Pattern Classification Approach Research Articles

Auditory perception and syntactic cognition: brain activity‐based decoding within and across subjects

The present magnetoencephalography study investigated whether the brain states of early syntactic and auditory-perceptual processes can be decoded from single-trial recordings with a multivariate pattern classification approach. In particular, it was investigated whether the early neural activation patterns in response to rule violations in basic auditory perception and in high cognitive processes (syntax) reflect a functional organization that largely generalizes across individuals or is subject-specific. On this account, subjects were auditorily presented with correct sentences, syntactically incorrect sentences, correct sentences including an interaural time difference change, and sentences containing both violations. For the analysis, brain state decoding was carried out within and across subjects with three pairwise classifications. Neural patterns elicited by each of the violation sentences were separately classified with the patterns elicited by the correct sentences. The results revealed the highest decoding accuracies over temporal cortex areas for all three classification types. Importantly, both the magnitude and the spatial distribution of decoding accuracies for the early neural patterns were very similar for within-subject and across-subject decoding. At the same time, across-subject decoding suggested a hemispheric bias, with the most consistent patterns in the left hemisphere. Thus, the present data show that not only auditory-perceptual processing brain states but also cognitive brain states of syntactic rule processing can be decoded from single-trial brain activations. Moreover, the findings indicate that the neural patterns in response to syntactic cognition and auditory perception reflect a functional organization that is highly consistent across individuals.

ERP-based multivariate pattern classification predicts errors before an overt response is executed

Event Abstract Back to Event ERP-based multivariate pattern classification predicts errors before an overt response is executed Stefan Bode1* and Jutta Stahl2 1 University of Melbourne, Melbourne School of Psychological Sciences, Australia 2 University of Cologne, Department of Psychology, Germany Error detection is an important aspect of cognitive control. The earliest electrophysiological marker of error processing is the error-related negativity (ERN) that peaks ~80-100ms after an overt response and could be related to the detection of a mismatch between required and performed response representations. It is unknown, however, whether error processing could take place even before the execution of an already initiated erroneous response. We used a highly sensitive multivariate pattern classification approach for electroencephalography that can detect choice-related information before it is reflected in average event-related potentials (ERP) (Bode et al., 2012, J Neurosci). Data from 18 participants (from a larger sample of 94) that made ≥50 errors in an Eriksen-flanker-task were analysed. In this task, participants were instructed to respond to the central letters on the screen and ignore the flanking letters that could be congruent (HHHHH, SSSSS) or incongruent (HHSHH, SSHSS). After performing a current source density analysis on the data to reduce local redundancies, the spatial-temporal patterns within all 10ms time-windows across all 61 channels were analysed using linear support vector machine classifiers. A 10-fold cross-validation procedure was employed for training and testing on randomly drawn single trials. We could predict errors from correct responses from ~40ms before response initiation, significantly preceding the ERN. This effect was mainly driven by congruent trials, supporting error detection theories, which implies that for errors in congruent trials mismatch detection is facilitated. We show that the brain might process these error signals even faster than postulated so far. References Bode S, Sewell D, Lilburn S, Forte J, Smith PL, Stahl J (2012). Predicting perceptual decision biases from early brain activity. J Neurosci, 32, 12488-98. Keywords: error processing, Electroencephalography (EEG), multivariate pattern classification, prediction of errors, pre-response signals Conference: ACNS-2012 Australasian Cognitive Neuroscience Conference, Brisbane, Australia, 29 Nov - 2 Dec, 2012. Presentation Type: Oral Presentation Topic: Executive Processes Citation: Bode S and Stahl J (2012). ERP-based multivariate pattern classification predicts errors before an overt response is executed. Conference Abstract: ACNS-2012 Australasian Cognitive Neuroscience Conference. doi: 10.3389/conf.fnhum.2012.208.00019 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 25 Oct 2012; Published Online: 07 Nov 2012. * Correspondence: Dr. Stefan Bode, University of Melbourne, Melbourne School of Psychological Sciences, Melbourne, Australia, 314026@frontiersin.org Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Stefan Bode Jutta Stahl Google Stefan Bode Jutta Stahl Google Scholar Stefan Bode Jutta Stahl PubMed Stefan Bode Jutta Stahl Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Intelligent Risk Assessment for Pipeline Third-Party Interference

In petroleum industry, pipeline is singled out as it is the safest and the most economically viable means of transporting large quantities of oil and natural gas. However, accidents to pipelines because of the third-party interference have been recorded. An intelligent risk assessment approach is proposed to estimate the risk of each pipeline section and classify various risk patterns, using self-organization mapping neural network theory, which incorporates the factors of pipeline laying conditions, historical damage records, safety-related actions, management measures, and the environment around the underling pipeline. A field case study of Shaanxi–Beijing gas pipeline in China is undertook so that the effectiveness of the proposed risk pattern classification approach could be verified, which helps safety engineer to take effective and accurate safety measures according to different risk patterns.

