Search NASA⌕ Search

SEARCH · Search NASA

Results for “model inference”

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 145 records · Page 8

The clustering of the SDSS-IV extended Baryon Oscillation Spectroscopic Survey DR14 quasar sample: structure growth rate measurement from the anisotropic quasar power spectrum in the redshift range 0.8 < z < 2.2

We analyse the clustering of the Sloan Digital Sky Survey IV extended Baryon Oscillation Spectroscopic Survey Data Release 14 quasar sample (DR14Q).We measure the redshift space distortions using the power-spectrum monopole, quadrupole, and hexadecapole inferred from 148 659 quasars between redshifts 0.8 and 2.2, covering a total sky footprint of 2112.9 deg 2 . We constrain the logarithmic growth of structure times the amplitude of dark matter density fluctuations, fσ 8 , and the Alcock–Paczynski dilation scales that allow constraints to be placed on the angular diameter distance D A (z) and the Hubble H(z) parameter. At the effective redshift of z eff = 1.52, fσ 8 (z eff ) = 0.420 ± 0.076, H(z eff ) = [162 ± 12] (r$fid\atop{s}$ /r s ) kms -1 Mpc -1 , and DA(z eff ) = [1.85 ± 0.11] × 10 3 (r s/ r$fid\atop{s}$) Mpc, where r s is the comoving sound horizon at the baryon drag epoch and the superscript ‘fid’ stands for its fiducial value. The errors take into account the full error budget, including systematics and statistical contributions. These results are in full agreement with the current Λ-Cold Dark Matter cosmological model inferred from Planck measurements. Finally, we compare our measurements with other eBOSS companion papers and find excellent agreement, demonstrating the consistency and complementarity of the different methods used for analysing the data.

79 ASTRONOMY AND ASTROPHYSICS↗

Generic inference of inflation models by non-Gaussianity and primordial power spectrum reconstruction

We present a generic inference method for inflation models from observational data by the usage of higher-order statistics of the curvature perturbation on uniform density hypersurfaces. This method is based on the calculation of the posterior for the primordial non-Gaussianity parameters f{sub NL} and g{sub NL}, which in general depend on specific parameters of inflation and reheating models, and enables to discriminate among the still viable inflation models. To keep analyticity as far as possible to dispense with numerically expensive sampling techniques a saddle-point approximation is introduced, whose precision is validated for a numerical toy example. The mathematical formulation is done in a generic way so that the approach remains applicable to cosmic microwave background data as well as to large scale structure data. Additionally, we review a few currently interesting inflation models and present numerical toy examples thereof in two and three dimensions to demonstrate the efficiency of the higher-order statistics method. A second quantity of interest is the primordial power spectrum. Here, we present two Bayesian methods to infer it from observational data, the so called critical filter and an extension thereof with smoothness prior, both allowing for a non-parametric spectrum reconstruction. These methods are able to reconstruct the spectra of the observed perturbations and the primordial ones of curvature perturbation even in case of non-Gaussianity and partial sky coverage. We argue that observables like T- and B-modes permit to measure both spectra. This also allows to infer the level of non-Gaussianity generated since inflation.

79 ASTRONOMY AND ASTROPHYSICS↗

PHYSICAL PROPERTIES OF YOUNG BROWN DWARFS AND VERY LOW MASS STARS INFERRED FROM HIGH-RESOLUTION MODEL SPECTRA

By comparing near-infrared spectra with atmospheric models, we infer the effective temperature, surface gravity, projected rotational velocity, and radial velocity for 21 very low mass stars and brown dwarfs. The unique sample consists of two sequences in spectral type from M6-M9, one of 5-10 Myr objects and one of >1 Gyr field objects. A third sequence is comprised of only {approx}M6 objects with ages ranging from <1 Myr to >1 Gyr. Spectra were obtained in the J band at medium (R {approx} 2000) and high (R {approx} 20,000) resolutions with NIRSPEC on the Keck II telescope. Synthetic spectra were generated from atmospheric structures calculated with the PHOENIX model atmosphere code. Using multi-dimensional least-squares fitting and Monte Carlo routines we determine the best-fit model parameters for each observed spectrum and note which spectral regions provide consistent results. We identify successes in the reproduction of observed features by atmospheric models, including pressure-broadened K I lines, and investigate deficiencies in the models, particularly missing FeH opacity, that will need to be addressed in order to extend our analysis to cooler objects. The precision that can be obtained for each parameter using medium- and high-resolution near-infrared spectra is estimated and the implications for future studies of very low mass stars and brown dwarfs are discussed.

