Search NASA⌕ Search

SEARCH · Search NASA

Results for “512-S”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2012

Performance of a 512 x 512 Gated CMOS Imager with a 250 ps Exposure Time

We describe the performance of a 512x512 gated CMOS read out integrated circuit (ROIC) with a 250 ps exposure time. A low-skew, H-tree trigger distribution system is used to locally generate individual pixel gates in each 8x8 neighborhood of the ROIC. The temporal width of the gate is voltage controlled and user selectable via a precision potentiometer. The gating implementation was first validated in optical tests of a 64x64 pixel prototype ROIC developed as a proof-of-concept during the early phases of the development program. The layout of the H-Tree addresses each quadrant of the ROIC independently and admits operation of the ROIC in two modes. If “common mode” triggering is used, the camera provides a single 512x512 image. If independent triggers are used, the camera can provide up to four 256x256 images with a frame separation set by the trigger intervals. The ROIC design includes small (sub-pixel) optical photodiode structures to allow test and characterization of the ROIC using optical sources prior to bump bonding. Reported test results were obtained using short pulse, second harmonic Ti:Sapphire laser systems operating at λ~ 400 nm at sub-ps pulse widths.

47 OTHER INSTRUMENTATION↗
DOE OSTI2011

Identification of γ -ray emission from 3C 345 and NRAO 512

For more than 15 years, since the days of the Energetic Gamma-Ray Experiment Telescope (EGRET) on board the Compton Gamma-Ray Observatory (CGRO; 1991–2000), it has remained an open question why the prominent blazar 3C 345 was not reliably detected at γ-ray energies ≥ 20 MeV. Recently a bright γ-ray source (0FGL J1641.4+3939/1FGL J1642.5+3947), potentially associated with 3C 345, was detected by the Large Area Telescope (LAT) on Fermi . Multiwavelength observations from radio bands to X-rays (mainly GASP-WEBT and Swift) of possible counterparts (3C 345, NRAO 512, B3 1640 + 396) were combined with 20 months of Fermi -LAT monitoring data (August 2008 – April 2010) to associate and identify the dominating γ-ray emitting counterpart of 1FGL J1642.5+3947. The source 3C 345 is identified as the main contributor for this γ-ray emitting region. However, after November 2009 (15 months), a significant excess of photons from the nearby quasar NRAO 512 started to contribute and thereafter was detected with increasing γ-ray activity, possibly adding flux to 1FGL J1642.5+3947. As a result, for the same time period and during the summer of 2010, an increase of radio, optical and X-ray activity of NRAO 512 was observed. No γ-ray emission from B3 1640 + 396 was detected.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Ultra-Fast Non-Volatile Resistive Switching Devices with Over 512 Distinct and Stable Levels for Memory and Neuromorphic Computing

Low-current multilevel programmability with inherent non-volatility and high stability of resistance states is required for both multi-bit memory storage and deep learning accelerators but is difficult to achieve. Here, in a resistive switching system, this work realizes >512 (>9 bits) distinct non-volatile conductance levels with stable retention for each state with current levels down to the nanoampere range, highly promising for potential integration with small processing nodes with ultra-low power consumption requirements. This is achieved by demonstrating a new thin film design concept that encompasses three key features: an ultra-thin epitaxial oxygen ionic switching layer that provides a tunable energy barrier at the bottom electrode, an overcoat amorphous layer that acts as an ion migration barrier for stable state retention, and a partial conductive filament as a localized electronic transport channel to the epitaxial switching layer. A large dynamic resistance range of up to seven orders of magnitude is achieved with reset-free transitions among intermediate states, and programmability is demonstrated with ultra-fast (20 ns) pulses. Artificial neural network (ANN) simulations, based on the experimental performance and its non-idealities, demonstrate close-to-ideal inference accuracies for various Modified National Institute of Standards and Technology (MNIST) data sets.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Results for Actinide Removal Process 512-S Guard Filter Analyses

The Actinide Removal Process (ARP) removes Sr and actinides by adsorption to monosodium titanate (MST) from radioactive salt solutions that have been stored at the Savannah River Site (SRS). The MST solids are then removed by filtration. Material from the 512-S Late Wash Precipitate Tank (LWPT) is cross-flow filtered and the filtrate sent through a secondary filter (guard filter) to the Late Wash Hold Tank (LWHT) for further processing in the Modular Caustic Side Solvent Extraction (CSSX) Unit (MCU).

