-
Multistep severe plastic deformation to achieve non-rare earth bulk magnets with high α-MnBi phase content
Authors:
L. Weissitsch,
S. Wurster,
H. Krenn,
A. Bachmaier
Abstract:
The ferromagnetic α-MnBi phase as non-rare earth magnetic material has gained increasing interest, but fabrication of large volumes containing a significant amount of α-MnBi is still challenging. Targeting successful processing strategies, we apply multistep severe plastic deformation with intermediate magnetic field assisted annealing. The subsequent severe plastic deformation renewed the high am…
▽ More
The ferromagnetic α-MnBi phase as non-rare earth magnetic material has gained increasing interest, but fabrication of large volumes containing a significant amount of α-MnBi is still challenging. Targeting successful processing strategies, we apply multistep severe plastic deformation with intermediate magnetic field assisted annealing. The subsequent severe plastic deformation renewed the high amount of material defects and lead to microstructural refinement of the previously formed α-MnBi phase. Thus, α-MnBi phase content is enhanced during final annealing. Secondly, the magnetic coercivity increases. These results suggest that further optimization will pave the way towards non-rare-earth bulk magnetic materials with enhanced α-MnBi phase content.
△ Less
Submitted 19 February, 2024;
originally announced February 2024.
-
Synthesis of hard magnetic α-MnBi phase by high pressure torsion and field assisted annealing
Authors:
Lukas Weissitsch,
Stefan Wurster,
Martin Stueckler,
Timo Mueller,
Heinz Krenn,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
Bulk composites using powder blends of Mn and Bi with equal atomic ratios are consolidated and severely deformed by high-pressure torsion (HPT). Subsequent annealing treatments lead to the formation of the ferromagnetic and rare-earth free hard magnetic α-MnBi phase. The initial phase formation is studied by in-situ high-energy XRD annealing experiments showing a beneficial influence of HPT-deform…
▽ More
Bulk composites using powder blends of Mn and Bi with equal atomic ratios are consolidated and severely deformed by high-pressure torsion (HPT). Subsequent annealing treatments lead to the formation of the ferromagnetic and rare-earth free hard magnetic α-MnBi phase. The initial phase formation is studied by in-situ high-energy XRD annealing experiments showing a beneficial influence of HPT-deformation on the amount of α-MnBi. The nucleation sites are strongly increased, thus the volumetric α-MnBi phase content exceeds 50 vol.%. A lower HPT-pressure of 2 GPa is found to be preferred over 5 GPa to obtain high α-MnBi contents. Annealing is done with or without applying an external magnetic field. Additionally, influences of the HPT-induced shear deformation are discussed and correlated with preferred diffusion path ways fostering the formation of anisotropic α-MnBi phase.
△ Less
Submitted 22 December, 2023;
originally announced December 2023.
-
Adaptively Placed Multi-Grid Scene Representation Networks for Large-Scale Data Visualization
Authors:
Skylar Wolfgang Wurster,
Tianyu Xiong,
Han-Wei Shen,
Hanqi Guo,
Tom Peterka
Abstract:
Scene representation networks (SRNs) have been recently proposed for compression and visualization of scientific data. However, state-of-the-art SRNs do not adapt the allocation of available network parameters to the complex features found in scientific data, leading to a loss in reconstruction quality. We address this shortcoming with an adaptively placed multi-grid SRN (APMGSRN) and propose a do…
▽ More
Scene representation networks (SRNs) have been recently proposed for compression and visualization of scientific data. However, state-of-the-art SRNs do not adapt the allocation of available network parameters to the complex features found in scientific data, leading to a loss in reconstruction quality. We address this shortcoming with an adaptively placed multi-grid SRN (APMGSRN) and propose a domain decomposition training and inference technique for accelerated parallel training on multi-GPU systems. We also release an open-source neural volume rendering application that allows plug-and-play rendering with any PyTorch-based SRN. Our proposed APMGSRN architecture uses multiple spatially adaptive feature grids that learn where to be placed within the domain to dynamically allocate more neural network resources where error is high in the volume, improving state-of-the-art reconstruction accuracy of SRNs for scientific data without requiring expensive octree refining, pruning, and traversal like previous adaptive models. In our domain decomposition approach for representing large-scale data, we train an set of APMGSRNs in parallel on separate bricks of the volume to reduce training time while avoiding overhead necessary for an out-of-core solution for volumes too large to fit in GPU memory. After training, the lightweight SRNs are used for realtime neural volume rendering in our open-source renderer, where arbitrary view angles and transfer functions can be explored. A copy of this paper, all code, all models used in our experiments, and all supplemental materials and videos are available at https://github.com/skywolf829/APMGSRN.
△ Less
Submitted 6 April, 2024; v1 submitted 16 July, 2023;
originally announced August 2023.
-
Neural Stream Functions
Authors:
Skylar Wolfgang Wurster,
Hanqi Guo,
Tom Peterka,
Han-Wei Shen
Abstract:
We present a neural network approach to compute stream functions, which are scalar functions with gradients orthogonal to a given vector field. As a result, isosurfaces of the stream function extract stream surfaces, which can be visualized to analyze flow features. Our approach takes a vector field as input and trains an implicit neural representation to learn a stream function for that vector fi…
▽ More
We present a neural network approach to compute stream functions, which are scalar functions with gradients orthogonal to a given vector field. As a result, isosurfaces of the stream function extract stream surfaces, which can be visualized to analyze flow features. Our approach takes a vector field as input and trains an implicit neural representation to learn a stream function for that vector field. The network learns to map input coordinates to a stream function value by minimizing the inner product of the gradient of the neural network's output and the vector field. Since stream function solutions may not be unique, we give optional constraints for the network to learn particular stream functions of interest. Specifically, we introduce regularizing loss functions that can optionally be used to generate stream function solutions whose stream surfaces follow the flow field's curvature, or that can learn a stream function that includes a stream surface passing through a seeding rake. We also discuss considerations for properly visualizing the trained implicit network and extracting artifact-free surfaces. We compare our results with other implicit solutions and present qualitative and quantitative results for several synthetic and simulated vector fields.
△ Less
Submitted 16 July, 2023;
originally announced July 2023.
