Search NASA⌕ Search

SEARCH · Search NASA

Results for “Discriminant Analysis”

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 109 records · Page 6

Zone-based analysis for automated detection of abnormalities in chest radiographs

Purpose: The aim of this study was to develop an automated method for detection of local texture-based and density-based abnormalities in chest radiographs. Methods: The method was based on profile analysis to detect abnormalities in chest radiographs. In the method, one density-based feature, Density Symmetry Index, and two texture-based features, Roughness Maximum Index and Roughness Symmetry Index, were used to detect abnormalities in the lung fields. In each chest radiograph, the lung fields were divided into four zones initially and then the method was applied to each zone separately. For each zone, Density Symmetry Index was obtained from the projection profile of each zone, and Roughness Maximum Index and Roughness Symmetry Index were obtained by measuring the roughness of the horizontal profiles via moving average technique. Linear discriminant analysis was used to classify normal and abnormal cases based on the three indices. The discriminant performance of the method was evaluated using ROC analysis. Results: The method was evaluated on a database of 250 normal and 250 abnormal chest images. In the optimized conditions, the zone-based performance Az of the method for zones 1, 2, 3, and 4 were 0.917, 0.897, 0.892, and 0.814, respectively, and the case-based performance Az of the method was 0.842. Our previous method for detection of gross abnormalities was also evaluated on the same database. The case-based performance of our previous method was 0.689. Conclusions: In comparing the previous method and the new method proposed in this study, there was a great improvement by the new method for detection of local texture-based and density-based abnormalities. The new method combined with the previous one has potential for screening abnormalities in chest radiographs.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Synthesis route attribution of sulfur mustard by multivariate data analysis of chemical signatures

A multivariate model was developed here to attribute samples to a synthetic method used in the production of sulfur mustard (HD). Eleven synthetic methods were used to produce 66 samples for model construction. Three chemists working in both participating laboratories took part in the production, with the aim to introduce variability while reducing the influence of laboratory or chemist specific impurities in multivariate analysis. A gas chromatographic/mass spectrometric data set of peak areas for 103 compounds was subjected to orthogonal partial least squares - discriminant analysis to extract chemical attribution signature profiles and to construct multivariate models for classification of samples. For one- and two-step routes, model quality allowed the classification of an external test set (16/16 samples) according to synthesis conditions in the reaction yielding sulfur mustard. Classification of samples according to first-step methodology was considerably more difficult, given the high purity and uniform quality of the intermediate thiodiglycol produced in the study. Model performance in classification of aged samples was also investigated.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Elemental analysis of cotton by laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy (LIBS) has been applied to the elemental characterization of unprocessed cotton. This research is important in forensic and fraud detection applications to establish an elemental fingerprint of U.S. cotton by region, which can be used to determine the source of the cotton. To the best of our knowledge, this is the first report of a LIBS method for the elemental analysis of cotton. The experimental setup consists of a Nd:YAG laser that operates at the fundamental wavelength as the LIBS excitation source and an echelle spectrometer equipped with an intensified CCD camera. The relative concentrations of elements Al, Ba, Ca, Cr, Cu, Fe, Mg, and Sr from both nutrients and environmental contributions were determined by LIBS. Principal component analysis was used to visualize the differences between cotton samples based on the elemental composition by region in the U.S. Linear discriminant analysis of the LIBS data resulted in the correct classification of >97% of the cotton samples by U.S. region and >81% correct classification by state of origin.

74 ATOMIC AND MOLECULAR PHYSICS↗

Serum Proteome Profiles in Stricturing Crohn’s Disease: A pilot study.