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2024

A Modular 512-Channel Neural Signal Acquisition ASIC for High-Density 4096 Channel Electrophysiology

The complexity of information processing in the brain requires the development of technologies that can provide spatial and temporal resolution by means of dense electrode arrays paired with high-channel-count signal acquisition electronics. In this work, we present an ultra-low noise modular 512-channel neural recording circuit that is scalable to up to 4096 simultaneously recording channels. The neural readout application-specific integrated circuit (ASIC) uses a dense 8.2 mm × 6.8 mm 2D layout to enable high-channel count, creating an ultra-light 350 mg flexible module. The module can be deployed on headstages for small animals like rodents and songbirds, and it can be integrated with a variety of electrode arrays. The chip was fabricated in a TSMC 0.18 µm 1.8 V CMOS technology and dissipates a total of 125 mW. Each DC-coupled channel features a gain and bandwidth programmable analog front-end along with 14 b analog-to-digital conversion at speeds up to 30 kS/s. Additionally, each front-end includes programmable electrode plating and electrode impedance measurement capability. We present both standalone and in vivo measurements results, demonstrating the readout of spikes and field potentials that are modulated by a sensory input.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Principles of Neutron Scattering from Condensed Matter. By Andrew T. Boothroyd. Oxford University Press, 2020. Pp. 512. Price GBP 55.00. ISBN 9780198862314.

Neutron scattering user facilities are increasingly used by ‘non-expert’ users, opening neutron scattering to new science fields and greatly increasing the scientific impact of these facilities. This book aims to bring the needed introduction to the expanding area of neutron scattering techniques to this wider audience of scientists using neutron scattering as part of their experimental portfolio. In my opinion, it completely succeeds in that goal and provides the right balance of introduction and in-depth discussion. I also like the layout of the book, with some figures and notes in the margin making it easy to read and navigate. As a ‘long-time’ neutron scatterer, I am excited to add this book to my book shelf.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2014

SU-E-T-512: Electromagnetic Simulations of the Dielectric Wall Accelerator

Purpose: To characterize and parametrically study the key components of a dielectric wall accelerator through electromagnetic modeling and particle tracking. Methods: Electromagnetic and particle tracking simulations were performed using a commercial code (CST Microwave Studio, CST Inc.) utilizing the finite integration technique. A dielectric wall accelerator consists of a series of stacked transmission lines sequentially fired in synchrony with an ion pulse. Numerous properties of the stacked transmission lines, including geometric, material, and electronic properties, were analyzed and varied in order to assess their impact on the transverse and axial electric fields. Additionally, stacks of transmission lines were simulated in order to quantify the parasitic effect observed in closely packed lines. Particle tracking simulations using the particle-in-cell method were performed on the various stacks to determine the impact of the above properties on the resultant phase space of the ions. Results: Examination of the simulation results show that novel geometries can shape the accelerating pulse in order to reduce the energy spread and increase the average energy of accelerated ions. Parasitic effects were quantified for various geometries and found to vary with distance from the end of the transmission line and along the beam axis. An optimal arrival time of an ion pulse relative to the triggering of the transmission lines for a given geometry was determined through parametric study. Benchmark simulations of single transmission lines agree well with published experimental results. Conclusion: This work characterized the behavior of the transmission lines used in a dielectric wall accelerator and used this information to improve them in novel ways. Utilizing novel geometries, we were able to improve the accelerating gradient and phase space of the accelerated particle bunch. Through simulation, we were able to discover and optimize design issues with the device at low cost. Funding: Morgridge Institute for Research, Madison WI; Conflict of Interest: Dr. Mackie is an investor and board member at CPAC, a company developing compact accelerator designs similar to those discussed in this work, but designs discussed are not directed by CPAC. Funding: Morgridge Institute for Research, Madison WI; Conflict of Interest: Dr. Mackie is an investor and board member at CPAC, a company developing compact accelerator designs similar to those discussed in this work, but designs discussed are not directed by CPAC.