-
Germany's nationwide travel experiment in 2022: public transport for 9 Euro per month -- First findings of an empirical study
Authors:
Allister Loder,
Fabienne Cantner,
Lennart Adenaw,
Nico Nachtigall,
David Ziegler,
Felix Gotzler,
Markus B. Siewert,
Stefan Wurster,
Sebastian Goerg,
Markus Lienkamp,
Klaus Bogenberger
Abstract:
In spring 2022, the German federal government agreed on a set of policy measures that aimed at reducing households' financial burden resulting from a recent price increase, especially in energy and mobility. These included among others, a nationwide public transport ticket for 9~Euro per month for three months in June, July, and August 2022. In transport policy research this is an almost unprecede…
▽ More
In spring 2022, the German federal government agreed on a set of policy measures that aimed at reducing households' financial burden resulting from a recent price increase, especially in energy and mobility. These included among others, a nationwide public transport ticket for 9~Euro per month for three months in June, July, and August 2022. In transport policy research this is an almost unprecedented behavioral experiment. It allows us to study not only behavioral responses in mode choice and induced demand but also to assess the effectiveness of these instruments. We observe this natural experiment with a three-wave survey and a smartphone-based travel diary with passive tracking on an initial sample of 2,261 participants with a focus on the Munich metropolitan region. This area is chosen as it offers a variety of mode options with a dense and far-reaching public transport network that even provides good access to many leisure destinations. The app has been providing data from 756 participants until the end of September, the three-wave survey by 1,402, and the app and the three waves by 637 participants. In this paper, we report on the study design, the recruitment and study participation as well as the impacts of the policy measures on the self-reported and app-observed travel behavior; we present results on consumer choices for a successor ticket to the 9-Euro-Ticket that started in May 2023. We find a substantial shift in the modal share towards public transport from the car in our sample during the 9-Euro-Ticket period in travel distance (around 5 %) and in trip frequency (around 7 %). The mobility outcomes of the 9-Euro-Ticket however provide evidence that cheap public transport as a policy instrument does not suffice to incentive sustainable travel behavior choices and that other policy instruments are required in addition.
△ Less
Submitted 14 June, 2023;
originally announced June 2023.
-
Exchange Bias Demonstrated in Bulk Nanocomposites Processed by High Pressure Torsion
Authors:
Michael Zawodzki,
Lukas Weissitsch,
Heinz Krenn,
Stefan Wurster,
Andrea Bachmaier
Abstract:
Ferromagnetic (Fe or Fe20Ni80) and antiferromagnetic (NiO) phases were deformed by high pressure torsion, a severe plastic deformation technique, to manufacture bulk sized nanocomposites and demonstrate an exchange bias, which has been reported predominantly for bilayer thin films. High pressure torsion deformation at elevated temperatures proved to be the key to obtain homogeneous bulk nanocompos…
▽ More
Ferromagnetic (Fe or Fe20Ni80) and antiferromagnetic (NiO) phases were deformed by high pressure torsion, a severe plastic deformation technique, to manufacture bulk sized nanocomposites and demonstrate an exchange bias, which has been reported predominantly for bilayer thin films. High pressure torsion deformation at elevated temperatures proved to be the key to obtain homogeneous bulk nanocomposites. X-ray diffraction investigations detected nanocrystallinity of the ferromagnetic and antiferromagnetic phases. Furthermore, an additional phase was identified by X-ray diffraction, which formed during deformation at elevated temperatures through the reduction of NiO by Fe. Depending on the initial powder composition of Fe50NiO50 or Fe10Ni40NiO50 the new phase was magnetite or maghemite, respectively. Magnetometry measurements demonstrated an exchange bias in the high pressure torsion processed bulk nanocomposites. Additionally, a tailoring of magnetic parameters was demonstrated by the application of different strains or post-process annealing. A correlation between the amount of applied strain and exchange bias was found. The increase of exchange bias through applied strain was related to the microstructural refinement of the nanocomposite. The nanocrystalline maghemite was considered to have a crucial impact on the observed changes of exchange bias through applied strain.
△ Less
Submitted 12 January, 2023;
originally announced January 2023.
-
A nation-wide experiment: fuel tax cuts and almost free public transport for three months in Germany -- Report 4 Third wave results
Authors:
Allister Loder,
Fabienne Cantner,
Andrea Cadavid,
Markus B. Siewert,
Stefan Wurster,
Sebastian Goerg,
Klaus Bogenberger
Abstract:
In spring 2022, the German federal government agreed on a set of measures that aimed at reducing households' financial burden resulting from a recent price increase, especially in energy and mobility. These measures included among others, a nation-wide public transport ticket for 9EUR per month and a fuel tax cut that reduced fuel prices by more than 15%. In transportation policy and travel behavi…
▽ More
In spring 2022, the German federal government agreed on a set of measures that aimed at reducing households' financial burden resulting from a recent price increase, especially in energy and mobility. These measures included among others, a nation-wide public transport ticket for 9EUR per month and a fuel tax cut that reduced fuel prices by more than 15%. In transportation policy and travel behavior research this is an almost unprecedented behavioral experiment. It allows to study not only behavioral responses in mode choice and induced demand but also to assess the effectiveness of transport policy instruments. We observe this natural experiment with a three-wave survey and an app-based travel diary on a sample 2'263 individuals; for the Munich Study, 919 participants in the survey-and-app group and 425 in the survey-only group have been successfully recruited, while 919 participants have been recruited through a professional panel provider to obtain a representative nation-wide reference group for the three-wave survey. In this fourth report we present the results of the third wave. At the end of the study, all three surveys have been completed by 1'484 participants and 642 participants completed all three surveys and used the travel diary throughout the entire study. Based on our results we conclude that when offering a 49EUR-Ticket as a successor to the 9~EUR-Ticket and a local travel pass for 30EUR/month more than 60% of all 9~EUR-Ticket owners would buy one of the two new travel passes. In other words, a substantial increase in travel pass ownership in Germany can be expected, with our modest estimate being around 20%. With the announcement of the introduction of a successor ticket in 2023 as well as with the prevailing high inflation, this study will continue into the year 2023 to monitor the impact on mobility and daily activities.
△ Less
Submitted 19 October, 2022;
originally announced October 2022.