Background: Crohn’s disease (CD) is a form of inflammatory bowel disease (IBD) with different described behaviors, including stricture. At present, there are no laboratory studies that can differentiate stricturing CD from other phenotypes of IBD. We performed a pilot study to examine differences in the proteome among patients with stricturing Crohn’s disease, non-stricturing Crohn’s disease, and ulcerative colitis (UC). Methods: Serum samples were selected from the Ocean State Crohn’s and Colitis Area Registry (OSCCAR), an established cohort of patients with IBD. Crohn’s disease patients with surgically-resected stricture were matched with similar patients with Crohn’s disease without known stricture, and with UC. Serum samples from each patient were digested and analyzed using liquid chromatography-mass spectrometry to characterize the proteome. Statistical analyses were performed to identify peptides and proteins that can differentiate CD with stricture. Results: Samples from 9 patients in each group (27 total patients) were analyzed. Baseline demographic characteristics were similar among the three groups. We quantified 7668 peptides and 897 proteins for analysis. ROC analysis identified a subset of peptides with an area under the curve greater than 0.9, indicating greater separation potential. Partial least squares discriminant analysis was able to distinguish among the three groups with up to 70% accuracy by peptides, and up to 80% accuracy by proteins. We identified the significantly different proteins and peptides, and determined their function based on previously published literature. Conclusions: The serum of patients with stricturing CD, non-stricturing CD, and UC are distinguishable via proteomic analysis. Some of the proteins that differentiate the stricturing phenotype have been implicated in complement activation, fibrinolytic pathways, and lymphocyte adhesion.

Environmental Molecular Sciences Laboratory↗

Comparative analysis of co-registered SIR-A, Seasat and Landsat images

A number of areas were covered by Seasat SAR and SIR-A, thus providing information at two different incidence angles (20 deg and 50 deg respectively). These angles correspond to scattering from two different surface roughness scales. The results of the analysis from two areas will be presented. The first is the San Rafael area in Utah where Seasat/SIR-A/Landsat data were coregistered and discriminant analysis is being applied to classify surface units. The second area is the Chotts area in Algeria where appreciable changes have been observed between Seasat and SIR-A images. However, it is not clear if the changes are due to the different observation geometry or changes in the surface between 1978 and 1981.

Elachi, C.↗

HELIOSEISMOLOGY OF PRE-EMERGING ACTIVE REGIONS. I. OVERVIEW, DATA, AND TARGET SELECTION CRITERIA

This first paper in a series describes the design of a study testing whether pre-appearance signatures of solar magnetic active regions were detectable using various tools of local helioseismology. The ultimate goal is to understand flux-emergence mechanisms by setting observational constraints on pre-appearance subsurface changes, for comparison with results from simulation efforts. This first paper provides details of the data selection and preparation of the samples, each containing over 100 members, of two populations: regions on the Sun that produced a numbered NOAA active region, and a 'control' sample of areas that did not. The seismology is performed on data from the GONG network; accompanying magnetic data from SOHO/MDI are used for co-temporal analysis of the surface magnetic field. Samples are drawn from 2001-2007, and each target is analyzed for 27.7 hr prior to an objectively determined time of emergence. The results of two analysis approaches are published separately: one based on averages of the seismology- and magnetic-derived signals over the samples, another based on Discriminant Analysis of these signals, for a statistical test of detectable differences between the two populations. We include here descriptions of a new potential-field calculation approach and the algorithm for matching sample distributions over multiple variables. We describe known sources of bias and the approaches used to mitigate them. We also describe unexpected bias sources uncovered during the course of the study and include a discussion of refinements that should be included in future work on this topic.

79 ASTRONOMY AND ASTROPHYSICS↗

Methods for presentation and display of multivariate data

Methods for the presentation and display of multivariate data are discussed with emphasis placed on the multivariate analysis of variance problems and the Hotelling T(2) solution in the two-sample case. The methods utilize the concepts of stepwise discrimination analysis and the computation of partial correlation coefficients.

Myers, R. H.↗

Nuclear Forensic Inferences Using Iterative Multidimensional Statistics

Nuclear forensics involves the analysis of interdicted nuclear material for specific material characteristics (referred to as 'signatures') that imply specific geographical locations, production processes, culprit intentions, etc. Predictive signatures rely on expert knowledge of physics, chemistry, and engineering to develop inferences from these material characteristics. Comparative signatures, on the other hand, rely on comparison of the material characteristics of the interdicted sample (the 'questioned sample' in FBI parlance) with those of a set of known samples. In the ideal case, the set of known samples would be a comprehensive nuclear forensics database, a database which does not currently exist. In fact, our ability to analyze interdicted samples and produce an extensive list of precise materials characteristics far exceeds our ability to interpret the results. Therefore, as we seek to develop the extensive databases necessary for nuclear forensics, we must also develop the methods necessary to produce the necessary inferences from comparison of our analytical results with these large, multidimensional sets of data. In the work reported here, we used a large, multidimensional dataset of results from quality control analyses of uranium ore concentrate (UOC, sometimes called 'yellowcake'). We have found that traditional multidimensional techniques, such as principal components analysis (PCA), are especially useful for understanding such datasets and drawing relevant conclusions. In particular, we have developed an iterative partial least squares-discriminant analysis (PLS-DA) procedure that has proven especially adept at identifying the production location of unknown UOC samples. By removing classes which fell far outside the initial decision boundary, and then rebuilding the PLS-DA model, we have consistently produced better and more definitive attributions than with a single pass classification approach. Performance of the iterative PLS-DA method compared favorably to that of classification and regression tree (CART) and k nearest neighbor (KNN) algorithms, with the best combination of accuracy and robustness, as tested by classifying samples measured independently in our laboratories against the vendor QC based reference set.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Chemical studies of H chondrites. 6: Antarctic/non-Antarctic compositional differences revisited

