Search NASA⌕ Search

SEARCH · Search NASA

Results for “maximum likelihood”

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 685 records · Page 38

Minimal spanning tree algorithm for γ-ray source detection in sparse photon images: cluster parameters and selection strategies

We present that the minimal spanning tree (MST) algorithm is a graph-theoretical cluster-finding method. We previously applied it to γ-ray bidimensional images, showing that it is quite sensitive in finding faint sources. Possible sources are associated with the regions where the photon arrival directions clusterize. MST selects clusters starting from a particular “tree” connecting all the point of the image and performing a cut based on the angular distance between photons, with a number of events higher than a given threshold. In this paper, we show how a further filtering, based on some parameters linked to the cluster properties, can be applied to reduce spurious detections. We find that the most efficient parameter for this secondary selection is the magnitudeM of a cluster, defined as the product of its number of events by its clustering degree. We test the sensitivity of the method by means of simulated and real Fermi-Large Area Telescope (LAT) fields. Our results show that √M is strongly correlated with other statistical significance parameters, derived from a wavelet based algorithm and maximum likelihood (ML) analysis, and that it can be used as a good estimator of statistical significance of MST detections. Finally, we apply the method to a 2-year LAT image at energies higher than 3 GeV, and we show the presence of new clusters, likely associated with BL Lac objects.

79 ASTRONOMY AND ASTROPHYSICS↗

A Hybrid Semi-supervised Classification Scheme for Mining Multisource Geospatial Data

Supervised learning methods such as Maximum Likelihood (ML) are often used in land cover (thematic) classification of remote sensing imagery. ML classifier relies exclusively on spectral characteristics of thematic classes whose statistical distributions (class conditional probability densities) are often overlapping. The spectral response distributions of thematic classes are dependent on many factors including elevation, soil types, and ecological zones. A second problem with statistical classifiers is the requirement of large number of accurate training samples (10 to 30 |dimensions|), which are often costly and time consuming to acquire over large geographic regions. With the increasing availability of geospatial databases, it is possible to exploit the knowledge derived from these ancillary datasets to improve classification accuracies even when the class distributions are highly overlapping. Likewise newer semi-supervised techniques can be adopted to improve the parameter estimates of statistical model by utilizing a large number of easily available unlabeled training samples. Unfortunately there is no convenient multivariate statistical model that can be employed for mulitsource geospatial databases. In this paper we present a hybrid semi-supervised learning algorithm that effectively exploits freely available unlabeled training samples from multispectral remote sensing images and also incorporates ancillary geospatial databases. We have conducted several experiments on real datasets, and our new hybrid approach shows over 25 to 35% improvement in overall classification accuracy over conventional classification schemes.

54 ENVIRONMENTAL SCIENCES↗

Measuring key Sm isotope ratios in irradiated UO 2 for use in plutonium discrimination nuclear forensics

As part of an experimental validation of a nuclear forensics methodology for Pu source reactor-type discrimination, destructive analysis has been performed here on two irradiated UO 2 pellets with different irradiation histories. Analysis has focused on measuring key Sm fission product isotope ratios used in a previously published maximum likelihood methodology to determine the most likely irradiation history of the pellets. A total of 21 Sm isotope ratios were measured within the irradiated pellets, and generally agreed within 20% of the irradiations as simulated using the Monte Carlo Radiation Transport and material depletion code, MCNP6. Results indicate the chosen approach can accurately measure the isotope ratios within 5% experimental error.

07 ISOTOPE AND RADIATION SOURCES↗

Scalable computations for nonstationary Gaussian processes

Nonstationary Gaussian process models can capture complex spatially varying dependence structures in spatial datasets. However, the large number of observations in modern datasets makes fitting such models computationally intractable with conventional dense linear algebra. In addition, derivative-free or even first-order optimization methods can be very slow to converge when estimating many spatially varying parameters. In this paper, we present a computational framework which couples an algebraic block diagonal plus low-rank covariance matrix approximation with stochastic trace estimation to facilitate the efficient use of second-order solvers for maximum likelihood estimation of Gaussian process models with many parameters. We demonstrate the effectiveness of these methods by simultaneously fitting 192 parameters in the popular nonstationary model of Paciorek and Schervish using 107,600 sea surface temperature anomaly measurements.

