m(x,t). T1-MRI; The MRI Device - Lisa Lemen ‘Classical’ NMR; FID Imaging in 1D via k-Space - Nathan Yanasak Spin-Echo; S-E/Spin Warp in a 2D Slice - Ronald Price Magnetic resonance imaging not only reveals the structural, anatomic details of the body, as does CT, but also it can provide information on the physiological status and pathologies of its tissues, like nuclear medicine. It can display high-quality slice and 3D images of organs and vessels viewed from any perspective, with resolution better than 1 mm. MRI is perhaps most extraordinary and notable for the plethora of ways in which it can create unique forms of image contrast, reflective of fundamentally different biophysical phenomena. As with ultrasound, there is no risk from ionizing radiation to the patient or staff, since no X-rays or radioactive nuclei are involved. Instead, MRI harnesses magnetic fields and radio waves to probe the stable nuclei of the ordinary hydrogen atoms (isolated protons) occurring in water and lipid molecules within and around cells. MRI consists, in essence, of creating spatial maps of the electromagnetic environments around these hydrogen nuclei. Spatial variations in the proton milieus can be related to clinical differences in the biochemical and physiological properties and conditions of the associated tissues. Imaging of proton density (PD), and of the tissue proton spin relaxation times known as T1 and T2, all can reveal important clinical information, but they do so with approaches so dissimilar from one another that each is chosen for only certain clinical situations. T1 and T2 in a voxel are determined by different aspects of the rotations and other motions of the water and lipid molecules involved, as constrained by the local biophysical surroundings within and between its cells – and they, in turn, depend on the type of tissue and its state of health. Three other common applications of MRI exploit its capability to detect and image distinct movements of fluids: MR angiography (MRA), which rivals CT angiography but often requires no contrast medium, monitors the bulk flow of blood; functional MRI ( f MRI), distinguishes the perfusion of oxygenated blood from that of de-oxygenated, and lights up parts of the brain that are activated by a stimulus, rather like PET; and diffusion tensor imaging (DTI) indicates the diffusion of free water along tracts of axons, thereby bringing nerve trunks into view. There are variants on all of these themes, and on others as well. Magnetic Resonance Spectroscopy (MRS), for example, can perform non-invasive ‘virtual biopsies’ that allow identification of certain cancers and other lesions. And an MRI-guided needle biopsy can sample brain tissue from a region only millimeters in dimensions. MRI, however, involves deeper and more complex aspects of physics, technology, and biology than do most other imaging modalities, and it is widely considered to be correspondingly more difficult to learn. We could probably cover all of this rather comprehensively if we had 50 hours available rather than 2 ̶ but, to paraphrase a former Secretary of Defense, you tell your story in the time you have allotted. The four presenters and another physicist, Kevin King from GE, have combined their efforts to co-author a single slide show that describes essentials of MRI as simply as possible. It is obviously far from thorough, but hopefully it will succeed in explaining some of the basics in a simplified but still valid fashion; in providing a taste of the numerous capabilities and complexities of the modality; and in whetting your appetite to learn more. Part I. NMR, and Proton Density MRI of the 1D Patient (Wolbarst), begins with an introductory case study that illustrates a half dozen ways in which MRI provides valuable clinical information. It then explores the nuclear magnetic resonance (NMR) phenomenon, which underlies MRI. NMR can be introduced with either of two approaches. In the first, one thinks (loosely) of the nuclei of hydrogen atoms as (rotating and charged and therefore) magnetic objects, whose spin-axes tend to align in a strong external magnetic field, much like a compass needle. As with the Bohr atom, this spin-up/spin-down picture is a highly abridged version of the full quantum mechanical treatment, but still it leads to some useful, legitimate pictures of the NMR process occurring within a voxel: When RF photons of the correct (Larmor) frequency elevate protons in a fixed magnetic field out of their lower-energy spin state into the upper, the NMR phenomenon is indicated by the detectable absorption of RF power. With the addition of a linear gradient field along a multi-voxel, one-dimensional patient/phantom, as well, we can determine the water content of each compartment – an example of a real MRI study, albeit in 1D. Part I concludes with a discussion of the net magnetization at position x, m0(x), under conditions of dynamic thermal equilibrium, which leads into: Part II. Net Voxel Magnetization, m(x,t); T1-MRI; The MRI Device (Lemen), investigates the biophysics of the form of proton spin relaxation process characterized by the time T1. It then moves on to the creation of an MR image that displays the spatial variation in the values of this clinically relevant parameter, again in 1D. Finally, the design and workings of a clinical MRI machine are sketched, in preparation for: Part III. ‘Classical’ NMR; FID Imaging in 1D via k-Space (Yanasak) presents the second standard approach to NMR and MRI, the classical model. It focuses on the time dependence of the net nuclear magnetization, m(x,t), the overall magnetic field generated by the cohort of protons in the voxel at position x. Quite remarkably, this nuclear net magnetization itself acts in a strong magnetic field like a gyroscope in a gravitational field. This tack is better for explaining Free Induction Decay (FID), which involves a brief introduction to the Fourier transform and k-space. This leads to conventional Spin-Echo (S-E) reconstruction techniques for creating clinical images from raw data, and sets the stage for: Part IV. Spin-Echo; S-E / Spin Warp in a 2D Slice (Price) discusses application of the S- E sequence of radiofrequency pulses and gradient magnetic fields to the 1D patient. T2 is introduced but not explained. This Part also considers how to manipulate the image acquisition parameters so as to generate clinical pictures with contrast dominated by spatial variations in PD, T1, or T2. We conclude by demonstrating the spin-warp approach to imaging in 2D with a simple 2×2, 4-voxel example. Much of this material is presented in more detail in the chapter “MRI of the One-dimensional Patient, Part I”, in Advances in Medical Physics, Vol 5 (2014). Copies are on display at the Medical Physics Publishing booth. Learning Objectives: The participant will learn about the processes of NMR and T1 spin relaxation in a tissue voxel in a uniform magnetic field. The participant will learn about combining spin-up/spin-down NMR and T1 processes with a linear gradient to effect frequency-encoding of voxel spatial position. This approach can be used to create proton density and T1 MRI maps, respectively, of the contents of multi-voxel 1D phantoms. The participant will learn about how the ‘classical’ model of NMR it can generate Free Induction Decay (FID) images of 1D phantoms, which involves the use of the Fourier transform in k-space. This can lead simply into standard Spin-Echo images. The participant will learn about extending Spin-Echo imaging into 2 and more dimensions.