-
A nation-wide experiment: fuel tax cuts and almost free public transport for three months in Germany -- Report 3 Second wave results
Authors:
Allister Loder,
Fabienne Cantner,
Andrea Cadavid,
Markus B. Siewert,
Stefan Wurster,
Sebastian Goerg,
Klaus Bogenberger
Abstract:
In spring 2022, the German federal government agreed on a set of measures that aimed at reducing households' financial burden resulting from a recent price increase, especially in energy and mobility. These measures included among others, a nation-wide public transport ticket for 9\ EUR per month and a fuel tax cut that reduced fuel prices by more than 15\,\%. In transportation research this is an…
▽ More
In spring 2022, the German federal government agreed on a set of measures that aimed at reducing households' financial burden resulting from a recent price increase, especially in energy and mobility. These measures included among others, a nation-wide public transport ticket for 9\ EUR per month and a fuel tax cut that reduced fuel prices by more than 15\,\%. In transportation research this is an almost unprecedented behavioral experiment. It allows to study not only behavioral responses in mode choice and induced demand but also to assess the effectiveness of transport policy instruments. We observe this natural experiment with a three-wave survey and an app-based travel diary on a sample of hundreds of participants as well as an analysis of traffic counts. In this third report, we provide first findings from the second survey, conducted during the experiment.
△ Less
Submitted 5 September, 2022; v1 submitted 31 August, 2022;
originally announced August 2022.
-
Magnetoresistive behaviour of ternary Cu-based materials processed by high-pressure torsion
Authors:
M. Kasalo,
S. Wurster,
M. Stückler,
M. Zawodzki,
L. Weissitsch,
R. Pippan,
A. Bachmaier
Abstract:
Severe plastic deformation using high-pressure torsion of ternary Cu-based materials (CuFeCo and CuFeNi) was used to fabricate bulk samples with a nanocrystalline microstructure. The goal was to produce materials featuring the granular giant magnetoresistance effect, requiring interfaces between ferro- and nonmagnetic materials. This magnetic effect was found for both ternary systems; adequate sub…
▽ More
Severe plastic deformation using high-pressure torsion of ternary Cu-based materials (CuFeCo and CuFeNi) was used to fabricate bulk samples with a nanocrystalline microstructure. The goal was to produce materials featuring the granular giant magnetoresistance effect, requiring interfaces between ferro- and nonmagnetic materials. This magnetic effect was found for both ternary systems; adequate subsequent annealing had a positive influence. The as-deformed states, as well as microstructural changes upon thermal treatments, were studied using scanning electron microscopy and X-ray diffraction measurements. Deducing from electron microscopy, a single-phase structure was observed for all as-deformed samples, indicating the formation of a supersaturated solid solution. However, judging from the presence of the granular giant-magnetoresistive effect, small ferromagnetic particles have to be present. The highest drop in room temperature resistivity (2.45% at 1790 kA/m) was found in Cu62Fe19Ni19 after annealing for 1 h at 400 °C. Combining the results of classical microstructural studies and magnetic measurements, insights into the evolution of ferromagnetic particles are accessible.
△ Less
Submitted 25 August, 2022;
originally announced August 2022.
-
Magnetron Sputtered Non-Toxic and Precious Element-Free Ti-Zr-Ge Metallic Glass Nanofilms with Enhanced Biocorrosion Resistance
Authors:
Baran Sarac,
Matej Micusik,
Barbara Putz,
Stefan Wurster,
Elham Sharifikolouei,
Lixia Xi,
Maria Omastova,
Florian Spieckermann,
Christian Mitterer,
Jürgen Eckert
Abstract:
The chemical composition and structural state of advanced alloys are the decisive factors in optimum biomedical performance. This contribution presents unique Ti-Zr-Ge metallic glass thin-film compositions fabricated by magnetron sputter deposition targeted for nanocoatings for biofouling prevention. The amorphous nanofilms with nanoscale roughness exhibit a large relaxation and supercooled liquid…
▽ More
The chemical composition and structural state of advanced alloys are the decisive factors in optimum biomedical performance. This contribution presents unique Ti-Zr-Ge metallic glass thin-film compositions fabricated by magnetron sputter deposition targeted for nanocoatings for biofouling prevention. The amorphous nanofilms with nanoscale roughness exhibit a large relaxation and supercooled liquid regions as revealed by flash differential scanning calorimetry. Ti\textsubscript{68}Zr\textsubscript{8}Ge\textsubscript{24} shows the lowest corrosion (0.17 \textmu A cm\textsuperscript{\textminus2}) and passivation (1.22 \textmu A cm\textsuperscript{\textminus2}) current densities, with the lowest corrosion potential of \textminus0.648 V and long-range stability against pitting, corroborating its excellent performance in phosphate buffer solution at 37 {\textdegree}C. The oxide layer is comprised of TiO\textsubscript{2}, TiO\textsubscript{\emph{x}} and ZrO\textsubscript{\emph{x}}, as determined using X-ray photoelectron spectroscopy by short-term ion-etching of the surface layer. The two orders of magnitude increase in the oxide and interface resistance (from 14 to 1257 {\textOmega} cm\textsuperscript{2}) along with an order of magnitude decrease in the capacitance parameter of the oxide interface (from 1.402 x 10\textsuperscript{\textminus5} to 1.677 x 10\textsuperscript{\textminus6} S s\textsuperscript{n} cm\textsuperscript{\textminus2}) of the same composition is linked to the formation of carbonyl groups and reduction of the native oxide layer during linear sweep voltammetry.
△ Less
Submitted 1 August, 2022; v1 submitted 14 July, 2022;
originally announced July 2022.
-
Manufacturing of Textured Bulk Fe-SmCo$_{5}$ Magnets by Severe Plastic Deformation
Authors:
Lukas Weissitsch,
Martin Stückler,
Stefan Wurster,
Juraj Todt,
Peter Knoll,
Heinz Krenn,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
Exchange-coupling between soft- and hard-magnetic phases plays an important role in the engineering of novel magnetic materials. To achieve exchange coupling, a two-phase microstructure is necessary. This interface effect is further enhanced if both phase dimensions are reduced to the nanometer scale. At the same time, it is challenging to obtain large sample dimensions. In this study, powder blen…
▽ More
Exchange-coupling between soft- and hard-magnetic phases plays an important role in the engineering of novel magnetic materials. To achieve exchange coupling, a two-phase microstructure is necessary. This interface effect is further enhanced if both phase dimensions are reduced to the nanometer scale. At the same time, it is challenging to obtain large sample dimensions. In this study, powder blends and ball-milled powder blends of Fe-SmCo$_{5}$ are consolidated and are deformed by high-pressure torsion (HPT), as this technique allows us to produce bulk magnetic materials of reasonable sizes. Additionally, the effect of severe deformation by ball-milling and severe plastic deformation by HPT on exchange coupling in Fe-SmCo$_{5}$ composites is investigated. Due to the applied shear deformation, it is possible to obtain a texture in both phases, resulting in an anisotropic magnetic behavior and an improved magnetic performance.