43 PARTICLE ACCELERATORS↗
DOE OSTI2015

SU-E-T-512: Intrinsic Characteristics of the Nine Detectors and Evaluation of Their Performance in Non-Equilibrium Radiation Dosimetry

Purpose The aim for the study was to compare intrinsic characteristics of the nine detectors and evaluate their performance in non-equilibrium radiation dosimetry. Methods The intrinsic characteristics of the nine detectors that were evaluated are based on the composition and size of the active volume, operating voltage, initial recombination of the collected charge, temperature, the effective cross section of the detectors. The shortterm stability and collection efficiency has been investigated. The minimum radiation detection sensitivity and detectors leakage current has been measured. The sensitivity to changes in energy spectrum as well as change in incident beam angles were measured an analyzed. Results The short-term stability of the measurements within every detector showed consistency in the measured values with the highest value of the standard deviation of the mean not exceeding 0.5%. Air ion chamber detectors showed minimum sensitivity to change in incident beam angles while diode detectors underestimated measurements up to 16%. Comparing the slope of the tangents for detector’s sensitivity curve, diode detectors illustrate more sensitivity to change in photon spectrum than ion chamber detectors. The change in radiation detection sensitivity with increase in dose delivered has been observed for semiconductor detectors with maximum deviation 0.01% for doses between 1 Gy and 10 Gy. Leakage current has been mainly influenced by bias voltage (ion chamber detectors) and room light intensity (diode detectors). With dose per pulse varying from 1.47E−4 to 5.1E−4 Gy/pulse the maximum change in collection efficiency was 1.4% for the air ion chambers up to 8% for liquid filled ion chamber. Conclusion Broad range of measurements performed showed all the detectors susceptible to some limitations and while they are suitable for use in broad scope of applications, careful selection has to be made for particular range of measurements.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Analytic Framework for Optimal Sizing of Hydrogen Fueling Stations for Heavy Duty Vehicles at Ports (CRADA 512)

This CRADA presents the strategy that PNNL and Sandia National Laboratories (SNL) will take to support Seattle City Light (SCL) in properly sizing a hydrogen station for class 8 drayage trucks at the Port of Seattle (Port), as part of a coordinated plan of mixed uses for hydrogen. The proper sizing for the fueling station is unique from past experiences because it will be part of a coordinated plan for a multi-use scheme for the hydrogen, and not a stand-alone fueling station. The multiple use-streams, and their relative values, will impact all facets of station sizing and design, including fueling rates, storage technology and size, equipment selection, and safety considerations. The future deployment of this fueling station, and its proper sizing, is seen as a critical first step in the larger vision of utilizing hydrogen to address a range of issues for SCL and the Port. These include, but are not limited to, large scale fueling of medium duty (MD) and heavy duty (HD) vehicles to reduce emissions, support of adjacent light duty (LD) vehicles, support of critical port operation during extreme events (e.g. resiliency), mitigation of capital intensive electrical distribution system upgrades to support evolving port operations, a market resource that can be used by SCL to generate revenue, and support of planned future maritime operations that currently require heavy use of shipboard hydrogen. While the broader use of hydrogen by SCL and the Port is not the focus of this CRADA, how its use impacts the design of the initial vehicle fueling station, and the modular approach, are the central work. As such, this project will develop the framework and process for developing a modular station design that considers the broader range of coordinate uses of hydrogen as an energy resource.

