Application Of Pattern Recognition Methods Research Articles

Application of Pattern Recognition Methods to Study Spatial Localization of Polymetallic Mineralization in the Altai–Sayan Region

The Altai–Sayan mountain-folded belt is analyzed with the purpose of (1) revealing peculiarities of localization of large-scale polymetallic mineralization in the lineament-block structure of the region and (2) determining the geophysical and geomorphic peculiarities of the locations of these deposits using the Cora-3 pattern recognition algorithm. The lineament-block structure of the region is determined using morphostructural zoning. A spatial correlation between large and superlarge polymetallic deposits and morphostructural nodes is revealed. Based on this correlation, a dichotomy problem is solved, which is to divide the entire set of nodes in the region into two classes—ore-bearing and non-ore bearing. For this purpose, we used the Cora-3 logical recognition algorithm with training, for which the input data are geomorphological and geophysical parameters of the nodes. The training set of the algorithm was composed of the nodes where large and superlarge polymetal deposits are known. At the training stage, the algorithm identified the sets of the characteristic features that are peculiar to each class. Based on these features, all the nodes in the region were divided into ore-bearing and non-ore-bearing ones. As a result of recognition, the nodes in which deposits of the considered types and sizes are known were classified as ore-bearing, and, in addition to them, another 11 nodes were identified that meet the features determined in the work and can be considered potentially ore-bearing.

Identifying High-Risk Patterns in Single-Vehicle, Single-Occupant Road Traffic Accidents: A Novel Pattern Recognition Approach

Despite various interventions in road safety work, fatal and severe road traffic accidents (RTAs) remain a significant challenge, leading to human suffering and economic costs. Understanding the multicausal nature of RTAs, where multiple conditions and factors interact, is crucial for developing effective prevention measures in road safety work. This study investigates the multivariate statistical analysis of co-occurring conditions in RTAs, focusing on single-vehicle accidents with single occupancy and personal injury on Austrian roads outside built-up areas from 2012 to 2019. The aim is to detect recurring combinations of accident-related variables, referred to as blackpatterns (BPs), using the Austrian RTA database. This study proposes Fisher’s exact test to estimate the relationship between an accident-related variable and fatal and severe RTAs (severe casualties). In terms of pattern recognition, this study develops the maximum combination value (MCV) of accident-related variables, a procedure to search through all possible combinations of variables to find the one that has the highest frequency. The accident investigation proceeds with the application of pattern recognition methods, including binomial logistic regression and a newly developed method, the PATTERMAX method, created to accurately detect and analyse variable-specific BPs in RTA data. Findings indicate significant BPs contributing to severe accidents. The combination of binomial logistic regression and the PATTERMAX method appears to be a promising approach to investigate severe accidents, providing both insights into detailed variable combinations and their impact on accident severity.

Application of Pattern Recognition Methods to Study Spatial Localization of Polymetallic Mineralization in the Altai–Sayan Region

Application of Pattern Recognition Methods to Study Spatial Localization of Polymetallic Mineralization in the Altai–Sayan Region

An approach for assessment of concrete deterioration by surface waves

The aging of concrete infrastructure facilities is a global problem. External and internal harmful factors such as water ingress, freezing and thawing, fire, chemical attack and forces lead to concrete degradation, frequently progressing from its surface. It is important to know how deep the deterioration process is expanded from the surface into the depth of the concrete bulk and how much the concrete degraded. The purpose of the study was to demonstrate the feasibility of an approach based on the application of pattern recognition methods to the analysis of ultrasonic signals to assess two variable factors of interest - the material degradation degree and the thickness of the conditionally weak surface layer on the strong concrete underlay. A series of specimens was made according to the 2-factor grid experimental design in order to build a mathematical model for the evaluation. Ultrasonic signals were acquired in the surface transmission at 2 frequencies, 50 to 100 kHz by stepped profiling of the specimens’ surface. 2D spatiotemporal waveform matrices acquired by surface profiling, where the coordinates were the propagation distance and ultrasonic time, served as the material for processing. The approach based on the methods of pattern recognition included steps of transformation the signals by the discrete Fourier transform, calculation of statistical criteria and creation of decision rules as bilinear mathematical approximation surfaces of the criteria. Verification of the method by substituting data on a control series of specimens into the mathematical model built on the grid of specimens of a training series allowed a realistic determination of the both factors of interest for a specimen from the control series. However, the use of only formal statistical criteria was found not completely sufficient for reliable recognition in all control cases. Further improvement of recognition can apply additional decision rules based on wave propagation in the time domain and refining of the existing decision rules by implementation of weights of influence.