△ Less
Submitted 4 April, 2022;
originally announced April 2022.
-
GNN-Surrogate: A Hierarchical and Adaptive Graph Neural Network for Parameter Space Exploration of Unstructured-Mesh Ocean Simulations
Authors:
Neng Shi,
Jiayi Xu,
Skylar W. Wurster,
Hanqi Guo,
Jonathan Woodring,
Luke P. Van Roekel,
Han-Wei Shen
Abstract:
We propose GNN-Surrogate, a graph neural network-based surrogate model to explore the parameter space of ocean climate simulations. Parameter space exploration is important for domain scientists to understand the influence of input parameters (e.g., wind stress) on the simulation output (e.g., temperature). The exploration requires scientists to exhaust the complicated parameter space by running a…
▽ More
We propose GNN-Surrogate, a graph neural network-based surrogate model to explore the parameter space of ocean climate simulations. Parameter space exploration is important for domain scientists to understand the influence of input parameters (e.g., wind stress) on the simulation output (e.g., temperature). The exploration requires scientists to exhaust the complicated parameter space by running a batch of computationally expensive simulations. Our approach improves the efficiency of parameter space exploration with a surrogate model that predicts the simulation outputs accurately and efficiently. Specifically, GNN-Surrogate predicts the output field with given simulation parameters so scientists can explore the simulation parameter space with visualizations from user-specified visual map**s. Moreover, our graph-based techniques are designed for unstructured meshes, making the exploration of simulation outputs on irregular grids efficient. For efficient training, we generate hierarchical graphs and use adaptive resolutions. We give quantitative and qualitative evaluations on the MPAS-Ocean simulation to demonstrate the effectiveness and efficiency of GNN-Surrogate. Source code is publicly available at https://github.com/trainsn/GNN-Surrogate.
△ Less
Submitted 21 February, 2022; v1 submitted 17 February, 2022;
originally announced February 2022.
-
Oxide-stabilized microstructure of severe plastically deformed CuCo alloys
Authors:
Martin Stueckler,
Jakub Zalesak,
Timo Mueller,
Stefan Wurster,
Lukas Weissitsch,
Martin Meier,
Peter Felfer,
Christoph Gammer,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
Nanocrystalline materials are well known for their beneficial mechanical and physical properties. However, it is of utmost importance to stabilize the microstructure at elevated temperatures to broaden the window of application e.g. by pinning grain boundaries through impurities or oxides. CuCo alloys, severe plastically deformed using high-pressure torsion, were used to investigate the evolution…
▽ More
Nanocrystalline materials are well known for their beneficial mechanical and physical properties. However, it is of utmost importance to stabilize the microstructure at elevated temperatures to broaden the window of application e.g. by pinning grain boundaries through impurities or oxides. CuCo alloys, severe plastically deformed using high-pressure torsion, were used to investigate the evolution of oxides upon annealing. These investigations were performed using electron microscopy, synchrotron high-energy X-ray scattering and atom probe tomography. Monitoring the evolution of oxides by in-situ small angle scattering investigations indicates the growth of primary oxides as well as the formation of another species of oxide at elevated temperatures. Slightly different coarsening can be observed in heat-treated samples, which can be traced back to the influence of different oxide amounts. The initially supersaturated CuCo matrix changes by the decomposition process as well as Co- and Cu-based oxides formation and growth. It has been found that the major number of oxides which are present after annealing, can already be found in the starting materials.
△ Less
Submitted 11 January, 2022;
originally announced January 2022.
-
Soft Magnetic Properties of Ultra-Strong and Nanocrystalline Pearlitic Wires
Authors:
Stefan Wurster,
Martin Stückler,
Lukas Weissitsch,
Heinz Krenn,
Anton Hohenwarter,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
The paper describes the capability of magnetic softening of a coarse grained bulk material by a severe deformation technique. Connecting the microstructure with magnetic properties, the coercive field decreases dramatically for grains smaller than the magnetic exchange length. This makes the investigation of soft magnetic properties of severely drawn pearlitic wires very interesting. With the help…
▽ More
The paper describes the capability of magnetic softening of a coarse grained bulk material by a severe deformation technique. Connecting the microstructure with magnetic properties, the coercive field decreases dramatically for grains smaller than the magnetic exchange length. This makes the investigation of soft magnetic properties of severely drawn pearlitic wires very interesting. With the help of the starting two phase microstructure, it is possible to substantially refine the material, which allows the investigation of magnetic properties for nanocrystalline bulk material. Compared to the coarse grained initial, pearlitic state, the coercivities of the highly deformed wires decrease while the saturation magnetization values increase, even beyond the value expectable from the individual constituents. The lowest coercivity in the drawn state is found to be 520 A m for a wire of 24 um thickness and an annealing treatment has a further positive effect on it. The decreasing coercivity is discussed in the framework of two opposing models: grain refinement on the one hand and dissolution of cementite on the other hand. Auxiliary measurements give a clear indication for the latter model, delivering a sufficient description of the observed evolution of magnetic properties.
△ Less
Submitted 10 January, 2022;
originally announced January 2022.
-
Nanocrystalline FeCr alloys synthesised by severe plastic deformation -- a potential material for exchange bias and enhanced magnetostriction
Authors:
Lukas Weissitsch,
Stefan Wurster,
Alexander Paulischin,
Martin Stückler,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
This work gives insights into processing and characterisation of bulk nanocrystalline FeCr materials. The investigated FeCr alloys, consisting of 30, 50 and 70 at.% ferromagnetic Fe and remaining anti-ferromagnetic Cr, are processed by arc melting and subsequent severe plastic deformation by high pressure torsion. The physical similarities between elemental Fe and Cr in combination with the nanocr…
▽ More
This work gives insights into processing and characterisation of bulk nanocrystalline FeCr materials. The investigated FeCr alloys, consisting of 30, 50 and 70 at.% ferromagnetic Fe and remaining anti-ferromagnetic Cr, are processed by arc melting and subsequent severe plastic deformation by high pressure torsion. The physical similarities between elemental Fe and Cr in combination with the nanocrystalline structure of the as-deformed alloys, necessitates advanced characterisation techniques for the as-deformed state: In-situ annealing synchrotron X-ray diffraction measurements as well as electron microscopy experiments are linked to magnetostrictive measurements and reveal a single phase microstructure. Surprisingly, the nanocrystalline FeCr alloys remain supersaturated solid solutions upon annealing above 500°C, meaning a decomposition in a FeCr nanocomposite is suppressed. For the chosen annealing conditions grain growth is faster than decomposition and enhanced magnetostrictive values are found compared to materials in the as-deformed state.