60 APPLIED LIFE SCIENCES↗

TU-EF-BRA-04: Into 2, 3, and 4 Dimensions

NMR, and Proton Density MRI of the 1D Patient - Anthony Wolbarst Net Voxel Magnetization, m(x,t). T1-MRI; The MRI Device - Lisa Lemen ‘Classical’ NMR; FID Imaging in 1D via k-Space - Nathan Yanasak Spin-Echo; S-E/Spin Warp in a 2D Slice - Ronald Price Magnetic resonance imaging not only reveals the structural, anatomic details of the body, as does CT, but also it can provide information on the physiological status and pathologies of its tissues, like nuclear medicine. It can display high-quality slice and 3D images of organs and vessels viewed from any perspective, with resolution better than 1 mm. MRI is perhaps most extraordinary and notable for the plethora of ways in which it can create unique forms of image contrast, reflective of fundamentally different biophysical phenomena. As with ultrasound, there is no risk from ionizing radiation to the patient or staff, since no X-rays or radioactive nuclei are involved. Instead, MRI harnesses magnetic fields and radio waves to probe the stable nuclei of the ordinary hydrogen atoms (isolated protons) occurring in water and lipid molecules within and around cells. MRI consists, in essence, of creating spatial maps of the electromagnetic environments around these hydrogen nuclei. Spatial variations in the proton milieus can be related to clinical differences in the biochemical and physiological properties and conditions of the associated tissues. Imaging of proton density (PD), and of the tissue proton spin relaxation times known as T1 and T2, all can reveal important clinical information, but they do so with approaches so dissimilar from one another that each is chosen for only certain clinical situations. T1 and T2 in a voxel are determined by different aspects of the rotations and other motions of the water and lipid molecules involved, as constrained by the local biophysical surroundings within and between its cells – and they, in turn, depend on the type of tissue and its state of health. Three other common applications of MRI exploit its capability to detect and image distinct movements of fluids: MR angiography (MRA), which rivals CT angiography but often requires no contrast medium, monitors the bulk flow of blood; functional MRI ( f MRI), distinguishes the perfusion of oxygenated blood from that of de-oxygenated, and lights up parts of the brain that are activated by a stimulus, rather like PET; and diffusion tensor imaging (DTI) indicates the diffusion of free water along tracts of axons, thereby bringing nerve trunks into view. There are variants on all of these themes, and on others as well. Magnetic Resonance Spectroscopy (MRS), for example, can perform non-invasive ‘virtual biopsies’ that allow identification of certain cancers and other lesions. And an MRI-guided needle biopsy can sample brain tissue from a region only millimeters in dimensions. MRI, however, involves deeper and more complex aspects of physics, technology, and biology than do most other imaging modalities, and it is widely considered to be correspondingly more difficult to learn. We could probably cover all of this rather comprehensively if we had 50 hours available rather than 2 ̶ but, to paraphrase a former Secretary of Defense, you tell your story in the time you have allotted. The four presenters and another physicist, Kevin King from GE, have combined their efforts to co-author a single slide show that describes essentials of MRI as simply as possible. It is obviously far from thorough, but hopefully it will succeed in explaining some of the basics in a simplified but still valid fashion; in providing a taste of the numerous capabilities and complexities of the modality; and in whetting your appetite to learn more. Part I. NMR, and Proton Density MRI of the 1D Patient (Wolbarst), begins with an introductory case study that illustrates a half dozen ways in which MRI provides valuable clinical information. It then explores the nuclear magnetic resonance (NMR) phenomenon, which underlies MRI. NMR can be introduced with either of two approaches. In the first, one thinks (loosely) of the nuclei of hydrogen atoms as (rotating and charged and therefore) magnetic objects, whose spin-axes tend to align in a strong external magnetic field, much like a compass needle. As with the Bohr atom, this spin-up/spin-down picture is a highly abridged version of