97 MATHEMATICS AND COMPUTING↗

Selection of a phylogenetically informative region of the norovirus genome for outbreak linkage

The recognition of a common source norovirus outbreak is supported by finding identical norovirus sequences in patients. Norovirus sequencing has been established in many (national) public health laboratories and academic centers, but often partial and different genome sequences are used. Therefore, agreement on a target sequence of sufficient diversity to resolve links between outbreaks is crucial. Although harmonization of laboratory methods is one of the keystone activities of networks that have the aim to identify common source norovirus outbreaks, this has proven difficult to accomplish, particularly in the international context. Here, we aimed at providing a method enabling identification of the genomic region informative of a common source norovirus outbreak by bio-informatic tools. The data set of 502 unique full length capsid gene sequences available from the public domain, combined with epidemiological data including linkage information was used to build over 3,000 maximum likelihood (ML) trees for different sequence lengths and regions. All ML trees were evaluated for robustness and specificity of clustering of known linked norovirus outbreaks against the background diversity of strains. Great differences were seen in the robustness of commonly used PCR targets for cluster detection. The capsid gene region spanning nucleotides 900–1,400 was identified as the region optimally substituting for the full length capsid region. Reliability of this approach depends on the quality of the background data set, and we recommend periodic reassessment of this growing data set. The approach may be applicable to multiple sequence-based data sets of other pathogens.

59 BASIC BIOLOGICAL SCIENCES↗

Learning generative neural networks with physics knowledge

Deep generative neural networks have enabled modeling complex distributions, but incorporating physics knowledge into the neural networks is still challenging and is at the core of current physics-based machine learning research. To this end, we propose a physics generative neural network (PhysGNN), a new class of generative neural networks for learning unknown distributions in a physical system described by partial differential equations (PDE). PhysGNN couples PDE systems with generative neural networks. It is a fully differentiable model that allows back-propagation of gradients through both numerical PDE solvers and generative neural networks, and is trained by minimizing the discrete Wasserstein distance between generated and observed probability distributions of the PDE outputs using the stochastic gradient descent method. Moreover, PhysGNN does not require adversarial training like standard generative neural networks, which offers better stability than adversarial training. We show that PhysGNN can learn complex distributions in stochastic inverse problems, where conventional methods such as maximum likelihood estimation and momentum matching methods may be inapplicable when little knowledge is known about the form of unknown distributions or the physical model is too complex. Furthermore, our method allows physics-based generative neural network training for learning complex distributions in the context of differential equations.

97 MATHEMATICS AND COMPUTING↗

Sequential Monte Carlo sampling in hidden Markov models of nonlinear dynamical systems

We investigate the issue of which state functionals can have their uncertainty estimated efficiently in dynamical systems with uncertainty. Because of the high dimensionality and complexity of the problem, sequential Monte Carlo (SMC) methods are used. We investigate SMC methods where the proposal distribution is computed by maximum likelihood or by a linearization approach. We prove that the variance of the SMC method is bounded linearly in the number of time steps when the proposal distribution is truncated normal distribution. We also show that for a moderate large number of steps the error produced by approximation of dynamical systems linearly accumulates on the condition that the logarithm of the density function of noise is Lipschitz continuous. This finding is significant because the uncertainty in many dynamical systems, in particular, in chemical engineering systems, can be assumed to have this nature. We demonstrate our findings for a simple test case from chemical engineering. The theoretical findings provide a foundation for the parallel software SISTOS.