We report data for the trace elements Au, Co, Sb, Ga, Rb, Ag, Se, Cs, Te, Zn, Cd, Bi, T1, and In (ordered by putative volatility during nebular condensation and accretion) determined by radiochemical neutron activation analysis of 14 additional H5 and H6 chondrite falls. Data for the 10 most volatile elements (Rb to In) treated by the multivariate techniques of linear discriminant analysis and logistic regression in these and 44 other falls are compared with those of 59 H4-6 chondrites from Antarctica. Various populations are tested by the multivariate techniques, using the previously developed method of randomization-simulation to assess significance levels. An earlier conclusion, based on fewer examples, that H4-6 chondrite falls are compositionally distinguishable from the Antarctic suite is verified by the additional data. This distinctiveness is highly significant because of the presence of samples from Victoria Land in the Antarctic population, which differ compositionally from falls beyond any reasonable doubt. However, it cannot be proven unequivocally that falls and Antarctic samples from Queen Maud Land are compositionally distinguishable. Trivial causes (e.g., analyst bias, weathering) cannot explain the Victoria Land (Antarctic)/non-Antarctic compositional difference for paradigmatic H4-6 chondrites. This seems to reflect a time-dependent variation of near-Earth meteoroid source regions differing in average thermal history.

Wolf, Stephen F.↗

Remote Sensing as a Landscape Epidemiologic Tool to Identify Villages at High Risk for Malaria Transmission

A landscape approach using remote sensing and Geographic Information System (GIS) technologies was developed to discriminate between villages at high and low risk for malaria transmission, as defined by adult Anopheles albimanus abundance. Satellite data for an area in southern Chiapas, Mexico were digitally processed to generate a map of landscape elements. The GIS processes were used to determine the proportion of mapped landscape elements surrounding 40 villages where An. albimanus data had been collected. The relationships between vector abundance and landscape element proportions were investigated using stepwise discriminant analysis and stepwise linear regression. Both analyses indicated that the most important landscape elements in terms of explaining vector abundance were transitional swamp and unmanaged pasture. Discriminant functions generated for these two elements were able to correctly distinguish between villages with high ind low vector abundance, with an overall accuracy of 90%. Regression results found both transitional swamp and unmanaged pasture proportions to be predictive of vector abundance during the mid-to-late wet season. This approach, which integrates remotely sensed data and GIS capabilities to identify villages with high vector-human contact risk, provides a promising tool for malaria surveillance programs that depend on labor-intensive field techniques. This is particularly relevant in areas where the lack of accurate surveillance capabilities may result in no malaria control action when, in fact, directed action is necessary. In general, this landscape approach could be applied to other vector-borne diseases in areas where: 1. the landscape elements critical to vector survival are known and 2. these elements can be detected at remote sensing scales.

Beck, Louisa R.↗

A novel gene expression-based prognostic scoring system to predict survival in gastric cancer

Analysis of gene expression patterns in gastric cancer (GC) can help to identify a comprehensive panel of gene biomarkers for predicting clinical outcomes and to discover potential new therapeutic targets. Here, a multi-step bioinformatics analytic approach was developed to establish a novel prognostic scoring system for GC. We first identified 276 genes that were robustly differentially expressed between normal and GC tissues, of which, 249 were found to be significantly associated with overall survival (OS) by univariate Cox regression analysis. The biological functions of 249 genes are related to cell cycle, RNA/ncRNA process, acetylation and extracellular matrix organization. A network was generated for view of the gene expression architecture of 249 genes in 265 GCs. Finally, we applied a canonical discriminant analysis approach to identify a 53-gene signature and a prognostic scoring system was established based on a canonical discriminant function of 53 genes. The prognostic scores strongly predicted patients with GC to have either a poor or good OS. Our study raises the prospect that the practicality of GC patient prognosis can be assessed by this prognostic scoring system.