Highly specific fingerprinting of alkaline earth metal ions by a tunable plasmonic nanosensor array based on nanoaggregation of metallochromic dyes-AuNPs.

Metal ions, specifically alkaline earth metal ions (AEMIs; Mg2+, Ca2+, Sr2+, and Ba2+), have essential roles in industrial processes, medical testing, and environmental evaluation; therefore, developing sensitive detection methods capable of their contents is highly required. To this aim, we have designed an absorbance nanosensor array using three metallochromic dyes decorated on AuNPs and have monitored variations in AuNP plasmonic profiles upon the addition of AEMIs in different buffer and pH solutions. The array is designed in a tunable size of 2 × 3 × 1(2/3); as the type buffer and pH of solution are fixed, the number of dyes can be changed in three individual modes, three binary modes, and a ternary mode, respectively. Owing to the different binding affinities of AEMIs toward dyes in different buffer and pH solutions, fingerprint-like plasmonic profiles with different levels of aggregation AuNPs were generated for all modes of array. These aggregation AuNP-based fingerprint profiles in the wavelengths of 500, 525, 550, 575, 600, 625, 650, 675, 700, 725, and 750nm were used to discriminate the AEMIs by applying pattern recognition methods including linear discrimination analysis (LDA) and hierarchical clustering analysis (HCA) to identify each AEMI in the range 2.1-24.7μM. Accordingly, limits of detection (LODs) values of 0.013 (±3.13), 0.014 (±2.99), 0.020 (±4.17), and 0.017 (±4.31) μM were obtained the Mg2+, Ca2+, Sr2+, and Ba2+, respectively. The results revealed that all the modes of array could well differentiate complex mixtures of the AEMIs. Our suggested array also exhibited a good performance in the differentiation of AEMIs in real samples and a certified reference material (CRM) sample.

Particle tracking and radiation field characterization with Timepix3 in ATLAS

Four hybrid pixel detectors of Timepix3 technology, installed in the ATLAS experiment, were continuously taking data from April 2018 until the end of the Run-2 data taking period (December 2019). These detectors are synchronized with each other and the LHC orbit clock. They are capable of resolving the bunch structure of the LHC beams due to their time resolution of ∼2ns. Thus, they allow the characterization of the radiation field inside and outside bunch-crossing periods. This is shown for Timepix3 detectors at the extended barrel (x=-3.58 m, y=0.97 m, z=2.83 m). We apply pattern recognition methods to decompose the radiation field and determine the directionality of the minimum ionizing particles (MIP) component of the radiation field.

Identification of Catecholamine Neurotransmitters Using a Fluorescent Electronic Tongue.

Catecholamine neurotransmitters, specifically, dopamine (DA), epinephrine (EP), and norepinephrine (NE), are known as substantial indicators of various neurological diseases. Developing rapid detection methods capable of simultaneously screening their concentrations is highly desired for early clinical diagnosis of such diseases. To this aim, we have designed an optical sensor array using three fluorescent dyes with distinct emission bands and have monitored variations in their emission profiles upon the addition of DA, EP, and NE in the presence of gold ions. Because of the different reducing power of catecholamines, differently sized gold nanoparticles (GNPs) with different levels of aggregation were generated, resulting in different amounts of spectral overlap between the absorption band of the in situ generated plasmonic GNPs and the emission bands of the fluorescent dyes. These energy-transfer-based fingerprint profiles were used to discriminate the neurotransmitters by applying pattern recognition methods including linear discriminant analysis (LDA) and artificial neural networks (ANN) and to determine their concentration using multiple linear regression (MLR). Our proposed array also showed a good performance in the discrimination of DA, EP, and NE in complex biological media such as human urine.