97 MATHEMATICS AND COMPUTING↗

DeepCMB: Lensing reconstruction of the cosmic microwave background with deep neural networks

Next-generation cosmic microwave background (CMB) experiments will have lower noise and therefore increased sensitivity, enabling improved constraints on fundamental physics parameters such as the sum of neutrino masses and the tensor-to-scalar ratio r. Achieving competitive constraints on these parameters requires high signal-to-noise extraction of the projected gravitational potential from the CMB maps. Standard methods for reconstructing the lensing potential employ the quadratic estimator (QE). However, the QE performs suboptimally at the low noise levels expected in upcoming experiments. Other methods, like maximum likelihood estimators (MLE), are under active development. In this work, we demonstrate reconstruction of the CMB lensing potential with deep convolutional neural networks (CNN) - ie, a ResUNet. The network is trained and tested on simulated data, and otherwise has no physical parametrization related to the physical processes of the CMB and gravitational lensing. We show that, over a wide range of angular scales, ResUNets recover the input gravitational potential with a higher signal-to-noise ratio than the QE method, reaching levels comparable to analytic approximations of MLE methods. We demonstrate that the network outputs quantifiably different lensing maps when given input CMB maps generated with different cosmologies. We also show we can use the reconstructed lensing map for cosmological parameter estimation. This application of CNN provides a few innovations at the intersection of cosmology and machine learning. First, while training and regressing on images, we predict a continuous-variable field rather than discrete classes. Second, we are able to establish uncertainty measures for the network output that are analogous to standard methods. We expect this approach to excel in capturing hard-to-model non-Gaussian astrophysical foreground and noise contributions.

79 ASTRONOMY AND ASTROPHYSICS↗

Monte Carlo studies of medium-size telescope designs for the Cherenkov Telescope Array

In this paper, we present studies for optimizing the next generation of ground-based imaging atmospheric Cherenkov telescopes (IACTs). Results focus on mid-sized telescopes (MSTs) for CTA, detecting very high energy gamma rays in the energy range from a few hundred GeV to a few tens of TeV. We describe a novel, flexible detector Monte Carlo package, FAST (FAst Simulation for imaging air cherenkov Telescopes), that we use to simulate different array and telescope designs. The simulation is somewhat simplified to allow for efficient exploration over a large telescope design parameter space. We investigate a wide range of telescope performance parameters including optical resolution, camera pixel size, and light collection area. In order to ensure a comparison of the arrays at their maximum sensitivity, we analyze the simulations with the most sensitive techniques used in the field, such as maximum likelihood template reconstruction and boosted decision trees for background rejection. Choosing telescope design parameters representative of the proposed Davies–Cotton (DC) and Schwarzchild–Couder (SC) MST designs, we compare the performance of the arrays by examining the gamma-ray angular resolution and differential point-source sensitivity. We further investigate the array performance under a wide range of conditions, determining the impact of the number of telescopes, telescope separation, night sky background, and geomagnetic field. We find a 30–40% improvement in the gamma-ray angular resolution at all energies when comparing arrays with an equal number of SC and DC telescopes, significantly enhancing point-source sensitivity in the MST energy range. Finally, we attribute the increase in point-source sensitivity to the improved optical point-spread function and smaller pixel size of the SC telescope design.

42 ENGINEERING↗

Error handling strategies in multiphase inverse modeling

Parameter estimation by inverse modeling involves the repeated evaluation of a function of residuals. These residuals represent both errors in the model and errors in the data. In practical applications of inverse modeling of multiphase flow and transport, the error structure of the final residuals often significantly deviates from the statistical assumptions that underlie standard maximum likelihood estimation using the least-squares method. Large random or systematic errors are likely to lead to convergence problems, biased parameter estimates, misleading uncertainty measures, or poor predictive capabilities of the calibrated model. The multiphase inverse modeling code iTOUGH2 supports strategies that identify and mitigate the impact of systematic or non-normal error structures. We discuss these approaches and provide an overview of the error handling features implemented in iTOUGH2.

54 ENVIRONMENTAL SCIENCES↗

Solving inverse problems in stochastic models using deep neural networks and adversarial training

Inverse problems associated with stochastic models constitute a significant portion of scientific and engineering applications. In such cases the unknown quantities are distributions. The applicability of traditional methods is limited because of their demanding assumptions or prohibitive computational consumption; for example, maximum likelihood methods require closed-form density functions, and Markov Chain Monte Carlo needs a large number of simulations. We propose a new method that estimates the unknown distribution by matching the statistical properties between observed and simulated random processes. We leverage the expressive power of neural networks to approximate the unknown distribution and use a discriminative neural network for computing the statistical discrepancies between the observed and simulated random processes. Here we demonstrated numerically that the proposed methods can estimate both the model parameters and learn complicated unknown distributions.