08 HYDROGEN↗
DOE OSTI2016

Evaluation of no-MST operations in the SRS ARP for Hanford LAWPS

The Savannah River Site (SRS) Actinide Removal Process has been processing salt waste since 2008. This process includes a filtration step in the 512-S facility. Initial operations included the addition, or strike, of monosodium titanate (MST) to remove soluble actinides and strontium. The added MST and any entrained sludge solids were then separated from the supernate by cross flow filtration. During this time, the filter operations have, on many occasions, been the bottleneck process limiting the rate of salt processing. Recently, 512-S- has started operations utilizing “No-MST” where the MST actinide removal strike was not performed and the supernate was simply pre-filtered prior to Cs removal processing. Direct filtration of decanted tank supernate, as demonstrated in 512-S, is the proposed method of operation for the Hanford Low Activity Waste Pretreatment System (LAWPS) facility. Processing decanted supernate without MST solids has been demonstrated for cross flow filtration to provide a significant improvement in production with the SRS Salt Batches 8 and 9 feed chemistries. The average filtration rate for the first 512-S batch processing cycle using No-MST has increased filtrate production by over 35% of the historical average. The increase was sustained for more than double the amount of filtrate batches processed before cleaning of the filter was necessary. While there are differences in the design of the 512-S and Hanford filter systems, the 512-S system should provide a reasonable indication of LAWPS filter performance with similar feed properties. Based on the data from the 512-S facility and with favorable feed properties, the LAWPS filter, as currently sized at over twice the size of the 512-S filter (532 square feet filtration area versus 235 square feet), has the potential to provide sustained filtrate production at the upper range of the planned LAWPS production rate of 17 gpm.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2012

Fast DRR generation for 2D to 3D registration on GPUs

Purpose: The generation of digitally reconstructed radiographs (DRRs) is the most time consuming step on the CPU in intensity based two-dimensional x-ray to three-dimensional (CT or 3D rotational x-ray) medical image registration, which has application in several image guided interventions. This work presents optimized DRR rendering on graphical processor units (GPUs) and compares performance achievable on four commercially available devices. Methods: A ray-cast based DRR rendering was implemented for a 512 Multiplication-Sign 512 Multiplication-Sign 72 CT volume. The block size parameter was optimized for four different GPUs for a region of interest (ROI) of 400 Multiplication-Sign 225 pixels with different sampling ratios (1.1%-9.1% and 100%). Performance was statistically evaluated and compared for the four GPUs. The method and the block size dependence were validated on the latest GPU for several parameter settings with a public gold standard dataset (512 Multiplication-Sign 512 Multiplication-Sign 825 CT) for registration purposes. Results: Depending on the GPU, the full ROI is rendered in 2.7-5.2 ms. If sampling ratio of 1.1%-9.1% is applied, execution time is in the range of 0.3-7.3 ms. On all GPUs, the mean of the execution time increased linearly with respect to the number of pixels if sampling was used. Conclusions: The presented results outperform other results from the literature. This indicates that automatic 2D to 3D registration, which typically requires a couple of hundred DRR renderings to converge, can be performed quasi on-line, in less than a second or depending on the application and hardware in less than a couple of seconds. Accordingly, a whole new field of applications is opened for image guided interventions, where the registration is continuously performed to match the real-time x-ray.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2016

Optical CT imaging of solid radiochromic dosimeters in mismatched refractive index solutions using a scanning laser and large area detector