79 ASTRONOMY AND ASTROPHYSICS↗

RAM‐SCB simulations of electron transport and plasma wave scattering during the October 2012 “double‐dip” storm

Abstract Mechanisms for electron injection, trapping, and loss in the near‐Earth space environment are investigated during the October 2012 “double‐dip” storm using our ring current‐atmosphere interactions model with self‐consistent magnetic field (RAM‐SCB). Pitch angle and energy scattering are included for the first time in RAM‐SCB using L and magnetic local time (MLT)‐dependent event‐specific chorus wave models inferred from NOAA Polar‐orbiting Operational Environmental Satellites (POES) and Van Allen Probes Electric and Magnetic Field Instrument Suite and Integrated Science observations. The dynamics of the source (approximately tens of keV) and seed (approximately hundreds of keV) populations of the radiation belts simulated with RAM‐SCB is compared with Van Allen Probes Magnetic Electron Ion Spectrometer observations in the morning sector and with measurements from NOAA 15 satellite in the predawn and afternoon MLT sectors. We find that although the low‐energy ( E < 100 keV) electron fluxes are in good agreement with observations, increasing significantly by magnetospheric convection during both SYM‐H dips while decreasing during the intermediate recovery phase, the injection of high‐energy electrons is underestimated by this mechanism throughout the storm. Local acceleration by chorus waves intensifies the electron fluxes at E ≥50 keV considerably, and RAM‐SCB simulations overestimate the observed trapped fluxes by more than an order of magnitude; the precipitating fluxes simulated with RAM‐SCB are weaker, and their temporal and spatial evolutions agree well with POES/Medium Energy Proton and Electron Detectors data.

58 GEOSCIENCES↗

Impact of California’s air pollution laws on black carbon and their implications for direct radiative forcing

We examine the temporal and the spatial trends in the concentrations of black carbon (BC) - recorded by the IMPROVE monitoring network for the past 20 years - in California. Annual average BC concentrations in California have decreased by about 50% from 0.46 11g m-3 in 1989 to 0.24 11 gm-3 in 2008 compared to the corresponding reductions in diesel BC emissions (also about 50%) from a peak of 0.013 Tg Yr -1 in 1990 to 0.006 Tg Tr -1 by 2008. We attribute the observed negative trends to the reduction in vehicular emissions due to stringent statewide regulations. Our conclusion that the reduction in diesel emissions is a primary cause of the observed BC reduction is also substantiated by a significant decrease in the ratio of BC to non-BC aerosols. The absorption efficiency of aerosols at visible wavelengths - determined from the observed scattering coefficient and the observed BC - also decreased by about 50% leading to a model-inferred negative direct radiative forcing (a cooling effect) of -1.4 w m-2 (±60%) over California.

54 ENVIRONMENTAL SCIENCES↗

Mapping Glacier Basal Sliding Applying Machine Learning

During the RESOLVE project (“High-resolution imaging in subsurface geophysics: development of a multi-instrument platform for interdisciplinary research”), continuous surface displacement and seismic array observations were obtained on Glacier d’Argentière in the French Alps for 35 days in May 2018. The data set is used to perform a detailed study of targeted processes within the highly dynamic cryospheric environment. In particular, the physical processes controlling glacial basal motion are poorly understood and remain challenging to observe directly. Especially in the Alpine region for temperate based glaciers where the ice rapidly responds to changing climatic conditions and thus, processes are strongly intermittent in time and heterogeneous in space. Spatially dense seismic and Global Positioning System (GPS) measurements are analyzed applying machine learning to gain insight into the processes controlling glacial motions of Glacier d’Argentière. Using multiple bandpass-filtered copies of the continuous seismic waveforms, we compute energy-based features, develop a matched field beamforming catalog and include meteorological observations. Features describing the data are analyzed with a gradient boosting decision tree model to directly estimate the GPS displacements from the seismic noise. We posit that features of the seismic noise provide direct access to the dominant parameters that drive displacement on the highly variable and unsteady surface of the glacier. The machine learning model infers daily fluctuations and longer term trends. The results show on-ice displacement rates are strongly modulated by activity at the base of the glacier. The techniques presented provide a new approach to study glacial basal sliding and discover its full complexity.