42 ENGINEERING↗

Latent map Gaussian processes for mixed variable metamodeling

Gaussian processes (GPs) are ubiquitously used in sciences and engineering as metamodels. Standard GPs, however, can only handle numerical or quantitative variables. Here we introduce latent map Gaussian processes (LMGPs) that inherit the attractive properties of GPs and are also applicable to mixed data which have both quantitative and qualitative inputs. The core idea behind LMGPs is to learn a continuous, low-dimensional latent space or manifold which encodes all qualitative inputs. To learn this manifold, we first assign a unique prior vector representation to each combination of qualitative inputs. We then use a low-rank linear map to project these priors on a manifold that characterizes the posterior representations. As the posteriors are quantitative, they can be directly used in any standard correlation function such as the Gaussian or Matern. Hence, the optimal map and the corresponding manifold, along with other hyperparameters of the correlation function, can be systematically learned via maximum likelihood estimation. Through a wide range of analytic and real-world examples, we demonstrate the advantages of LMGPs over state-of-the-art methods in terms of accuracy and versatility. In particular, we show that LMGPs can handle variable-length inputs, have an explainable neural network interpretation, and provide insights into how qualitative inputs affect the response or interact with each other. We also employ LMGPs in Bayesian optimization and illustrate that they can discover optimal compound compositions more efficiently than conventional methods that convert compositions to qualitative variables via manual featurization.

42 ENGINEERING↗

Insight into a highly polymorphic endophyte isolated from the roots of the halophytic seepweed Suaeda salsa: Laburnicola rhizohalophila sp. nov. (Didymosphaeriaceae, Pleosporales)

In this work, we surveyed root endophytic fungi of the coastal halophyte Suaeda salsa and detected a population of a novel species that we described here as Laburnicola rhizohalophila sp. nov. No sexual sporulating structure was observed. Instead, it produced a large amount of thalloconidia, 0–1 transverse septa, hyaline to darkly pigmented, often peanut-shaped and sometimes dumbbell-shaped, both ends enlarged with numerous oil droplets inside the hyphal cells. Surprisingly, a high degree of phenotypic and physiological intraspecific variation (e.g., salinity tolerance, growth under different carbon:nitrogen ratios, and carbon utilization pattern) was recorded. The inoculation test indicated that the isolates could successfully infect host roots and form microsclerotia-like structures in cortical cells, a typical trait of dark septate endophytes (DSEs). Furthermore, most isolates were shown to promote host seedling growth. To evaluate conspecificity and infer its phylogenetic affinity, multiloci data including nuclear rRNA loci (ITS1 and 2, partial 28S), partial RNA Polymerase II second-largest subunit ( rpb2 ), and partial translation elongation factor-1α ( tef1 ) were characterized. Genealogical concordance phylogenetic species recognition (GCPSR) detected a genetically isolated clade of L. rhizohalophila within the Pleosporales in the Didymosphaeriaceae. Maximum likelihood phylogenetic reconstruction revealed that the endophytic fungus was genetically close to L aburnicola dactylidis but separated by a relatively long genetic distance. Our work highlights that the pleosporalean taxa might represent an underexplored reservoir of root DSEs.

59 BASIC BIOLOGICAL SCIENCES↗

Elective Clinical Target Volumes for Conformal Therapy in Anorectal Cancer: A Radiation Therapy Oncology Group Consensus Panel Contouring Atlas