Purpose: The practical use of the PRESAGE® solid plastic dosimeter is limited by the inconvenience of immersing it in high-viscosity oils to achieve refractive index matching for optical computed tomography (CT) scanning. The oils are slow to mix and difficult to clean from surfaces, and the dosimeter rotation can generate dynamic Schlieren inhomogeneity patterns in the reference liquid, limiting the rotational and overall scan speed. Therefore, it would be beneficial if lower-viscosity, water-based solutions with slightly unmatched refractive index could be used instead. The purpose of this work is to demonstrate the feasibility of allowing mismatched conditions when using a scanning laser system with a large acceptance angle detector. A fiducial-based ray path measurement technique is combined with an iterative CT reconstruction algorithm to reconstruct images. Methods: A water based surrounding liquid with a low viscosity was selected for imaging PRESAGE® solid dosimeters. Liquid selection was optimized to achieve as high a refractive index as possible while avoiding rotation-induced Schlieren effects. This led to a refractive index mismatch of 6% between liquid and dosimeters. Optical CT scans were performed with a fan-beam scanning-laser optical CT system with a large area detector to capture most of the refracted rays. A fiducial marker placed on the wall of a cylindrical sample occludes a given light ray twice. With knowledge of the rotation angle and the radius of the cylindrical object, the actual internal path of each ray through the dosimeter can be calculated. Scans were performed with 1024 projections of 512 data samples each, and rays were rebinned to form 512 parallel-beam projections. Reconstructions were performed on a 512 × 512 grid using 100 iterations of the SIRT iterative CT algorithm. Proof of concept was demonstrated with a uniformly attenuating solution phantom. PRESAGE® dosimeters (11 cm diameter) were irradiated with Cobalt-60 irradiator to achieve either a uniform dose or a 2-level “step-dose” pattern. Results: With 6% refractive index mismatching, a circular field of view of 85% of the diameter of a cylindrical sample can be reconstructed accurately. Reconstructed images of the test solution phantom were uniform (within 3%) inside this radius. However, the dose responses of the PRESAGE® samples were not spatially uniform, with variations of at least 5% in sensitivity. The variation appears as a “cupping” artifact with less sensitivity in the middle than at the periphery of the PRESAGE® cylinder. Polarization effects were also detected for these samples. Conclusions: The fiducial-based ray path measurement scheme, coupled with an iterative reconstruction algorithm, enabled optical CT scanning of PRESAGE® dosimeters immersed in mismatched refractive index solutions. However, improvements to PRESAGE® dose response uniformity are required.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

Sample variance in N-body simulations and impact on tomographic shear predictions

We study the effects of sample variance in N-body simulations as a function of the size of the simulation box, namely in connection with predictions on tomographic shear spectra. We make use of a set of eight ΛCDM simulations in boxes of 128, 256, and 512 h{sup −1} Mpc a side, for a total of 24, differing just by the initial seeds. Among the simulations with 128 and 512 h{sup −1} Mpc a side, we suitably select those closest and farthest from average. Numerical and linear spectra P(k,z) are suitably connected at low k to evaluate the effects of sample variance on shear spectra C{sub ij}(ℓ) for 5 or 10 tomographic bands. We find that shear spectra obtained by using 128 h{sup −1} Mpc simulations can vary up to ∼25%, just because of the seed. The sample variance lowers to ∼3.3% when using 512 h{sup −1} Mpc. These percentages could however slightly vary if other sets of the same number of realizations were considered. Accordingly, in order to match the ∼1% precision expected for the data, if still using eight boxes, we require a size of ∼1300–1700 h{sup −1} Mpc for them.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

The eleventh and twelfth data release of the Sloan Digital Sky Survey: Final data from SDSS-III