the full quantum mechanical treatment, but still it leads to some useful, legitimate pictures of the NMR process occurring within a voxel: When RF photons of the correct (Larmor) frequency elevate protons in a fixed magnetic field out of their lower-energy spin state into the upper, the NMR phenomenon is indicated by the detectable absorption of RF power. With the addition of a linear gradient field along a multi-voxel, one-dimensional patient/phantom, as well, we can determine the water content of each compartment – an example of a real MRI study, albeit in 1D. Part I concludes with a discussion of the net magnetization at position x, m0(x), under conditions of dynamic thermal equilibrium, which leads into: Part II. Net Voxel Magnetization, m(x,t); T1-MRI; The MRI Device (Lemen), investigates the biophysics of the form of proton spin relaxation process characterized by the time T1. It then moves on to the creation of an MR image that displays the spatial variation in the values of this clinically relevant parameter, again in 1D. Finally, the design and workings of a clinical MRI machine are sketched, in preparation for: Part III. ‘Classical’ NMR; FID Imaging in 1D via k-Space (Yanasak) presents the second standard approach to NMR and MRI, the classical model. It focuses on the time dependence of the net nuclear magnetization, m(x,t), the overall magnetic field generated by the cohort of protons in the voxel at position x. Quite remarkably, this nuclear net magnetization itself acts in a strong magnetic field like a gyroscope in a gravitational field. This tack is better for explaining Free Induction Decay (FID), which involves a brief introduction to the Fourier transform and k-space. This leads to conventional Spin-Echo (S-E) reconstruction techniques for creating clinical images from raw data, and sets the stage for: Part IV. Spin-Echo; S-E / Spin Warp in a 2D Slice (Price) discusses application of the S- E sequence of radiofrequency pulses and gradient magnetic fields to the 1D patient. T2 is introduced but not explained. This Part also considers how to manipulate the image acquisition parameters so as to generate clinical pictures with contrast dominated by spatial variations in PD, T1, or T2. We conclude by demonstrating the spin-warp approach to imaging in 2D with a simple 2×2, 4-voxel example. Much of this material is presented in more detail in the chapter “MRI of the One-dimensional Patient, Part I”, in Advances in Medical Physics, Vol 5 (2014). Copies are on display at the Medical Physics Publishing booth. Learning Objectives: The participant will learn about the processes of NMR and T1 spin relaxation in a tissue voxel in a uniform magnetic field. The participant will learn about combining spin-up/spin-down NMR and T1 processes with a linear gradient to effect frequency-encoding of voxel spatial position. This approach can be used to create proton density and T1 MRI maps, respectively, of the contents of multi-voxel 1D phantoms. The participant will learn about how the ‘classical’ model of NMR it can generate Free Induction Decay (FID) images of 1D phantoms, which involves the use of the Fourier transform in k-space. This can lead simply into standard Spin-Echo images. The participant will learn about extending Spin-Echo imaging into 2 and more dimensions.

60 APPLIED LIFE SCIENCES↗

Roundtable on Producing and Managing Large Scientific Data with Artificial Intelligence and Machine Learning (Brochure)

The US Department of Energy’s (DOE’s) twelve Basic Energy Sciences (BES) scientific user facilities provide access to the world’s most advanced research instruments, annually serving over 16,000 users with impact reported in nearly 7,000 publications and producing unprecedented quantities of scientific data. While impressive, reaching the full potential of these rapidly growing facilities will require new innovations to solve a variety of technical challenges in data acquisition, control, modeling, and analysis. Artificial intelligence and machine learning (AI/ML) have opened corresponding new avenues in optimization, efficient surrogate models, data analytics, and inverse problems. These intriguing capabilities suggest that AI/ML can greatly accelerate the quest to probe and understand fundamental phenomena across a vast range of lengths and timescales, potentially leading to transformative advances across scientific disciplines.