58 GEOSCIENCES↗

Chronic gamma radiation resistance in fungi correlates with resistance to chromium and elevated temperatures, but not with resistance to acute irradiation

Abstract Exposure to chronic ionizing radiation (CIR) from nuclear power plant accidents, acts of terrorism, and space exploration poses serious threats to humans. Fungi are a group of highly radiation-resistant eukaryotes, and an understanding of fungal CIR resistance mechanisms holds the prospect of protecting humans. We compared the abilities of 95 wild-type yeast and dimorphic fungal isolates, representing diverse Ascomycota and Basidiomycota , to resist exposure to five environmentally-relevant stressors: CIR (long-duration growth under 36 Gy/h) and acute (10 kGy/h) ionizing radiation (IR), heavy metals (chromium, mercury), elevated temperature (up to 50 °C), and low pH (2.3). To quantify associations between resistances to CIR and these other stressors, we used correlation analysis, logistic regression with multi-model inference, and customized machine learning. The results suggest that resistance to acute IR in fungi is not strongly correlated with the ability of a given fungal isolate to grow under CIR. Instead, the strongest predictors of CIR resistance in fungi were resistance to chromium (III) and to elevated temperature. These results suggest fundamental differences between the mechanisms of resistance to chronic and acute radiation. Convergent evolution towards radioresistance among genetically distinct groups of organisms is considered here.

Shuryak, Igor↗

Study of heat sources interacting in integrated circuits by laser mirage effect

This work exploits the mirage effect to analyze multiple heat sources thermally interacting in an integrated circuit (IC) by means of a probe IR laser beam, which strikes on the die lateral walls and passes through the die substrate. Under such conditions, the criteria for locating such hot spots, as well as their relative power dissipation, are discussed on the basis of a theoretical model inferred in this work. Finally, the technique feasibility is shown in a real application scenario, obtaining 5-μm spatial lateral resolution and an error in power dissipation measurements below 5%. This method may become a practical alternative to usual off-chip techniques for inspecting hot spots in ICs and to experimentally characterize heat flow in the semiconductor substrate.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Thermonuclear neutron emission from a sheared-flow stabilized Z-pinch

We report the fusion Z-pinch experiment (FuZE) is a sheared-flow stabilized Z-pinch designed to study the effects of flow stabilization on deuterium plasmas with densities and temperatures high enough to drive nuclear fusion reactions. Results from FuZE show high pinch currents and neutron emission durations thousands of times longer than instability growth times. While these results are consistent with thermonuclear neutron emission, energetically resolved neutron measurements are a stronger constraint on the origin of the fusion production. This stems from the strong anisotropy in energy created in beam-target fusion, compared to the relatively isotropic emission in thermonuclear fusion. In dense Z-pinch plasmas, a potential and undesirable cause of beam-target fusion reactions is the presence of fast-growing, “sausage” instabilities. This work introduces a new method for characterizing beam instabilities by recording individual neutron interactions in plastic scintillator detectors positioned at two different angles around the device chamber. Histograms of the pulse-integral spectra from the two locations are compared using detailed Monte Carlo simulations. These models infer the deuteron beam energy based on differences in the measured neutron spectra at the two angles, thereby discriminating beam-target from thermonuclear production. An analysis of neutron emission profiles from FuZE precludes the presence of deuteron beams with energies greater than 4.65 keV with a statistical uncertainty of 4.15 keV and a systematic uncertainty of 0.53 keV. This analysis demonstrates that axial, beam-target fusion reactions are not the dominant source of neutron emission from FuZE. These data are promising for scaling FuZE up to fusion reactor conditions.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Tracing the cosmic metal evolution in the low-redshift intergalactic medium