The third generation of the Sloan Digital Sky Survey (SDSS-III) took data from 2008 to 2014 using the original SDSS wide-field imager, the original and an upgraded multi-object fiber-fed optical spectrograph, a new near-infrared high-resolution spectrograph, and a novel optical interferometer. All of the data from SDSS-III are now made public. In particular, this paper describes Data Release 11 (DR11) including all data acquired through 2013 July, and Data Release 12 (DR12) adding data acquired through 2014 July (including all data included in previous data releases), marking the end of SDSS-III observing. Relative to our previous public release (DR10), DR12 adds one million new spectra of galaxies and quasars from the Baryon Oscillation Spectroscopic Survey (BOSS) over an additional 3000 deg 2 of sky, more than triples the number of H-band spectra of stars as part of the Apache Point Observatory (APO) Galactic Evolution Experiment (APOGEE), and includes repeated accurate radial velocity measurements of 5500 stars from the Multi-object APO Radial Velocity Exoplanet Large-area Survey (MARVELS). The APOGEE outputs now include the measured abundances of 15 different elements for each star. In total, SDSS-III added 5200 deg 2 of ugriz imaging; 155,520 spectra of 138,099 stars as part of the Sloan Exploration of Galactic Understanding and Evolution 2 (SEGUE-2) survey; 2,497,484 BOSS spectra of 1,372,737 galaxies, 294,512 quasars, and 247,216 stars over 9376 deg 2 ; 618,080 APOGEE spectra of 156,593 stars; and 197,040 MARVELS spectra of 5513 stars. Since its first light in 1998, SDSS has imaged over 1/3 of the Celestial sphere in five bands and obtained over five million astronomical spectra.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

SOLIDS PRECIPITATION EVENT IN MCU CAUSAL ANALYSIS AND RECOMMENDATIONS FROM SOLIDS RECOVERY TEAM

A process upset occurred in the Modular Caustic-Side Solvent Extraction Unit (MCU) facility on April 6th, 2014. During recovery efforts, a significant amount of solids were found in the Salt Solution Feed Tank (SSFT), Salt Solution Receipt Tanks (SSRTs), two extraction contactors, and scrub contactors. The solids were identified by Savannah River National Laboratory (SRNL) as primarily sodium oxalate and sodium alumina silicate (NAS) with the presence of some aluminum hydroxide. NAS solids have been present in the SSFT since simulant runs during cold chemical startup of MCU in 2007, and have not hindered operations since that time. During the process upset in April 2014, the oxalate solids partially blocked the aqueous outlet of the extraction contactors, causing salt solution to exit through the contactor organic outlet to the scrub contactors with the organic phase. This salt solution overwhelmed the scrub contactors and passed with the organic phase to the strip section of MCU. The partially reversed flow of salt solution resulted in a Strip Effluent (SE) stream that was high in Isopar™ L, pH and sodium. The primary cause of the excessive solids accumulation in the SSRTs and SSFT at MCU is attributed to an increase in the frequency of oxalic acid cleaning of the 512-S primary filter. Agitation in the SSRTs at MCU in response to cold weather likely provided the primary mechanism to transfer the solids to the contactors. Sources of the sodium oxalate solids are attributed to the oxalic acid cleaning solution used to clean the primary filter at the Actinide Removal Process (ARP) filtration at 512-S, as well as precipitation from the salt batch feed, which is at or near oxalate saturation. The Solids Recovery Team was formed to determine the cause of the solids formation and develop recommendations to prevent or mitigate this event in the future. A total of 53 recommendations were generated. These recommendations were organized into 4 focus areas: • Improve understanding of oxalate equilibrium and kinetics in salt solutions • Reduction/elimination of oxalic acid cleaning in 512-S • Flowsheet optimization • Improving diagnostic capability The recommendations implemented prior to resumption of MCU operations provide a risk mitigation or detection function through additional sampling and observation. The longer term recommendations provide a framework to increase the basic process knowledge of both oxalate chemistry and filtration behavior and then facilitate decisions that improve the salt flowsheet as a system.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2012

Effect of twinning on texture evolution of depleted uranium using a viscoplastic self-consistent model