59 BASIC BIOLOGICAL SCIENCES↗

Part 1: Tracing Russian VX to its synthetic routes by multivariate statistics of chemical attribution signatures

Chemical attribution signatures (CAS) associated with different synthetic routes used for the production of Russian VX (VR) were identified here. The goal of the study was to retrospectively determine the production method employed for an unknown VR sample. Six different production methods were evaluated, carefully chosen to include established synthetic routes used in the past for large scale production of the agent, routes involving general phosphorus-sulfur chemistry pathways leading to the agent, and routes whose main characteristic is their innate simplicity in execution. Two laboratories worked in parallel and synthesized a total of 37 batches of VR via the six synthetic routes following predefined synthesis protocols. The chemical composition of impurities and byproducts in each route was analyzed by GC/MS-EI and 49 potential CAS were recognized as important markers in distinguishing these routes using Principal Component Analysis (PCA). The 49 potential CAS included expected species based on knowledge of reaction conditions and pathways but also several novel compounds that were fully identified and characterized by a combined analysis that included MS-CI, MS-EI and HR-MS. The CAS profiles of the calibration set were then analyzed using partial least squares discriminant analysis (PLS-DA) and a cross validated model was constructed. The model allowed the correct classification of an external test set without any misclassifications, demonstrating the utility of this methodology for attributing VR samples to a particular production method. This work is part one of a three-part series in this Forensic VSI issue of a Sweden-United States collaborative effort towards the understanding of the CAS of VR in diverse batches and matrices. This part focuses on the CAS in synthesized batches of crude VR and in the following two parts of the series the influence of food matrices on the CAS profiles are investigated.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Sampling-based Sublinear Low-rank Matrix Arithmetic Framework for Dequantizing Quantum Machine Learning

We present an algorithmic framework for quantum-inspired classical algorithms on close-to-low-rank matrices, generalizing the series of results started by Tang’s breakthrough quantum-inspired algorithm for recommendation systems [STOC’19]. Motivated by quantum linear algebra algorithms and the quantum singular value transformation (SVT) framework of Gilyén et al. [STOC’19], we develop classical algorithms for SVT that run in time independent of input dimension, under suitable quantum-inspired sampling assumptions. Our results give compelling evidence that in the corresponding QRAM data structure input model, quantum SVT does not yield exponential quantum speedups. Since the quantum SVT framework generalizes essentially all known techniques for quantum linear algebra, our results, combined with sampling lemmas from previous work, suffice to generalize all prior results about dequantizing quantum machine learning algorithms. In particular, our classical SVT framework recovers and often improves the dequantization results on recommendation systems, principal component analysis, supervised clustering, support vector machines, low-rank regression, and semidefinite program solving. We also give additional dequantization results on low-rank Hamiltonian simulation and discriminant analysis. Our improvements come from identifying the key feature of the quantum-inspired input model that is at the core of all prior quantum-inspired results: ℓ 2 -norm sampling can approximate matrix products in time independent of their dimension. We reduce all our main results to this fact, making our exposition concise, self-contained, and intuitive.

Computer Science↗

Landsat analysis of tropical forest succession employing a terrain model

Landsat multispectral scanner (MSS) data have yielded a dual classification of rain forest and shadow in an analysis of a semi-deciduous forest on Mindonoro Island, Philippines. Both a spatial terrain model, using a fifth side polynomial trend surface analysis for quantitatively estimating the general spatial variation in the data set, and a spectral terrain model, based on the MSS data, have been set up. A discriminant analysis, using both sets of data, has suggested that shadowing effects may be due primarily to local variations in the spectral regions and can therefore be compensated for through the decomposition of the spatial variation in both elevation and MSS data.