Background correction method for improving the automated detection of radioisotopes from airborne gamma-ray surveys

An altitude-based background correction strategy was developed for use in the application of pattern recognition methods to the classification of gamma-ray spectra collected during airborne surveys. Application of this methodology helped to suppress the background spectral variation that serves to obscure the photopeaks associated with low levels of gamma-ray emission. The correction method was implemented by optimizing a database of background gamma-ray spectra collected at various locations and altitudes. Given this background database, a field-collected spectrum was corrected by performing linear regression onto a background spectrum from the database at a matching altitude. The residuals about the regression fit were then digitally filtered and submitted to nonparametric linear discriminant analysis for the purpose of computing classification models for targeted radioisotopes. The resulting classifiers were applied to predict the presence or absence of specific radioisotope signatures in data acquired during airborne surveys. Employing data provided by the U.S Environmental Protection Agency Airborne Spectral Photometric Environmental Collection Technology (ASPECT) program, classification models were computed to detect the presence of cesium-137 (137Cs) and cobalt-60 (60Co). The optimized classifiers were tested over 12 diverse locations, with nine of these data sets containing the target radioisotopes. Correct classification percentages of 99.4% and 99.8% were obtained for the 137Cs and 60Co classifiers, respectively, on the basis of comparisons to visual inspections of the corresponding spectra.

A comparison of various MRI feature types for characterizing whole brain anatomical differences using linear pattern recognition methods

There is a widespread interest in applying pattern recognition methods to anatomical neuroimaging data, but so far, there has been relatively little investigation into how best to derive image features in order to make the most accurate predictions. In this work, a Gaussian Process machine learning approach was used for predicting age, gender and body mass index (BMI) of subjects in the IXI dataset, as well as age, gender and diagnostic status using the ABIDE and COBRE datasets. MRI data were segmented and aligned using SPM12, and a variety of feature representations were derived from this preprocessing. We compared classification and regression accuracy using the different sorts of features, and with various degrees of spatial smoothing. Results suggested that feature sets that did not ignore the implicit background tissue class, tended to result in better overall performance, whereas some of the most commonly used feature sets performed relatively poorly.

Automated differentiation between meningioma and healthy brain tissue based on optical coherence tomography ex vivo images using texture features.

Brain tissue analysis is highly desired in neurosurgery, such as tumor resection. To guarantee best life quality afterward, exact navigation within the brain during the surgery is essential. So far, no method has been established that perfectly fulfills this need. Optical coherence tomography (OCT) is a promising three-dimensional imaging tool to support neurosurgical resections. We perform a preliminary study toward in vivo brain tumor removal assistance by investigating meningioma, healthy white, and healthy gray matter. For that purpose, we utilized a commercially available OCT device (Thorlabs Callisto) and measured eight samples of meningioma, three samples of healthy white, and two samples of healthy gray matter ex vivo directly after removal. Structural variations of different tissue types, especially meningioma, can already be seen in the raw OCT images. Nevertheless, an automated differentiation approach is desired, so that neurosurgical guidance can be delivered without a-priori knowledge of the surgeon. Therefore, we employ different algorithms to extract texture features and apply pattern recognition methods for their classification. With these postprocessing steps, an accuracy of nearly 98% was found.

Application of pattern recognition in detection of buried archaeological sites based on analysing environmental variables, Khorramabad Plain, West Iran

Archaeologists continue to search for techniques that enable them to analyze archaeological data efficiently with artificial intelligence approaches increasingly employed to create new knowledge from archaeological data. The purpose of this paper is to investigate the application of Pattern Recognition methods in detection of buried archaeological sites of the semi-arid Khorramabad Plain located in west Iran. This environment has provided suitable conditions for human habitation for over 40,000years. However, environmental changes in the late Pleistocene and Holocene have caused erosion and sedimentation resulting in burial of some archaeological sites making archaeological landscape reconstructions more challenging. In this paper, the environmental variables that have influenced formation of archaeological sites of the Khorramabad Plain are identified through the application of Arc GIS. These variables are utilized to create an accurate predictive model based on the application of One-Class classification Pattern Recognition techniques. These techniques can be built using data from one class only, when the data from other classes are difficult to obtain, and are highly suitable in this context. The experimental results of this paper confirm one-class classifiers, including Auto-encoder Neural Network, k-means, principal component analysis data descriptor, minimum spanning tree data descriptor, k-nearest neighbour and Gaussian distribution as promising applications in creating an effective model for detecting buried archaeological sites. Among the investigated classifiers, minimum spanning tree data descriptor achieved the best performance on the Khorramabad Plain data set.