△ Less
Submitted 27 October, 2021;
originally announced October 2021.
-
Reinforcement Learning for Load-balanced Parallel Particle Tracing
Authors:
Jiayi Xu,
Hanqi Guo,
Han-Wei Shen,
Mukund Raj,
Skylar W. Wurster,
Tom Peterka
Abstract:
We explore an online reinforcement learning (RL) paradigm to dynamically optimize parallel particle tracing performance in distributed-memory systems. Our method combines three novel components: (1) a work donation algorithm, (2) a high-order workload estimation model, and (3) a communication cost model. First, we design an RL-based work donation algorithm. Our algorithm monitors workloads of proc…
▽ More
We explore an online reinforcement learning (RL) paradigm to dynamically optimize parallel particle tracing performance in distributed-memory systems. Our method combines three novel components: (1) a work donation algorithm, (2) a high-order workload estimation model, and (3) a communication cost model. First, we design an RL-based work donation algorithm. Our algorithm monitors workloads of processes and creates RL agents to donate data blocks and particles from high-workload processes to low-workload processes to minimize program execution time. The agents learn the donation strategy on the fly based on reward and cost functions designed to consider processes' workload changes and data transfer costs of donation actions. Second, we propose a workload estimation model, hel** RL agents estimate the workload distribution of processes in future computations. Third, we design a communication cost model that considers both block and particle data exchange costs, hel** RL agents make effective decisions with minimized communication costs. We demonstrate that our algorithm adapts to different flow behaviors in large-scale fluid dynamics, ocean, and weather simulation data. Our algorithm improves parallel particle tracing performance in terms of parallel efficiency, load balance, and costs of I/O and communication for evaluations with up to 16,384 processors.
△ Less
Submitted 31 January, 2022; v1 submitted 12 September, 2021;
originally announced September 2021.
-
Deep Hierarchical Super Resolution for Scientific Data
Authors:
Skylar W. Wurster,
Hanqi Guo,
Han-Wei Shen,
Thomas Peterka,
Jiayi Xu
Abstract:
We present a novel technique for hierarchical super resolution (SR) with neural networks (NNs), which upscales volumetric data represented with an octree data structure to a high-resolution uniform grid with minimal seam artifacts on octree node boundaries. Our method uses existing state-of-the-art SR models and adds flexibility to upscale input data with varying levels of detail across the domain…
▽ More
We present a novel technique for hierarchical super resolution (SR) with neural networks (NNs), which upscales volumetric data represented with an octree data structure to a high-resolution uniform grid with minimal seam artifacts on octree node boundaries. Our method uses existing state-of-the-art SR models and adds flexibility to upscale input data with varying levels of detail across the domain, instead of only uniform grid data that are supported in previous approaches. The key is to use a hierarchy of SR NNs, each trained to perform 2x SR between two levels of detail, with a hierarchical SR algorithm that minimizes seam artifacts by starting from the coarsest level of detail and working up. We show that our hierarchical approach outperforms baseline interpolation and hierarchical upscaling methods, and demonstrate the usefulness of our proposed approach across three use cases including data reduction using hierarchical downsampling+SR instead of uniform downsampling+SR, computation savings for hierarchical finite-time Lyapunov exponent field calculation, and super-resolving low-resolution simulation results for a high-resolution approximation visualization.
△ Less
Submitted 14 October, 2022; v1 submitted 30 May, 2021;
originally announced July 2021.
-
Sampling the Cu-Fe-Co phase diagram by severe plastic deformation for enhanced soft magnetic properties
Authors:
Martin Stückler,
Stefan Wurster,
Lukas Weissitsch,
Heinz Krenn,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
Initial powder mixtures of Cu, Fe and Co are exposed to severe plastic deformation by high-pressure torsion to prepare solid solutions. A broad range of compositions is investigated, whereas this study aims at the synthesis of soft magnetic materials and therefore at the formation of a homogeneous and nanocrystalline microstructure. For intermediate ferromagnetic contents, high-pressure torsion at…
▽ More
Initial powder mixtures of Cu, Fe and Co are exposed to severe plastic deformation by high-pressure torsion to prepare solid solutions. A broad range of compositions is investigated, whereas this study aims at the synthesis of soft magnetic materials and therefore at the formation of a homogeneous and nanocrystalline microstructure. For intermediate ferromagnetic contents, high-pressure torsion at room temperature yields single-phase supersaturated solid solutions. For higher ferromagnetic contents, two consecutive steps of high-pressure torsion deformation at different temperatures yield the desired nanocrystalline microstructure. Depending on the Co-to-Fe-ratio, either a single-phase supersaturated solid solution or a nanocomposite forms. The composite exhibits an enhanced magnetic moment, indicating the formation of a (Fe,Co)-alloy upon severe plastic deformation. Soft magnetic properties are verified for large Co-to-Fe-ratios and this microstructure is found to remain stable up to 400 °C.
△ Less
Submitted 6 April, 2021;
originally announced April 2021.
-
On the magnetic nanostructure of a Co-Cu alloy processed by high-pressure torsion
Authors:
Martin Stückler,
Christian Teichert,
Aleksandar Matković,
Heinz Krenn,
Lukas Weissitsch,
Stefan Wurster,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
In this study, a preparation route of Co-Cu alloys with soft magnetic properties by high-pressure torsion deformation is introduced. Nanocrystalline, supersaturated single-phase microstructures are obtained after deformation of Co-Cu alloys, which are prepared from an initial powder mixture with Co-contents above 70 wt.%. Isochronal annealing treatments up to 400°C further reveal a remarkable micr…
▽ More
In this study, a preparation route of Co-Cu alloys with soft magnetic properties by high-pressure torsion deformation is introduced. Nanocrystalline, supersaturated single-phase microstructures are obtained after deformation of Co-Cu alloys, which are prepared from an initial powder mixture with Co-contents above 70 wt.%. Isochronal annealing treatments up to 400°C further reveal a remarkable microstructural stability. Only at 600°C, the supersaturated phase decomposes into two fcc-phases. The coercivity, measured by SQUID as a function of annealing temperature, remains significantly below the value for bulk-Co in all states investigated. In order to understand the measured magnetic properties in detail, a quantitative analysis of the magnetic microstructure is carried out by magnetic force microscopy and correlated to the observed changes in coercivity. Our results show that the rising coercivity can be explained by a magnetic hardening effect occurring in context with spinodal decomposition.