Ductility and fracture toughness is a major stumbling block in using depleted uranium as a structural material. The ability to correctly model deformation of uranium can be used to create process path methods to improve its structural design ability. The textural evolution of depleted uranium was simulated using a visco-plastic self consistent model and analyzed by comparing pole figures of the simulations and experimental samples. Depleted uranium has the same structure as alpha uranium, which is an orthorhombic phase of uranium. Both deformation slip and twin systems were compared. The VPSC model was chosen to simulate this material because the model encompasses both low-symmetry materials as well as twinning in materials. This is of particular interest since depleted uranium has a high propensity for twinning, which dominates deformation and texture evolution. Simulated results were compared to experimental results to measure the validity of the model. One specific twin system, the {l_brace}176{r_brace}[512] twin, was of specific notice. The VPSC model was used to simulate the influence of this twin on depleted uranium and was compared with a mechanically shocked depleted uranium sample. Under high strain rate shock deformation conditions, the {l_brace}176{r_brace}[512] twin system appears to be a dominant deformation system. By simulating a compression process using the VPSC model with the {l_brace}176{r_brace}[512] twin as the dominant deformation mode, a favorable comparison could be made between the experimental and simulated textures. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Report on the AAPM grand challenge on deep generative modeling for learning medical image statistics

Abstract Background The findings of the 2023 AAPM Grand Challenge on Deep Generative Modeling for Learning Medical Image Statistics are reported in this Special Report. Purpose The goal of this challenge was to promote the development of deep generative models for medical imaging and to emphasize the need for their domain‐relevant assessments via the analysis of relevant image statistics. Methods As part of this Grand Challenge, a common training dataset and an evaluation procedure was developed for benchmarking deep generative models for medical image synthesis. To create the training dataset, an established 3D virtual breast phantom was adapted. The resulting dataset comprised about 108 000 images of size 512 512. For the evaluation of submissions to the Challenge, an ensemble of 10 000 DGM‐generated images from each submission was employed. The evaluation procedure consisted of two stages. In the first stage, a preliminary check for memorization and image quality (via the Fréchet Inception Distance [FID]) was performed. Submissions that passed the first stage were then evaluated for the reproducibility of image statistics corresponding to several feature families including texture, morphology, image moments, fractal statistics, and skeleton statistics. A summary measure in this feature space was employed to rank the submissions. Additional analyses of submissions was performed to assess DGM performance specific to individual feature families, the four classes in the training data, and also to identify various artifacts. Results Fifty‐eight submissions from 12 unique users were received for this Challenge. Out of these 12 submissions, 9 submissions passed the first stage of evaluation and were eligible for ranking. The top‐ranked submission employed a conditional latent diffusion model, whereas the joint runners‐up employed a generative adversarial network, followed by another network for image superresolution. In general, we observed that the overall ranking of the top 9 submissions according to our evaluation method (i) did not match the FID‐based ranking, and (ii) differed with respect to individual feature families. Another important finding from our additional analyses was that different DGMs demonstrated similar kinds of artifacts. Conclusions This Grand Challenge highlighted the need for domain‐specific evaluation to further DGM design as well as deployment. It also demonstrated that the specification of a DGM may differ depending on its intended use.

Radiology, Nuclear Medicine & Medical Imaging↗
DOE OSTI2017

Fast, Simple Calcium Imaging Segmentation with Fully Convolutional Networks

Calcium imaging is a technique for observing neuron activity as a series of images showing indicator fluorescence over time. Manually segmenting neurons is time-consuming, leading to research on automated calcium imaging segmentation (ACIS). We evaluated several deep learning models for ACIS on the Neurofinder competition datasets and report our best model: U-Net2DS, a fully convolutional network that operates on 2D mean summary images. U-Net2DS requires minimal domain-specific pre/post-processing and parameter adjustment, and predictions are made on full 512×512 images at ≈9K images per minute. It ranks third in the Neurofinder competition (F1=0.57) and is the best model to exclusively use deep learning. We also demonstrate useful segmentations on data from outside the competition. The model’s simplicity, speed, and quality results make it a practical choice for ACIS and a strong baseline for more complex models in the future.

Klibisz, Alex↗