Steps towards incorporating heterogeneities into program theory: A case study of a data-driven approach

This paper describes a framework that can help refine program theory through data explorations and stakeholder dialogue. The framework incorporates the following steps: a recognition that program implementation might need to be multi-phased for a number of interventions, the need to take stock of program theory, the application of pattern recognition methods to help identify heterogeneous program mechanisms, and stakeholder dialogue to refine the program. As part of the data exploration, a method known as developmental trajectories is implemented to learn about heterogeneous trajectories of outcomes in longitudinal evaluations. This method identifies trajectory clusters and also can estimate different treatment impacts for the various groups. The framework is highlighted with data collected in an evaluation of an alcohol risk-reduction program delivered in a college fraternity setting. The framework discussed in the paper is informed by a realist focus on “what works for whom under what contexts.” The utility of the framework in contributing to a dialogue on heterogeneous mechanism and subsequent implementation is described. The connection of the ideas in paper to a ‘learning through principled discovery’ approach is also described.

The adaptation of GDL motion recognition system to sport and rehabilitation techniques analysis.

The main novelty of this paper is presenting the adaptation of Gesture Description Language (GDL) methodology to sport and rehabilitation data analysis and classification. In this paper we showed that Lua language can be successfully used for adaptation of the GDL classifier to those tasks. The newly applied scripting language allows easily extension and integration of classifier with other software technologies and applications. The obtained execution speed allows using the methodology in the real-time motion capture data processing where capturing frequency differs from 100 Hz to even 500 Hz depending on number of features or classes to be calculated and recognized. Due to this fact the proposed methodology can be used to the high-end motion capture system. We anticipate that using novel, efficient and effective method will highly help both sport trainers and physiotherapist in they practice. The proposed approach can be directly applied to motion capture data kinematics analysis (evaluation of motion without regard to the forces that cause that motion). The ability to apply pattern recognition methods for GDL description can be utilized in virtual reality environment and used for sport training or rehabilitation treatment.

An integrated approach to the taxonomic identification of prehistoric shell ornaments.

Shell beads appear to have been one of the earliest examples of personal adornments. Marine shells identified far from the shore evidence long-distance transport and imply networks of exchange and negotiation. However, worked beads lose taxonomic clues to identification, and this may be compounded by taphonomic alteration. Consequently, the significance of this key early artefact may be underestimated. We report the use of bulk amino acid composition of the stable intra-crystalline proteins preserved in shell biominerals and the application of pattern recognition methods to a large dataset (777 samples) to demonstrate that taxonomic identification can be achieved at genus level. Amino acid analyses are fast (<2 hours per sample) and micro-destructive (sample size <2 mg). Their integration with non-destructive techniques provides a valuable and affordable tool, which can be used by archaeologists and museum curators to gain insight into early exploitation of natural resources by humans. Here we combine amino acid analyses, macro- and microstructural observations (by light microscopy and scanning electron microscopy) and Raman spectroscopy to try to identify the raw material used for beads discovered at the Early Bronze Age site of Great Cornard (UK). Our results show that at least two shell taxa were used and we hypothesise that these were sourced locally.

Application of pattern recognition methods to automatic identification of microscopic images of rocks registered under different polarization and lighting conditions

72 1024x768 The paper presents the research results regarding automatic classification of rock images, taken under an optical microscope in varying lighting conditions and with different polarization states. Classification was conducted with the use of 4 pattern recognition methods: nearest neighbor, K-nearest neighbors, nearest mode, and optimal spherical neighborhoods. The research was conducted on thin sections of 5 rocks. During research the CIELAB color space and a 9D feature space were used. The research results indicate that changing both lighting conditions along with polarization states results in worsening the classification outcome, although not substantially. The conducted research indicates that during automatic classification, of rocks photographed in varying lighting and polarization conditions, the highest percentage of correctly classified rocks apx 97% is given by the nearest neighbor method. The results also indicate the optimal spherical neighborhoods method is the safest method of those tested , which means that it returns the lowest number of incorrect classifications. Normal 0 21 false false false PL X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:Standardowy; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:Calibri,sans-serif; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:Times New Roman; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:Times New Roman; mso-bidi-theme-font:minor-bidi;}

Application of pattern recognition methods to solve optimization problems using imitation models

The summarizing lecture includes description of the approximate optimization method known as optimization-simulation.