△ Less
Submitted 15 February, 2021;
originally announced February 2021.
-
In situ AC-hysteresis measurements of SPD-processed Cu20(Fe15Co85)80
Authors:
Martin Stückler,
Stefan Wurster,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
The changes of magnetic properties upon heat treatment of a metastable supersaturated solid solution processed by severe plastic deformation are investigated by in-situ AC-hysteresis measurements. Data are analyzed in the framework of dynamic loss theory, with correlative investigations of the microstructural properties. The evolution of hysteresis upon annealing points out that the single-phase s…
▽ More
The changes of magnetic properties upon heat treatment of a metastable supersaturated solid solution processed by severe plastic deformation are investigated by in-situ AC-hysteresis measurements. Data are analyzed in the framework of dynamic loss theory, with correlative investigations of the microstructural properties. The evolution of hysteresis upon annealing points out that the single-phase supersaturated solid solution remains stable up to 400°C, then hindering of domain wall motion sets in at this temperature. At 600°C, a multi phase microstructure is present, causing a significant increase in coercivity.
△ Less
Submitted 18 January, 2021;
originally announced January 2021.
-
Processing of nanostructured bulk Fe-Cr alloys by severe plastic deformation
Authors:
Lukas Weissitsch,
Martin Stückler,
Stefan Wurster,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
The processing of binary alloys consisting of ferromagnetic Fe and antiferromagnetic Cr by severe plastic deformation (SPD) with different chemical compositions has been investigated. Although the phase diagram exhibits a large gap in the thermodynamical equilibrium at lower temperatures, it is shown that techniques based on SPD help to overcome common processing limits. Different processing route…
▽ More
The processing of binary alloys consisting of ferromagnetic Fe and antiferromagnetic Cr by severe plastic deformation (SPD) with different chemical compositions has been investigated. Although the phase diagram exhibits a large gap in the thermodynamical equilibrium at lower temperatures, it is shown that techniques based on SPD help to overcome common processing limits. Different processing routes including initial ball milling (BM) and arc melting (AM) and a concatenation with annealing treatments prior to high-pressure torsion (HPT) deformation are compared in this work. Investigation of the deformed microstructures by electron microscopy and synchrotron X-ray diffraction reveal homogeneous, nanocrystalline microstructures for HPT deformed AM alloys. HPT deformation of powder blends and BM powders leads to an exorbitant increase in hardness or an unusual fast formation of a $σ$-phase and therefore impede successful processing.
△ Less
Submitted 11 January, 2021;
originally announced January 2021.
-
Strain Induced Anisotropic Magnetic Behaviour and Exchange Coupling Effect in Fe-SmCo$_{5}$ Permanent Magnets Generated by High Pressure Torsion
Authors:
Lukas Weissitsch,
Martin Stückler,
Stefan Wurster,
Peter Knoll,
Heinz Krenn,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
High-pressure torsion (HPT), a technique of severe plastic deformation (SPD), is shown as a promising processing method for exchange-spring magnetic materials in bulk form. Powder mixtures of Fe and SmCo$_{5}$ are consolidated and deformed by HPT exhibiting sample dimensions of several millimetres, being essential for bulky magnetic applications. The structural evolution during HPT deformation of…
▽ More
High-pressure torsion (HPT), a technique of severe plastic deformation (SPD), is shown as a promising processing method for exchange-spring magnetic materials in bulk form. Powder mixtures of Fe and SmCo$_{5}$ are consolidated and deformed by HPT exhibiting sample dimensions of several millimetres, being essential for bulky magnetic applications. The structural evolution during HPT deformation of Fe-SmCo$_{5}$ compounds at room- and elevated- temperatures of chemical compositions consisting of 87, 47, 24 and 10 wt.% Fe is studied and microstructurally analysed. Electron microscopy and synchrotron X-ray diffraction reveal a dual-phase nanostructured composite for the as-deformed samples with grain refinement after HPT deformation. SQUID magnetometry measurements show hysteresis curves of an exchange coupled nanocomposite at room temperature, while for low temperatures a decoupling of Fe and SmCo$_{5}$ is observed. Furthermore, exchange interactions between the hard- and soft-magnetic phase can explain a shift of the hysteresis curve. Strong emphasis is devoted to the correlation between the magnetic properties and the evolving nano-structure during HPT deformation, which is conducted for a 1:1 composition ratio of Fe to SmCo$_{5}$. SQUID magnetometry measurements show an increasing saturation magnetisation for increasing strain $γ$ and a maximum of the coercive field strength at a shear strain of $γ$ = 75.
△ Less
Submitted 19 January, 2021; v1 submitted 11 January, 2021;
originally announced January 2021.
-
A new approach to evaluate the elastic modulus of metallic foils
Authors:
C. O. W. Trost,
S. Wurster,
C. Freitag,
A. Steinberger,
M. J. Cordill
Abstract:
The accurate determination of the elastic properties is non-trivial for metallic foils. The measured elastic modulus is often described in literature as significantly smaller than the respective modulus of the bulk counterparts. This paper describes a straight forward way to minimize the influence of the measurement on the calculated values using a standard tensile testing device combined with a l…
▽ More
The accurate determination of the elastic properties is non-trivial for metallic foils. The measured elastic modulus is often described in literature as significantly smaller than the respective modulus of the bulk counterparts. This paper describes a straight forward way to minimize the influence of the measurement on the calculated values using a standard tensile testing device combined with a laser speckle extensometer and advanced data evaluation. The elasticmodulus obtained with a monotonic tensile procedure is compared to values obtained frommultiple loading unloading cycles in the elastic regime. The latter were found to lead to more reproducible and reasonable values with significantly smaller standard deviations. To enable interpretation of the modulus values electron backscatter diffraction and X ray diffraction were used to emphasize the effects of texture and microstructure. Two different electrodeposited copper foilswere characterized since copper iswidely used in many industrial applications.