Purpose: To develop a Radiation Therapy Oncology Group (RTOG) atlas of the elective clinical target volume (CTV) definitions to be used for planning pelvic intensity-modulated radiotherapy (IMRT) for anal and rectal cancers. Methods and Materials: The Gastrointestinal Committee of the RTOG established a task group (the nine physician co-authors) to develop this atlas. They responded to a questionnaire concerning three elective CTVs (CTVA: internal iliac, presacral, and perirectal nodal regions for both anal and rectal case planning; CTVB: external iliac nodal region for anal case planning and for selected rectal cases; CTVC: inguinal nodal region for anal case planning and for select rectal cases), and to outline these areas on individual computed tomographic images. The imaging files were shared via the Advanced Technology Consortium. A program developed by one of the co-authors (I.E.N.) used binomial maximum-likelihood estimates to generate a 95% group consensus contour. The computer-estimated consensus contours were then reviewed by the group and modified to provide a final contouring consensus atlas. Results: The panel achieved consensus CTV definitions to be used as guidelines for the adjuvant therapy of rectal cancer and definitive therapy for anal cancer. The most important difference from similar atlases for gynecologic or genitourinary cancer is mesorectal coverage. Detailed target volume contouring guidelines and images are discussed. Conclusion: This report serves as a template for the definition of the elective CTVs to be used in IMRT planning for anal and rectal cancers, as part of prospective RTOG trials.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Role of Radiation Dose in the Risk of Secondary Leukemia After a Solid Tumor in Childhood Treated Between 1980 and 1999

Purpose: The purpose of this study was to estimate the risk of secondary leukemia as a function of radiation dose, taking into account heterogeneous radiation dose distribution. Methods and Materials: We analyzed a case-control study that investigated the risk of secondary leukemia and myelodysplasia after a solid tumor in childhood; it included 61 patients with leukemia matched with 196 controls. Complete clinical, chemotherapy, and radiotherapy histories were recorded for each patient in the study. Average radiation dose to each of seven bone marrow components for each patient was incorporated into the models, and corresponding risks were summed up. Conditional maximum likelihood methods were used to estimate risk parameters. Results: Whatever the model, we failed to evidence a role for the radiation dose to active bone marrow in the risk of later leukemia, myelodysplasia, or myeloproliferative syndrome, when adjusting for epipodophyllotoxin and anthracycline doses. This result was confirmed when fitting models that included total dose of radiation delivered during radiotherapy, when fitting models taking into account dose per fraction, and when restricting the analysis to acute myeloid leukemia. Conclusions: In contrast to results found in similar studies that included children treated before the use of epipodophyllotoxins, this study failed to show a role for radiotherapy in the risk of secondary leukemia after childhood cancer in children treated between 1980 and 1999. This discrepancy was probably due to a competitive mechanism between these two carcinogens.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Normal Tissue Complication Probability Estimation by the Lyman-Kutcher-Burman Method Does Not Accurately Predict Spinal Cord Tolerance to Stereotactic Radiosurgery

Purpose: To determine whether normal tissue complication probability (NTCP) analyses of the human spinal cord by use of the Lyman-Kutcher-Burman (LKB) model, supplemented by linear-quadratic modeling to account for the effect of fractionation, predict the risk of myelopathy from stereotactic radiosurgery (SRS). Methods and Materials: From November 2001 to July 2008, 24 spinal hemangioblastomas in 17 patients were treated with SRS. Of the tumors, 17 received 1 fraction with a median dose of 20 Gy (range, 18-30 Gy) and 7 received 20 to 25 Gy in 2 or 3 sessions, with cord maximum doses of 22.7 Gy (range, 17.8-30.9 Gy) and 22.0 Gy (range, 20.2-26.6 Gy), respectively. By use of conventional values for {alpha}/{beta}, volume parameter n, 50% complication probability dose TD{sub 50}, and inverse slope parameter m, a computationally simplified implementation of the LKB model was used to calculate the biologically equivalent uniform dose and NTCP for each treatment. Exploratory calculations were performed with alternate values of {alpha}/{beta} and n. Results: In this study 1 case (4%) of myelopathy occurred. The LKB model using radiobiological parameters from Emami and the logistic model with parameters from Schultheiss overestimated complication rates, predicting 13 complications (54%) and 18 complications (75%), respectively. An increase in the volume parameter (n), to assume greater parallel organization, improved the predictive value of the models. Maximum-likelihood LKB fitting of {alpha}/{beta} and n yielded better predictions (0.7 complications), with n = 0.023 and {alpha}/{beta} = 17.8 Gy. Conclusions: The spinal cord tolerance to the dosimetry of SRS is higher than predicted by the LKB model using any set of accepted parameters. Only a high {alpha}/{beta} value in the LKB model and only a large volume effect in the logistic model with Schultheiss data could explain the low number of complications observed. This finding emphasizes that radiobiological models traditionally used to estimate spinal cord NTCP may not apply to the dosimetry of SRS. Further research with additional NTCP models is needed.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Applying a Hypoxia-Incorporating TCP Model to Experimental Data on Rat Sarcoma