Neurodegeneration Across Stages of Cognitive Decline in Parkinson Disease

To assess regions and patterns of brain atrophy in patients with Parkinson disease (PD) with normal cognition (PD-NC), mild cognitive impairment (PD-MCI), and dementia-level cognitive deficits (PDD). Images were quantified using a region-of-interest approach and voxel-based morphometry analysis. We used a high-dimensional pattern classification approach to delineate brain regions that collectively formed the Spatial Pattern of Abnormalities for Recognition of PDD. The Parkinson's Disease and Movement Disorders Center at the University of Pennsylvania. Eighty-four PD patients (61 PD-NC, 12 PD-MCI, and 11 PDD) and 23 healthy control subjects (HCs) underwent magnetic resonance imaging of the brain. The PD-NC patients did not demonstrate significant brain atrophy compared with HCs. Compared with PD-NC patients, PD-MCI patients had hippocampal atrophy (β = -0.37; P = .001), and PDD patients demonstrated hippocampal (β = -0.32; P = .004) and additional medial temporal lobe atrophy (β = -0.36; P = .003). The PD-MCI patients had a different pattern of atrophy compared with PD-NC patients (P = .04) and a similar pattern to that of PDD patients (P = .81), characterized by hippocampal, prefrontal cortex gray and white matter, occipital lobe gray and white matter, and parietal lobe white matter atrophy. In nondemented PD patients, there was a correlation between memory-encoding performance and hippocampal volume. Hippocampal atrophy is a biomarker of initial cognitive decline in PD, including impaired memory encoding and storage, suggesting heterogeneity in the neural substrate of memory impairment. Use of a pattern classification approach may allow identification of diffuse regions of cortical gray and white matter atrophy early in the course of cognitive decline.

Assessing and classifying risk of pipeline third-party interference based on fault tree and SOM

Accidents to pipelines because of the third-party interference have been recorded and they often result in catastrophic consequences for environment and society with a great deal of economic loss. The third-party interference resulting from complicated origins occurs randomly, and is hard to be forecasted or controlled in advance, so it becomes a serious threat to the safe operation of long transmission pipeline. This paper focuses on the application of self-organizing maps (SOMs) to assess the risk of third-party interference and classify their risk patterns. In this work, fault tree is used first to establish the risk assessment index system, and then SOM is used in multi-parameter risk pattern classification approach, which is proposed to present various risk maps, incorporating the factors of pipeline laying conditions, historical damage records, safety-related actions, management measures and the environment around the underling pipeline. A field case study of Shaanxi–Beijing gas pipeline in China is undertaken so that the effectiveness of the proposed approach could be verified. By taking the classification results into consideration, the decision maker may well get precious and differentiated information about the pipeline risk distribution of third-party interference and make appropriate safety-related actions to prevent the damage.

Structural damage detection using bi-temporal optical satellite images

Multi-temporal satellite imagery is now available at sub-metre accuracy and has been found to be very useful for performing rapid damage assessment on human settlement areas affected by large-scale disasters. In this article, a method of formulating structural damage detection measures based on pre- and post-disaster satellite images is proposed. To validate the proposed damage measures, building-based structural damage assessment is conducted. First, their effectiveness in representing multilevel structural damage is demonstrated using synthetic patterns of building damage. Second, the damage classification accuracy is evaluated by means of a pattern classification approach applied to a pair of bi-temporal satellite images, wherein earthquake damage to hundreds of buildings is assessed. The article concludes that the proposed damage detection measures, which are conceptually simple and computationally efficient, outperform traditional measures, such as linear correlation coefficients.

A classifier for Bangla handwritten numeral recognition

This paper presents a novel pattern classification approach – a kernel and Bayesian discriminant based classifier which utilizes the distribution characteristics of the samples in each class. A kernel combined with Bayesian discriminant in the subspace spanned by the eigenvectors which are associated with the smaller eigenvalues in each class is adopted as the classification criterion. To solve the problem of the matrix inverse, the smaller eigenvalues are substituted by a small threshold which is decided by minimizing the training error in a given database. Application of the proposed classifier to the issue of handwritten numeral recognition demonstrates that it is promising in practical applications.