△ Less
Submitted 29 September, 2020;
originally announced September 2020.
-
Fracture mechanics of micro samples: Fundamental considerations
Authors:
Reinhard Pippan,
Stefan Wurster,
Daniel Kiener
Abstract:
In this review article we consider the crack growth resistance ofmicrometer and submicrometer sized samples from the fracture mechanics point of view. Standard fracture mechanics test procedures were developed for macroscale samples, and reduction of the specimen dimensions by three to five orders of magnitude has severe consequences. This concerns the interpretation of results obtained by micro a…
▽ More
In this review article we consider the crack growth resistance ofmicrometer and submicrometer sized samples from the fracture mechanics point of view. Standard fracture mechanics test procedures were developed for macroscale samples, and reduction of the specimen dimensions by three to five orders of magnitude has severe consequences. This concerns the interpretation of results obtained by micro and nanomechanics, as well as the life time and failure prediction of micro and nano devices. We discuss the relevant fracture mechanics length scales and their relation to the material specific structural lengths in order to conduct rigorous fracture mechanics experiments. To ensure general validity and applicability of evaluation concepts, these scaling considerations are detailed for ideally brittle, semi brittle and micro ductile crack propagation, subject to both monotonic and cyclic loading. Special attention is devoted to the requirements for determining specimen size for various loading types to measure material characteristic crack propagation resistance at small scales. Finally, we discuss novel possibilities of micron and submicron fracture mechanics tests to improve the basic understanding of specific crack propagation processes.
△ Less
Submitted 4 September, 2020;
originally announced September 2020.
-
An analytical solution for the correct determination of crack lengths via cantilever stiffness
Authors:
Markus Alfreider,
Stefan Kolitsch,
Stefan Wurster,
Daniel Kiener
Abstract:
The present work provides an analytic solution for the stiffness to crack length relation in microscopic cantilever shaped fracture specimens based on classical beam theory and substitution of the crack by a virtual rotational spring element. The resulting compact relationship allows for accounting of the actual beamgeometry and agrees very well with accompanying finite element simulations. Compar…
▽ More
The present work provides an analytic solution for the stiffness to crack length relation in microscopic cantilever shaped fracture specimens based on classical beam theory and substitution of the crack by a virtual rotational spring element. The resulting compact relationship allows for accounting of the actual beamgeometry and agrees very well with accompanying finite element simulations. Compared with the only other model present in literature the proposed relationship reduces the deviation between model and data to a maximum of 1.6% fromthe previous minimumof 15%. Thus, the novel solution will help to reduce the necessity for individual simulations and aim to increase the comparability of elastic-plastic microcantilever fracture experiments in the future.
△ Less
Submitted 4 September, 2020;
originally announced September 2020.
-
Microstructural Changes Influencing the Magnetoresistive Behavior of Bulk Nanocrystalline Materials
Authors:
Stefan Wurster,
Martin Stueckler,
Lukas Weissitsch,
Timo Mueller,
Andrea Bachmaier
Abstract:
Bulk nanocrystalline materials of small and medium ferromagnetic content were produced using severe plastic deformation by high-pressure torsion at roomtemperature. Giant magnetoresistive behavior was found for as deformed materials, which was further improved by adjusting the microstructure with thermal treatments. The adequate range of annealing temperatures was assessed with in-situ synchrotron…
▽ More
Bulk nanocrystalline materials of small and medium ferromagnetic content were produced using severe plastic deformation by high-pressure torsion at roomtemperature. Giant magnetoresistive behavior was found for as deformed materials, which was further improved by adjusting the microstructure with thermal treatments. The adequate range of annealing temperatures was assessed with in-situ synchrotron diffraction measurements. Thermally treated CuCo materials show larger giant magnetoresistance after annealing for 1 h at 300C, while for CuFe this annealing temperature is too high and decreases the magnetoresistive properties. The improvement of magnetoresistivity by thermal treatments is discussed with respect to the microstructural evolution as observed by electron microscopy and ex situ synchrotron diffraction measurements.
△ Less
Submitted 7 September, 2020;
originally announced September 2020.
-
Intermixing of Fe and Cu on the atomic scale by high-pressure torsion as revealed by DC- and AC-SQUID susceptometry and atom probe tomography
Authors:
Martin Stückler,
Heinz Krenn,
Philipp Kürnsteiner,
Baptiste Gault,
Frédéric De Geuser,
Lukas Weissitsch,
Stefan Wurster,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
The capability of high-pressure torsion on the preparation of supersaturated solid solutions, consisting of Cu-14Fe (wt.%), is studied. From microstructural investigations a steady state is obtained with nanocrystalline grains. The as-deformed state is analyzed with atom probe tomography, revealing an enhanced solubility and the presence of Fe-rich particles. The DC-hysteresis loop shows suppresse…
▽ More
The capability of high-pressure torsion on the preparation of supersaturated solid solutions, consisting of Cu-14Fe (wt.%), is studied. From microstructural investigations a steady state is obtained with nanocrystalline grains. The as-deformed state is analyzed with atom probe tomography, revealing an enhanced solubility and the presence of Fe-rich particles. The DC-hysteresis loop shows suppressed long range interactions in the as-deformed state and evolves towards a typical bulk hysteresis loop when annealed at 500°C. AC-susceptometry measurements of the as-deformed state reveal the presence of a superparamagnetic blocking peak, as well as a magnetic frustrated phase, whereas the transition of the latter follows the Almeida-Thouless line, coinciding with the microstructural investigations by atom probe tomography. AC-susceptometry shows that the frustrated state vanishes for annealing at 250°C.
△ Less
Submitted 7 September, 2020;
originally announced September 2020.