Using the Cosmic Origins Spectrograph aboard the Hubble Space Telescope, we measured the abundances of six ions (C III, C IV, Si III, Si IV, N V, and O VI) in the low-redshift (z ≤ 0.4) intergalactic medium (IGM). Both C IV and Si IV have increased in abundance by a factor of ∼10 from z ≈ 5.5 to the present. We derive ion mass densities, ρ{sub ion} ≡ Ω{sub ion}ρ{sub cr}, with Ω{sub ion} expressed relative to the closure density. Our models of mass-abundance ratios, (Si III/Si IV) =0.67{sub −0.19}{sup +0.35}, (C III/C IV) =0.70{sub −0.20}{sup +0.43}, and (Ω{sub C} {sub III}+Ω{sub C} {sub IV})/(Ω{sub Si} {sub III}+Ω{sub Si} {sub IV})=4.9{sub −1.1}{sup +2.2}, are consistent with the photoionization parameter log U = –1.5 ± 0.4, hydrogen photoionization rate Γ{sub H} = (8 ± 2) × 10{sup –14} s{sup –1} at z < 0.4, and specific intensity I {sub 0} = (3 ± 1) × 10{sup –23} erg cm{sup –2} s{sup –1} Hz{sup –1} sr{sup –1} at the Lyman limit. Consistent ionization corrections for C and Si are scaled to an ionizing photon flux Φ{sub 0} = 10{sup 4} cm{sup –2} s{sup –1}, baryon overdensity Δ {sub b} ≈ 200 ± 50, and ''alpha-enhancement'' (Si/C enhanced to three times its solar ratio). We compare these metal abundances to the expected IGM enrichment and abundances in higher photoionized states of carbon (C V) and silicon (Si V, Si VI, and Si VII). Our ionization modeling infers IGM metal densities of (5.4 ± 0.5) × 10{sup 5} M {sub ☉} Mpc{sup –3} in the photoionized Lyα forest traced by the C and Si ions and (9.1 ± 0.6) × 10{sup 5} M {sub ☉} Mpc{sup –3} in hotter gas traced by O VI. Combining both phases, the heavy elements in the IGM have mass density ρ {sub Z} = (1.5 ± 0.8) × 10{sup 6} M {sub ☉} Mpc{sup –3} or Ω {sub Z} ≈ 10{sup –5}. This represents 10% ± 5% of the metals produced by (6 ± 2) × 10{sup 8} M {sub ☉} Mpc{sup –3} of integrated star formation with yield y{sub m} = 0.025 ± 0.010. The missing metals at low redshift may reside within galaxies and in undetected ionized gas in galactic halos and circumgalactic medium.

79 ASTRONOMY AND ASTROPHYSICS↗

Predicting resistive wall mode stability in NSTX through balanced random forests and counterfactual explanations

Abstract Recent progress in the disruption event characterization and forecasting framework has shown that machine learning guided by physics theory can be easily implemented as a supporting tool for fast computations of ideal stability properties of spherical tokamak plasmas. In order to extend that idea, a customized random forest (RF) classifier that takes into account imbalances in the training data is hereby employed to predict resistive wall mode (RWM) stability for a set of high beta discharges from the NSTX spherical tokamak. More specifically, with this approach each tree in the forest is trained on samples that are balanced via a user-defined over/under-sampler. The proposed approach outperforms classical cost-sensitive methods for the problem at hand, in particular when used in conjunction with a random under-sampler, while also resulting in a threefold reduction in the training time. In order to further understand the model’s decisions, a diverse set of counterfactual explanations based on determinantal point processes (DPP) is generated and evaluated. Via the use of DPP, the underlying RF model infers that the presence of hypothetical magnetohydrodynamic activity would have prevented the RWM from concurrently going unstable, which is a counterfactual that is indeed expected by prior physics knowledge. Given that this result emerges from the data-driven RF classifier and the use of counterfactuals without hand-crafted embedding of prior physics intuition, it motivates the usage of counterfactuals to simulate real-time control by generating the β N levels that would have kept the RWM stable for a set of unstable discharges.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

wavess 1.2: presenting an HLA-aware within-host virus sequence simulation framework

Motivation Understanding how virus sequences are shaped by selection can inform vaccine design and transmission inference. Modeling within-host evolution to interrogate these questions requires a detailed mechanistic framework that accurately captures sequence diversification. The CD8 + cytotoxic T-lymphocyte (CTL) response plays an important role in immune-mediated selection and can leave strong signatures in virus sequences; however, existing sequence-based within-host virus modeling frameworks do not explicitly include a human leukocyte antigen (HLA)-aware CTL response. Results We extended our previously published within-host sequence evolution simulator, wavess, to include an explicit CTL response, and share a method for identifying HLA-specific CTL epitopes given a founder virus sequence. We also updated the model to permit a variable recombination rate, which allows for modeling non-adjacent genes, segmented genomes, and recombination hotspots. These extensions to wavess allow for more accurate simulation of viruses and virus genes, particularly in regions of the genome where the immune response is dominated by CTLs (rather than antibodies). It also provides the foundation for investigations of how these newly-added biological mechanisms influence within-host evolution. Availability and implementation The core of wavess is written in Python 3, with helper functions written in R. It is available at https://github.com/MolEvolEpid/wavess.