Application of support vector machine based on pattern spectrum entropy in fault diagnostics of rolling element bearings

This paper presents a novel pattern classification approach for the fault diagnostics of rolling element bearings, which combines the morphological multi-scale analysis and the ‘one to others’ support vector machine (SVM) classifiers. The morphological pattern spectrum describes the shape characteristics of the inspected signal based on the morphological opening operation with multi-scale structuring elements. The pattern spectrum entropy and the barycenter scale location of the spectrum curve are extracted as the feature vectors presenting different faults of the bearing, which are more effective and representative than the kurtosis and the enveloping demodulation spectrum. The ‘one to others’ SVM algorithm is adopted to distinguish six kinds of fault signals which were measured in the experimental test rig under eight different working conditions. The recognition results of the SVM are ideal and more precise than those of the artificial neural network even though the training samples are few. The combination of the morphological pattern spectrum parameters and the ‘one to others’ multi-class SVM algorithm is suitable for the on-line automated fault diagnosis of the rolling element bearings. This application is promising and worth well exploiting.

Dissociable neural imprints of perception and grammar in auditory functional imaging.

In language processing, the relative contribution of early sensory and higher cognitive brain areas is still an open issue. A recent controversial hypothesis proposes that sensory cortices show sensitivity to syntactic processes, whereas other studies suggest a wider neural network outside sensory regions. The goal of the current event-related fMRI study is to clarify the contribution of sensory cortices in auditory syntactic processing in a 2 × 2 design. Two-word utterances were presented auditorily and varied both in perceptual markedness (presence or absence of an overt word category marking "-t"), and in grammaticality (syntactically correct or incorrect). A multivariate pattern classification approach was applied to the data, flanked by conventional cognitive subtraction analyses. The combination of methods and the 2 × 2 design revealed a clear picture: The cognitive subtraction analysis found initial syntactic processing signatures in a neural network including the left IFG, the left aSTG, the left superior temporal sulcus (STS), as well as the right STS/STG. Classification of local multivariate patterns indicated the left-hemispheric regions in IFG, aSTG, and STS to be more syntax-specific than the right-hemispheric regions. Importantly, auditory sensory cortices were only sensitive to the overt perceptual marking, but not to the grammaticality, speaking against syntax-inflicted sensory cortex modulations. Instead, our data provide clear evidence for a distinction between regions involved in pure perceptual processes and regions involved in initial syntactic processes.

Mapping Software Metrics to Module Complexity: A Pattern Classification Approach

A desirable software engineering goal is the prediction of software module complexity (a qualitative concept) using automatically generated software metrics (quantitative measurements). This goal may be couched in the language of pattern classification; namely, given a set of metrics (a pattern) for a software module, predict the class (level of complexity) to which the module belongs. To find this mapping from metrics to complexity, we present a classification strategy, stochastic metric selection, to determine the subset of software metrics that yields the greatest predictive power with respect to module complexity. We demonstrate the effectiveness of this strategy by empirically evaluating it using a publicly available dataset of metrics compiled from a medical imaging system and comparing the prediction results against several classification system benchmarks.

Static Security Evaluation in Power Systems usingMulti-Class SVM with Different Parameter SelectionMethods

Security evaluation is a major concern in real time operation of electric power systems. Traditional method of security evaluation performed by continuous load flow analysis involves long computer time and generates voluminous results. This paper presents a practical and feasible Support Vector Machine Based Pattern Classification (SVMBPC) approach for static security evaluation in power systems. The proposed approach classifies the security status of any given operating condition in one of the four classes - Secure, Critically Secure, Insecure and Highly Insecure based on the computation of a numeric value called security index. The feature selection stage uses a simple and straightforward forward sequential method to select the best feature set from a large set of variables. The static security classifier is designed by a multi-class SVM with different parameter tuning methods. The proposed approach is implemented in New England 39 bus and IEEE 118 bus systems and the results are validated.