Purpose: To verify whether a tumor control probability (TCP) model which mechanistically incorporates acute and chronic hypoxia is able to describe animal in vivo dose-response data, exhibiting tumor reoxygenation. Methods and Materials: The investigated TCP model accounts for tumor repopulation, reoxygenation of chronic hypoxia, and fluctuating oxygenation of acute hypoxia. Using the maximum likelihood method, the model is fitted to Fischer-Moulder data on Wag/Rij rats, inoculated with rat rhabdomyosarcoma BA1112, and irradiated in vivo using different fractionation schemes. This data set is chosen because two of the experimental dose-response curves exhibit an inverse dose behavior, which is interpreted as due to reoxygenation. The tested TCP model is complex, and therefore, in vivo cell survival data on the same BA1112 cell line from Reinhold were added to Fischer-Moulder data and fitted simultaneously with a corresponding cell survival function. Results: The obtained fit to the combined Fischer-Moulder-Reinhold data was statistically acceptable. The best-fit values of the model parameters for which information exists were in the range of published values. The cell survival curves of well-oxygenated and hypoxic cells, computed using the best-fit values of the radiosensitivities and the initial number of clonogens, were in good agreement with the corresponding in vitro and in situ experiments of Reinhold. The best-fit values of most of the hypoxia-related parameters were used to recompute the TCP for non-small cell lung cancer patients as a function of the number of fractions, TCP(n). Conclusions: The investigated TCP model adequately describes animal in vivo data exhibiting tumor reoxygenation. The TCP(n) curve computed for non-small cell lung cancer patients with the best-fit values of most of the hypoxia-related parameters confirms previously obtained abrupt reduction in TCP for n < 10, thus warning against the adoption of severely hypofractionated schedules.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Assessing the Uncertainty in QUANTEC's Dose–Response Relation of Lung and Spinal Cord With a Bootstrap Analysis

Purpose: To apply a statistical bootstrap analysis to assess the uncertainty in the dose–response relation for the endpoints pneumonitis and myelopathy reported in the QUANTEC review. Methods and Materials: The bootstrap method assesses the uncertainty of the estimated population-based dose-response relation due to sample variability, which reflects the uncertainty due to limited numbers of patients in the studies. A large number of bootstrap replicates of the original incidence data were produced by random sampling with replacement. The analysis requires only the dose, the number of patients, and the number of occurrences of the studied endpoint, for each study. Two dose–response models, a Poisson-based model and the Lyman model, were fitted to each bootstrap replicate using maximum likelihood. Results: The bootstrap analysis generates a family of curves representing the range of plausible dose–response relations, and the 95% bootstrap confidence intervals give an estimated upper and lower toxicity risk. The curve families for the 2 dose–response models overlap for doses included in the studies at hand but diverge beyond that, with the Lyman model suggesting a steeper slope. The resulting distributions of the model parameters indicate correlation and non-Gaussian distribution. For both data sets, the likelihood of the observed data was higher for the Lyman model in >90% of the bootstrap replicates. Conclusions: The bootstrap method provides a statistical analysis of the uncertainty in the estimated dose–response relation for myelopathy and pneumonitis. It suggests likely values of model parameter values, their confidence intervals, and how they interrelate for each model. Finally, it can be used to evaluate to what extent data supports one model over another. For both data sets considered here, the Lyman model was preferred over the Poisson-based model.

62 RADIOLOGY AND NUCLEAR MEDICINE↗