60 APPLIED LIFE SCIENCES↗

Graph neural network for neutrino physics event reconstruction

Liquid argon time projection chamber (LArTPC) detector technology offers a wealth of high-resolution information on particle interactions, and leveraging that information to its full potential requires sophisticated automated reconstruction techniques. Here, this article describes NUGRAPH 2, a graph neural network for low-level reconstruction of simulated neutrino interactions in a LArTPC detector. Simulated neutrino interactions in the MicroBooNE detector geometry are described as heterogeneous graphs, with energy depositions on each detector plane forming nodes on planar subgraphs. The network utilizes a multihead attention message-passing mechanism to perform background filtering and semantic labeling on these graph nodes, identifying those associated with the primary physics interaction with 98.0% efficiency and labeling them according to particle type with 94.9% efficiency. The network operates directly on detector observables across multiple two-dimensional representations but utilizes a three-dimensional-context-aware mechanism to encourage consistency between these representations. Model inference takes 0.12 s / event on a CPU and 0.005 s / event batched on a GPU. This architecture is designed to be a general-purpose solution for particle reconstruction in neutrino physics, with the potential for deployment across a broad range of detector technologies, and offers a core convolution engine that can be leveraged for a variety of tasks beyond the two described in this paper.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

BraggNN : fast X-ray Bragg peak analysis using deep learning

X-ray diffraction based microscopy techniques such as high-energy diffraction microscopy (HEDM) rely on knowledge of the position of diffraction peaks with high precision. These positions are typically computed by fitting the observed intensities in detector data to a theoretical peak shape such as pseudo-Voigt. As experiments become more complex and detector technologies evolve, the computational cost of such peak-shape fitting becomes the biggest hurdle to the rapid analysis required for real-time feedback in experiments. To this end, we propose BraggNN, a deep-learning based method that can determine peak positions much more rapidly than conventional pseudo-Voigt peak fitting. When applied to a test dataset, peak center-of-mass positions obtained from BraggNN deviate less than 0.29 and 0.57 pixels for 75 and 95% of the peaks, respectively, from positions obtained using conventional pseudo-Voigt fitting (Euclidean distance). When applied to a real experimental dataset and using grain positions from near-field HEDM reconstruction as ground-truth, grain positions using BraggNN result in 15% smaller errors compared with those calculated using pseudo-Voigt. Recent advances in deep-learning method implementations and special-purpose model inference accelerators allow BraggNN to deliver enormous performance improvements relative to the conventional method, running, for example, more than 200 times faster on a consumer-class GPU card with out-of-the-box software.

36 MATERIALS SCIENCE↗

A 3D Implementation of Convolutional Neural Network for Fast Inference

Low latency inference has many applications in edge machine learning. In this paper, we present a run-time configurable convolutional neural network (CNN) inference ASIC design for low-latency edge machine learning. By implementing a 5-stage pipelined CNN inference model in a 3D ASIC technology, we demonstrate that the model distributed on two dies utilizing face-to-face (F2F) 3D integration achieves superior performance. Our experimental results show that the design based on 3D integration achieves 43% better energy-delay product when compared to the traditional 2D technology.

Miniskar, Narasinga Rao↗

Increasing Rubisco as a simple means to enhance photosynthesis and productivity now without lowering nitrogen use efficiency