Editorial: People are approximators, not perfectors

This issue has three papers. The first, Finding failures from passed test cases: Improving the pattern classification approach to the testing of mesh simplification programs, by W. K. Chan, Jeffrey C. F. Ho, and T. H. Tse, applies metamorphic testing to test mesh simplification programs. The second, Fault localization based on information flow coverage, by Wes Masri, presents results on using information flow to aid fault localization. The third, Fault localization using a model checker, by Andreas Griesmayer, Stefan Staber, and Roderick Bloem, presents ideas using model checking to support fault localization. I recently read Norman's classic book, The Design of Everyday Things [1]. Though 22-years old, it's an excellent read and still very relevant. One point Norman made is that people are good at approximations, but not at being perfect. This cognitive aspect of people affects the design and production of everyday things (and software). For most of human history, we created each product by hand, and made each product a little better than the previous. That is, our ancestors used natural continuous product evolution. But during the industrial revolution, we invented the idea of mass production. Suddenly design became vastly more important. Design mistakes could not be corrected in the next product, but were immediately placed into thousands of products. Then we invented software, making things worse. Software drives the production cost to almost zero, putting even more pressure on designers. This led to decades of attempts to get software to be perfect the first time. We invented processes, formal methods, specifications, complicated test criteria, designed modeling languages, etc., all with the goal of getting the software right before producing thousands of copies. With great efforts from academic researchers, industrial research labs, and industry practitioners, we have made progress toward getting software right the first time. Our software has gotten better. But Norman's point makes one thing crystal clear: In a very real way, the efforts in software engineering to achieve perfection are quixotic tilting at windmills, fighting against basic human nature. Simply put: People are not naturally good at being perfect! However, we are good at approximating and improving. In fact, our ancestors have used that approach for thousands of years. And in software, we are coming full circle … Large segments of the software industry have recently seen enormous changes in the deployment process. Web applications, updates delivered through the web, and the ability to push updates onto mobile devices allow us to update software daily or even hourly. That is, we are returning to our roots, continuous evolution, just like our ancestors did for millennia. Creating frequent updates where we try something, identify problems, then quickly make changes, is a natural human process and is being used in the industry. I recently attended a talk by the Senior Engineering Director at Google, who said that Google emphasizes ‘fast fixing’ over ‘prevention,’ in part by having a very short test-fix-update evolutionary cycle [2]. This approach will be suspicious to many academics. Most of us learned our craft in an environment where new versions come out every 4 or 5 years, and from teachers who emphasized safety critical software, where evolutionary test-fix-update process is more risky (if we crash the plane, don't expect another chance!). But for a large (and growing) part of the software industry, this process works. And yes, this has majorimpacts on software. A few years ago I published a paper on how quality is more important in web software [3]. Norman and Copeland have helped me to add a missing piece to that article: continuous evolution to approximate and improve our software. I believe this to be a good thing. And every testing researcher should pause and ask a question: Which of my ideas are relevant to this new reality?

Probabilistic Urban Structural Damage Classification Using Bitemporal Satellite Images

Recent research endeavors in civil engineering have attempted to apply remote sensing technology to urban damage assessment as an aid for post-disaster reconnaissance and recovery. In these attempts, urban structural damage is identified based on pre- and post-disaster satellite images with the use of a pattern classification approach. The result is usually presented in a damage map wherein categorical damage levels, such as “fully collapsed,” “partially collapsed,” or “intact,” are assigned to urban subregions or individual structures in images. However, a major limitation in past attempts is the use of deterministic approaches to classify damage levels. In general, these approaches are not able to capture the inherent uncertainties of structural damage and lack scalability when analyzing damage to built urban subregions of different sizes. To address this, a probabilistic classification framework by means of a multiclass classifier is proposed. By applying this probabilistic approach, classification of urban damage provides posterior probabilities, which can be used to quantify decision uncertainties and to obtain regional urban damage classification. Numerical experiments are conducted using satellite images acquired from a recent earthquake and a tsunami event, namely the 2003 Bam, Iran Earthquake, and the 2004 India Ocean Tsunami.

Attended and Unattended Products Direct Buying Options Using the Same Neural Circuits

Attended and Unattended Products Direct Buying Options Using the Same Neural Circuits

Prediction of the success of cardiac resuscitation: a pattern classification approach based on ECG spectral and temporal features

The ECG tracings recorded during a ventricular fibrillation (VF) using an automated external defibrillator (AED) contain useful information predictive of shock outcome. The focus is on the VF waveforms' morphology. The amplitude and the spectral properties of VF may predict the likelihood of successful defibrillation [1-4]. In almost all previous studies, the amplitude or the spectral properties of the ECG tracings have been singularly used. However, these approaches have led to methods lacking sufficient predictive power.