Barringer, T. H.↗

Elemental source attribution signatures for calcium ammonium nitrate (CAN) fertilizers used in homemade explosives

Calcium ammonium nitrate (CAN) is a widely available fertilizer composed of ammonium nitrate mixed with some form of calcium carbonate such as limestone or dolomite. CAN is also frequently used to make homemade explosives. The potential of using elemental profiling and chemometrics to match both pristine and reprocessed CAN fertilizers to their factories for use in future forensic investigations was examined. Inductively coupled plasma-mass spectrometry (ICP-MS) analysis was performed on 64 elements in 125 samples from 11 CAN stocks from 6 different CAN factories. Fisher ratio, degree-of-class-separation, and partial least squares discriminant analysis (PLSDA) were used to develop a model using the concentrations of Na, V, Mn, Cu, Ga, Sr, Ba and U to classify a validation set of CAN samples into 5 factory groups; one group was two factories from the same fertilizer company. In terms of the pristine CAN samples, i.e., unadulterated prills, 64% of the test samples were matched to their correct factory group with zero false positives. The same PLSDA model was used to correctly match 100% of the CAN samples that were reprocessed by crushing and mixing the CAN with powdered sugar. In the case of crushed CAN samples mixed with aluminum powder, correct matches were made for zero to 100% of the samples depending on the factory the CAN originated. Remarkably, for one factory, 100% of the ammonium nitrate samples that were extracted from CAN using tap or bottled water were matched to the correct CAN factory group. Lastly, the water-insoluble (calcium carbonate) portions of CAN provided a greater degree of discrimination between factories than the water-soluble portions of CAN. In summary, this work illustrates that sourcing unadulterated CAN fertilizer can potentially be done with high frequency and high confidence using elemental profiling and chemometrics while the sourcing of reprocessed CAN is dependent on how much an adulterant alters the recovered elemental profile of CAN.

chemical forensics↗

Association of Serum Bile Acid and Unsaturated Fatty Acid Profiles with the Risk of Diabetic Retinopathy in Type 2 Diabetic Patients

Aim: We aimed to identify the ability of serum bile acids (BAs) and unsaturated fatty acids (UFAs) profiles to predict the development of diabetic retinopathy (DR) in type 2 diabetes mellitus (T2DM) patients. Methods: We first used univariate and multivariate analysis to compare 15 serum BA and 11 UFA levels in healthy control (HC) group (n = 82), T2DM patients with DR (n = 58) and T2DM patients without DR (n = 60). Forty T2DM patients were considered for validation. Then, the receiver operating characteristic curve (ROC) and decision curve analysis were used to assess the diagnostic value and clinical benefit of serum biomarkers alone, clinical variables alone or in combination, and the area under the curve (AUC), integrated discrimination improvement (IDI), and net reclassification improvement (NRI) were used to further assess whether the addition of biomarkers significantly improved the predictive ability of the model. Results: Orthogonal partial least squares-discriminant analysis (OPLS-DA) of serum BAs and UFAs separated the three cohorts including HC, T2DM patients with or without DR. The difference in serum BA and UFA profiles of T2DM patients with or without DR was mainly manifested in the three metabolites of taurolithocholic acid (TLCA), tauroursodeoxycholic acid (TUDCA) and arachidonic acid (AA). Together, they had an AUC of 0.785 (0.918 for validation cohort) for predicting DR in T2DM patients. After adjusting for numerous confounding factors, TLCA, TUDCA, and AA were independent predictors that differentiated T2DM with or without DR. The results of AUC, IDI, and NRI demonstrated that adding these three biomarkers to a model with clinical variables statistically increased their predictive value and were replicated in our independent validation cohort. Conclusion: These findings highlight the association of three metabolites, TLCA, TUDCA and AA, with DR and may indicate their potential value in the pathogenesis of DR.

60 APPLIED LIFE SCIENCES↗

Multiparameter Methods are the New Field in Statistics

During the last decade statistical problems of a new kind have appeared more and more often, in which the dimensionality of observations is large, and the sample size is small or comparable with the dimensionality of observations. These problems may be characterized as multiparameter problems. The theory of multiparameter analysis, proposed by A.N.Kolmogorov and supported by the studies on the spectral theory of random matrices of V. A.Marchenko, L. A.Pastur, V. L.Girko, and others, revealed some specific phenomena appearing in statistics with a large number of weakly dependent variables. Particularly, there are stable relations between the principal parts of parameter set functions and the set of observed variables, which may be used to improve the statistical analysis. The use of these phenomena allows one to develop a thorough, always stable, and approximately optimal (irrespective of samples) versions of mostly used statistical procedures. The multiparameter versions of the discriminant analysis are of special interest.

97 MATHEMATICS AND COMPUTING↗