Summary Global demand for food may rise by 60% mid‐century. A central challenge is to meet this need using less land in a changing climate. Nearly all crop carbon is assimilated through Rubisco, which is catalytically slow, reactive with oxygen, and a major component of leaf nitrogen. Developing more efficient forms of Rubisco, or engineering CO 2 concentrating mechanisms into C 3 crops to competitively repress oxygenation, are major endeavors, which could hugely increase photosynthetic productivity (≥ 60%). New technologies are bringing this closer, but improvements remain in the discovery phase and have not been reduced to practice. A simpler shorter‐term strategy that could fill this time gap, but with smaller productivity increases (c. 10%) is to increase leaf Rubisco content. This has been demonstrated in initial field trials, improving the productivity of C 3 and C 4 crops. Combining three‐dimensional leaf canopies with metabolic models infers that a 20% increase in Rubisco increases canopy photosynthesis by 14% in sugarcane (C 4 ) and 9% in soybean (C 3 ). This is consistent with observed productivity increases in rice, maize, sorghum and sugarcane. Upregulation of Rubisco is calculated not to require more nitrogen per unit yield and although achieved transgenically to date, might be achieved using gene editing to produce transgene‐free gain of function mutations or using breeding.

Plant Sciences↗

Two-stage atlas subset selection in multi-atlas based image segmentation

Purpose: Fast growing access to large databases and cloud stored data presents a unique opportunity for multi-atlas based image segmentation and also presents challenges in heterogeneous atlas quality and computation burden. This work aims to develop a novel two-stage method tailored to the special needs in the face of large atlas collection with varied quality, so that high-accuracy segmentation can be achieved with low computational cost. Methods: An atlas subset selection scheme is proposed to substitute a significant portion of the computationally expensive full-fledged registration in the conventional scheme with a low-cost alternative. More specifically, the authors introduce a two-stage atlas subset selection method. In the first stage, an augmented subset is obtained based on a low-cost registration configuration and a preliminary relevance metric; in the second stage, the subset is further narrowed down to a fusion set of desired size, based on full-fledged registration and a refined relevance metric. An inference model is developed to characterize the relationship between the preliminary and refined relevance metrics, and a proper augmented subset size is derived to ensure that the desired atlases survive the preliminary selection with high probability. Results: The performance of the proposed scheme has been assessed with cross validation based on two clinical datasets consisting of manually segmented prostate and brain magnetic resonance images, respectively. The proposed scheme demonstrates comparable end-to-end segmentation performance as the conventional single-stage selection method, but with significant computation reduction. Compared with the alternative computation reduction method, their scheme improves the mean and medium Dice similarity coefficient value from (0.74, 0.78) to (0.83, 0.85) and from (0.82, 0.84) to (0.95, 0.95) for prostate and corpus callosum segmentation, respectively, with statistical significance. Conclusions: The authors have developed a novel two-stage atlas subset selection scheme for multi-atlas based segmentation. It achieves good segmentation accuracy with significantly reduced computation cost, making it a suitable configuration in the presence of extensive heterogeneous atlases.

60 APPLIED LIFE SCIENCES↗

SU-E-J-128: Two-Stage Atlas Selection in Multi-Atlas-Based Image Segmentation

Purpose: In the new era of big data, multi-atlas-based image segmentation is challenged by heterogeneous atlas quality and high computation burden from extensive atlas collection, demanding efficient identification of the most relevant atlases. This study aims to develop a two-stage atlas selection scheme to achieve computational economy with performance guarantee. Methods: We develop a low-cost fusion set selection scheme by introducing a preliminary selection to trim full atlas collection into an augmented subset, alleviating the need for extensive full-fledged registrations. More specifically, fusion set selection is performed in two successive steps: preliminary selection and refinement. An augmented subset is first roughly selected from the whole atlas collection with a simple registration scheme and the corresponding preliminary relevance metric; the augmented subset is further refined into the desired fusion set size, using full-fledged registration and the associated relevance metric. The main novelty of this work is the introduction of an inference model to relate the preliminary and refined relevance metrics, based on which the augmented subset size is rigorously derived to ensure the desired atlases survive the preliminary selection with high probability. Results: The performance and complexity of the proposed two-stage atlas selection method were assessed using a collection of 30 prostate MR images. It achieved comparable segmentation accuracy as the conventional one-stage method with full-fledged registration, but significantly reduced computation time to 1/3 (from 30.82 to 11.04 min per segmentation). Compared with alternative one-stage cost-saving approach, the proposed scheme yielded superior performance with mean and medium DSC of (0.83, 0.85) compared to (0.74, 0.78). Conclusion: This work has developed a model-guided two-stage atlas selection scheme to achieve significant cost reduction while guaranteeing high segmentation accuracy. The benefit in both complexity and performance is expected to be most pronounced with large-scale heterogeneous data.

60 APPLIED LIFE SCIENCES↗