Multivariate models of inter-subject anatomical variability

This paper presents a very selective review of some of the approaches for multivariate modelling of inter-subject variability among brain images. It focusses on applying probabilistic kernel-based pattern recognition approaches to pre-processed anatomical MRI, with the aim of most accurately modelling the difference between populations of subjects. Some of the principles underlying the pattern recognition approaches of Gaussian process classification and regression are briefly described, although the reader is advised to look elsewhere for full implementational details. Kernel pattern recognition methods require matrices that encode the degree of similarity between the images of each pair of subjects. This review focusses on similarity measures derived from the relative shapes of the subjects' brains. Pre-processing is viewed as generative modelling of anatomical variability, and there is a special emphasis on the diffeomorphic image registration framework, which provides a very parsimonious representation of relative shapes. Although the review is largely methodological, excessive mathematical notation is avoided as far as possible, as the paper attempts to convey a more intuitive understanding of various concepts. The paper should be of interest to readers wishing to apply pattern recognition methods to MRI data, with the aim of clinical diagnosis or biomarker development. It also tries to explain that the best models are those that most accurately predict, so similar approaches should also be relevant to basic science. Knowledge of some basic linear algebra and probability theory should make the review easier to follow, although it may still have something to offer to those readers whose mathematics may be more limited.

Applying pattern recognition methods plus quantum and physico-chemical molecular descriptors to analyze the anabolic activity of structurally diverse steroids

The great cost associated with the development of new anabolic-androgenic steroid (AASs) makes necessary the development of computational methods that shorten the drug discovery pipeline. Toward this end, quantum, and physicochemical molecular descriptors, plus linear discriminant analysis (LDA) were used to analyze the anabolic/androgenic activity of structurally diverse steroids and to discover novel AASs, as well as also to give a structural interpretation of their anabolic-androgenic ratio (AAR). The obtained models are able to correctly classify 91.67% (86.27%) of the AASs in the training (test) sets, respectively. The results of predictions on the 10% full-out cross-validation test also evidence the robustness of the obtained model. Moreover, these classification functions are applied to an "in house" library of chemicals, to find novel AASs. Two new AASs are synthesized and tested for in vivo activity. Although both AASs are less active than some commercially AASs, this result leaves a door open to a virtual variational study of the structure of the two compounds, to improve their biological activity. The LDA-assisted QSAR models presented here, could significantly reduce the number of synthesized and tested AASs, as well as could increase the chance of finding new chemical entities with higher AAR.

Pattern Recognition

Pattern recognition methods have been applied to a wide variety of chemical problems. In a typical pattern recognition study, samples are classified according to a specific property using measurements that are indirectly related to the property of interest. An empirical relationship or classification rule is developed from a set of samples for which the property of interest and the measurements are known. The classification rule can then be used to predict the property in samples that are not part of the original training set. In this review, the three major subdivisions of pattern recognition methodology are discussed and the analytical literature is surveyed. Much of the literature on pattern recognition focuses on novel and not so novel applications. Only the more interesting applications of pattern recognition methods are referenced in this review article.

Applying pattern recognition methods to analyze the molecular properties of a homologous series of nitrogen mustard agents

The purpose of this research was to analyze the pharmacological properties of a homologous series of nitrogen mustard (N-mustard) agents formed after inserting 1 to 9 methylene groups (-CH2-) between 2 -N(CH2CH2Cl)2 groups. These compounds were shown to have significant correlations and associations in their properties after analysis by pattern recognition methods including hierarchical classification, cluster analysis, nonmetric multi-dimensional scaling (MDS), detrended correspondence analysis, K-means cluster analysis, discriminant analysis, and self-organizing tree algorithm (SOTA) analysis. Detrended correspondence analysis showed a linear-like association of the 9 homologs, and hierarchical classification showed that each homolog had great similarity to at least one other member of the series-as did cluster analysis using paired-group distance measure. Nonmetric multi-dimensional scaling was able to discriminate homologs 2 and 3 (by number of methylene groups) from homologs 4, 5, and 6 as a group, and from homologs 7, 8, and 9 as a group. Discriminant analysis, K-means cluster analysis, and hierarchical classification distinguished the high molecular weight homologs from low molecular weight homologs. As the number of methylene groups increased the aqueous solubility decreased, dermal permeation coefficient increased, Log P increased, molar volume increased, parachor increased, and index of refraction decreased. Application of pattern recognition methods discerned useful interrelationships within the homologous series that will determine specific and beneficial clinical applications for each homolog and methods of administration.