97 MATHEMATICS AND COMPUTING↗

DPM: A deep learning PDE augmentation method with application to large-eddy simulation

A framework is introduced that leverages known physics to reduce overfitting in machine learning for scientific applications. The partial differential equation (PDE) that expresses the physics is augmented with a neural network that uses available data to learn a description of the corresponding unknown or unrepresented physics. Training within this combined system corrects for missing, unknown, or erroneously represented physics, including discretization errors associated with the PDE's numerical solution. For optimization of the network within the PDE, an adjoint PDE is solved to provide high-dimensional gradients, and a stochastic adjoint method (SAM) further accelerates training. Additionally, the approach is demonstrated for large-eddy simulation (LES) of turbulence. High-fidelity direct numerical simulations (DNS) of decaying isotropic turbulence provide the training data used to learn sub-filter-scale closures for the filtered Navier–Stokes equations. Out-of-sample comparisons show that the deep learning PDE method outperforms widely-used models, even for filter sizes so large that they become qualitatively incorrect. It also significantly outperforms the same neural network when a priori trained based on simple data mismatch, not accounting for the full PDE. Measures of discretization errors, which are well-known to be consequential in LES, point to the importance of the unified training formulation's design, which without modification corrects for them. For comparable accuracy, simulation runtime is significantly reduced. A relaxation of the typical discrete enforcement of the divergence-free constraint in the solver is also successful, instead allowing the DPM to approximately enforce incompressibility physics. Since the training loss function is not restricted to correspond directly to the closure to be learned, training can incorporate diverse data, including experimental data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Deep structural clustering for single-cell RNA-seq data jointly through autoencoder and graph neural network

Abstract Single-cell RNA sequencing (scRNA-seq) permits researchers to study the complex mechanisms of cell heterogeneity and diversity. Unsupervised clustering is of central importance for the analysis of the scRNA-seq data, as it can be used to identify putative cell types. However, due to noise impacts, high dimensionality and pervasive dropout events, clustering analysis of scRNA-seq data remains a computational challenge. Here, we propose a new deep structural clustering method for scRNA-seq data, named scDSC, which integrate the structural information into deep clustering of single cells. The proposed scDSC consists of a Zero-Inflated Negative Binomial (ZINB) model-based autoencoder, a graph neural network (GNN) module and a mutual-supervised module. To learn the data representation from the sparse and zero-inflated scRNA-seq data, we add a ZINB model to the basic autoencoder. The GNN module is introduced to capture the structural information among cells. By joining the ZINB-based autoencoder with the GNN module, the model transfers the data representation learned by autoencoder to the corresponding GNN layer. Furthermore, we adopt a mutual supervised strategy to unify these two different deep neural architectures and to guide the clustering task. Extensive experimental results on six real scRNA-seq datasets demonstrate that scDSC outperforms state-of-the-art methods in terms of clustering accuracy and scalability. Our method scDSC is implemented in Python using the Pytorch machine-learning library, and it is freely available at https://github.com/DHUDBlab/scDSC.

Gan, Yanglan↗

NaroNet: Discovery of tumor microenvironment elements from highly multiplexed images

Understanding the spatial interactions between the elements of the tumor microenvironment -i.e. tumor cells. fibroblasts, immune cells- and how these interactions relate to the diagnosis or prognosis of a tumor is one of the goals of computational pathology. We present NaroNet, a deep learning framework that models the multi-scale tumor microenvironment from multiplex-stained cancer tissue images and provides patient-level interpretable predictions using a seamless end-to-end learning pipeline. Trained only with multiplex-stained tissue images and their corresponding patient-level clinical labels, NaroNet unsupervisedly learns which cell phenotypes, cell neighborhoods, and neighborhood interactions have the highest influence to predict the correct label. To this end, NaroNet incorporates several novel and state-of-the-art deep learning techniques, such as patch-level contrastive learning, multi-level graph embeddings, a novel max-sum pooling operation, or a metric that quantifies the relevance that each microenvironment element has in the individual predictions. We validate NaroNet using synthetic data simulating multiplex-immunostained images where a patient label is artificially associated to the -adjustable- probabilistic incidence of different microenvironment elements. We then apply our model to two sets of images of human cancer tissues: 336 seven-color multiplex-immunostained images from 12 high-grade endometrial cancer patients; and 382 35-plex mass cytometry images from 215 breast cancer patients. In both synthetic and real datasets, NaroNet provides outstanding predictions of relevant clinical information while associating those predictions to the presence of specific microenvironment elements.

60 APPLIED LIFE SCIENCES↗