-
Tuneable Magneto-Resistance by Severe Plastic Deformation
Authors:
Stefan Wurster,
Lukas Weissitsch,
Martin Stueckler,
Peter Knoll,
Heinz Krenn,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
Bulk metallic samples were synthesized from different binary powder mixtures consisting of elemental Cu, Co, and Fe using severe plastic deformation. Small particles of the ferromagnetic phase originate in the conductive Cu phase, either by incomplete dissolution or by segregation phenomena during the deformation process. These small particles are known to give rise to granular giant magnetoresist…
▽ More
Bulk metallic samples were synthesized from different binary powder mixtures consisting of elemental Cu, Co, and Fe using severe plastic deformation. Small particles of the ferromagnetic phase originate in the conductive Cu phase, either by incomplete dissolution or by segregation phenomena during the deformation process. These small particles are known to give rise to granular giant magnetoresistance. Taking advantage of the simple production process, it is possible to perform a systematic study on the influence of processing parameters and material compositions on the magneto-resistance. Furthermore, it is feasible to tune the magnetoresistive behavior as a function of the specimens chemical composition. It was found that specimens of low ferromagnetic content show an almost isotropic drop in resistance in a magnetic field. With increasing ferromagnetic content, percolating ferromagnetic phases cause an anisotropy of the magnetoresistance. By changing the parameters of the high pressure torsion process, i.e., sample size, deformation temperature, and strain rate, it is possible to tailor the magnitude of giant magneto-resistance. A decrease in room temperature resistivity of approx. 3.5% was found for a bulk specimen containing an approximately equiatomic fraction of Co and Cu.
△ Less
Submitted 7 September, 2020;
originally announced September 2020.
-
Magnetic Binary Supersaturated Solid Solutions Processed by Severe Plastic Deformation
Authors:
Martin Stückler,
Heinz Krenn,
Reinhard Pippan,
Lukas Weissitsch,
Stefan Wurster,
Andrea Bachmaier
Abstract:
Samples consisting of one ferromagnetic and one diamagnetic component which are immiscible at the thermodynamic equilibrium (Co-Cu, Fe-Cu, Fe-Ag) are processed by high-pressure torsion at various compositions. The received microstructures are investigated by electron microscopy and synchrotron X-ray diffraction, showing a microstructural saturation. Results gained from microstructural investigatio…
▽ More
Samples consisting of one ferromagnetic and one diamagnetic component which are immiscible at the thermodynamic equilibrium (Co-Cu, Fe-Cu, Fe-Ag) are processed by high-pressure torsion at various compositions. The received microstructures are investigated by electron microscopy and synchrotron X-ray diffraction, showing a microstructural saturation. Results gained from microstructural investigations are correlated to magnetometry data. The Co-Cu samples show mainly ferromagnetic behavior and a decrease in coercivity with increasing Co-content. The saturation microstructure of Fe-Cu samples is found to be dual phase. Results of magnetic measurements also revealed the occurrence of two different magnetic phases in this system. For Fe-Ag, the microstructural and magnetic results indicate that no intermixing between the elemental phases takes place.
△ Less
Submitted 7 September, 2020;
originally announced September 2020.
-
Magnetic dilution by severe plastic deformation
Authors:
Martin Stückler,
Lukas Weissitsch,
Stefan Wurster,
Peter Felfer,
Heinz Krenn,
Reinhard Pippan,
Andrea Bachmaier
Abstract:
Mixtures of Fe and Cu powders are cold-compacted and subsequently deformed with severe plastic deformation by high-pressure torsion, leading to bulk samples. The dilution of Fe in the Cu matrix is investigated with SQUID-magnetometry, whereas the magnetic properties change as a function of Fe-content from a frustrated regime to a thermal activated behaviour. The magnetic properties are correlated…
▽ More
Mixtures of Fe and Cu powders are cold-compacted and subsequently deformed with severe plastic deformation by high-pressure torsion, leading to bulk samples. The dilution of Fe in the Cu matrix is investigated with SQUID-magnetometry, whereas the magnetic properties change as a function of Fe-content from a frustrated regime to a thermal activated behaviour. The magnetic properties are correlated with the microstructure, investigated by synchrotron X-ray diffraction and atom probe tomography. Annealing of the as-deformed states leads to demixing and grain growth, with the coercivity as a function of annealing temperature obeying the random anisotropy model. The presented results show that high-pressure torsion is a technique capable to affect the microstructure even on atomic length scales.
△ Less
Submitted 7 September, 2020;
originally announced September 2020.
-
Microstructure and Failure Characteristics of Nanostructured Molybdenum Copper Composites
Authors:
Katharina Schwarz,
Julian Rosalie,
Stefan Wurster,
Reinhard Pippan,
Anton Hohenwarter
Abstract:
Liquid-metal infiltrated Cu30Mo70 (weigth percentage) is subjected to severe plastic deformation using high pressure torsion. The initially equiaxed dual phase structure is gradually transformed into a lamellar structure composed of individual Cu and Mo layers. The thickness of the lamellae varies between the micrometer and nanometer ranges depending on the amount of applied strain. Consistent wit…
▽ More
Liquid-metal infiltrated Cu30Mo70 (weigth percentage) is subjected to severe plastic deformation using high pressure torsion. The initially equiaxed dual phase structure is gradually transformed into a lamellar structure composed of individual Cu and Mo layers. The thickness of the lamellae varies between the micrometer and nanometer ranges depending on the amount of applied strain. Consistent with the refinement of the microstructural features, strength and hardness substantially increase. In addition, an acceptable ductility is found in the intermediate deformation range. An assessment of the damage tolerance of the produced composites is performed by measuring the fracture toughness in different crack propagation directions. The results indicate the development of a pronounced anisotropy with increasing degree of deformation which is an effect of the concurrent alignment of the nanostructured lamellar composite into the shear plane.
△ Less
Submitted 4 September, 2020;
originally announced September 2020.
-
Human Gist Processing Augments Deep Learning Breast Cancer Risk Assessment
Authors:
Skylar W. Wurster,
Arkadiusz Sitek,
Jian Chen,
Karla Evans,
Gaeun Kim,
Jeremy M. Wolfe
Abstract:
Radiologists can classify a mammogram as normal or abnormal at better than chance levels after less than a second's exposure to the images. In this work, we combine these radiologists' gist inputs into pre-trained machine learning models to validate that integrating gist with a CNN model can achieve an AUC (area under the curve) statistically significantly higher than either the gist perception of…
▽ More
Radiologists can classify a mammogram as normal or abnormal at better than chance levels after less than a second's exposure to the images. In this work, we combine these radiologists' gist inputs into pre-trained machine learning models to validate that integrating gist with a CNN model can achieve an AUC (area under the curve) statistically significantly higher than either the gist perception of radiologists or the model without gist input.
△ Less
Submitted 27 November, 2019;
originally announced December 2019.