A comparison of fMRI adaptation and multivariate pattern classification analysis in visual cortex

Functional magnetic resonance imaging (fMRI) has become a ubiquitous tool in cognitive neuroscience. The technique allows noninvasive measurements of cortical responses in the human brain, but only on the millimeter scale. Because a typical voxel contains many thousands of neurons with varied properties, establishing the selectivity of their responses directly is impossible. In recent years, two methods using fMRI aimed at studying the selectivity of neuronal populations on a ‘subvoxel’ scale have been heavily used. The first technique, fMRI adaptation, relies on the observation that the blood oxygen level-dependent (BOLD) response in a given voxel is reduced after prolonged presentation of a stimulus, and that this reduction is selective to the characteristics of the repeated stimuli (adapters). The second technique, multivariate pattern analysis (MVPA), makes use of multivariate statistics to recover small biases in individual voxels in their responses to different stimuli. It is thought that these biases arise due to the uneven distribution of neurons (with different properties) sampled by the many voxels in the imaged volume. These two techniques have not been compared explicitly, however, and little is known about their relative sensitivities. Here, we compared fMRI results from orientation-specific visual adaptation and orientation–classification by MVPA, using optimized experimental designs for each, and found that the multivariate pattern classification approach was more sensitive to small differences in stimulus orientation than the adaptation paradigm. Estimates of orientation selectivity obtained with the two methods were, however, very highly correlated across visual areas.

Investigating the predictive value of whole-brain structural MR scans in autism: A pattern classification approach

Autistic spectrum disorder (ASD) is accompanied by subtle and spatially distributed differences in brain anatomy that are difficult to detect using conventional mass-univariate methods (e.g., VBM). These require correction for multiple comparisons and hence need relatively large samples to attain sufficient statistical power. Reports of neuroanatomical differences from relatively small studies are thus highly variable. Also, VBM does not provide predictive value, limiting its diagnostic value.Here, we examined neuroanatomical networks implicated in ASD using a whole-brain classification approach employing a support vector machine (SVM) and investigated the predictive value of structural MRI scans in adults with ASD. Subsequently, results were compared between SVM and VBM. We included 44 male adults; 22 diagnosed with ASD using “gold-standard” research interviews and 22 healthy matched controls.SVM identified spatially distributed networks discriminating between ASD and controls. These included the limbic, frontal-striatal, fronto-temporal, fronto-parietal and cerebellar systems. SVM applied to gray matter scans correctly classified ASD individuals at a specificity of 86.0% and a sensitivity of 88.0%. Cases (68.0%) were correctly classified using white matter anatomy. The distance from the separating hyperplane (i.e., the test margin) was significantly related to current symptom severity. In contrast, VBM revealed few significant between-group differences at conventional levels of statistical stringency.We therefore suggest that SVM can detect subtle and spatially distributed differences in brain networks between adults with ASD and controls. Also, these differences provide significant predictive power for group membership, which is related to symptom severity.

Combining different biometric traits with one-class classification

It has been demonstrated in the literature that the combining of different biometric traits is a powerful tool to overcome the limitations imposed by a single biometric system. The fusion of different systems can be approached in different ways. In this work, we consider the pattern classification approach, where the scores of the various systems are used as features to feed the classifiers. More specifically, we are interested in one-class classifiers, and we show that one-class classification could be considered as an alternative to biometric fusion, especially when the data are highly unbalanced or when data from only a single class are available. The results reported for one-class classification on two different databases compares with the standard two-class SVM and surpasses all the conventional classifier combination rules tested.

Finding failures from passed test cases: improving the pattern classification approach to the testing of mesh simplification programs

AbstractMesh simplification programs create three‐dimensional polygonal models similar to an original polygonal model, and yet use fewer polygons. They produce different graphics even though they are based on the same original polygonal model. This results in a test oracle problem. To address the problem, our previous work has developed a technique that uses a reference model of the program under test to train a classifier. Using such an approach may mistakenly mark a failure‐causing test case as passed. It lowers the testing effectiveness of revealing failures. This paper suggests piping the test cases marked as passed by a statistical pattern classification module to an analytical metamorphic testing (MT) module. We evaluate our approach empirically using three subject programs with over 2700 program mutants. The result shows that, using a resembling reference model to train a classifier, the integrated approach can significantly improve the failure detection effectiveness of the pattern classification approach. We also explain how MT in our design trades specificity for sensitivity. Copyright © 2009 John Wiley & Sons, Ltd.

Fuzzy declustering-based vector quantization

Vector quantization is a useful approach for multi-dimensional data compression and pattern classification. One of the most popular techniques for vector quantization design is the LBG (Linde, Buzo, Gray) algorithm. To address the problem of producing poor estimate of vector centroids which are subjected to biased data in vector quantization; we propose a fuzzy declustering strategy for the LBG algorithm. The proposed technique calculates appropriate declustering weights to adjust the global data distribution. Using the result of fuzzy declustering-based vector quantization design, we incorporate the notion of fuzzy partition entropy into the distortion measures that can be useful for classification of spectral features. Experimental results obtained from simulated and real data sets demonstrate the effective performance of the proposed approach.