Classification of echolocation calls from 14 species of bat by Support Vector Machines and Ensembles of Neural Networks

Calls from 14 species of bat were classified to genus and species using discriminant function analysis (DFA), support vector machines (SVM) and ensembles of neural networks (ENN). Both SVMs and ENNs outperformed DFA for every species while ENNs (mean identification rate – 97%) consistently outperformed SVMs (mean identification rate – 87%). Correct classification rates produced by the ENNs varied from 91% to 100%; calls from six species were correctly identified with 100% accuracy. Calls from the five species of Myotis, a genus whose species are considered difficult to distinguish acoustically, had correct identification rates that varied from 91 – 100%. Five parameters were most important for classifying calls correctly while seven others contributed little to classification performance.

A high performance liquid chromatography and electrospray ionization mass spectrometry method for the analysis of the natural medicine, Forsythia suspensa

Samples of Forsythia suspensa from raw (Laoqiao) and ripe (Qingqiao) fruit were analyzed with the use of HPLC-DAD and the EIS-MS techniques. Seventeen peaks were detected, and of these, twelve were identified. Most were related to the glucopyranoside molecular fragment. Samples collected from three geographical areas (Shanxi, Henan and Shandong Provinces), were discriminated with the use of hierarchical clustering analysis (HCA), discriminant analysis (DA), and principal component analysis (PCA) models, but only PCA was able to provide further information about the relationships between objects and loadings; eight peaks were related to the provinces of sample origin. The supervised classification models-K-nearest neighbor (KNN), least squares support vector machines (LS-SVM), and counter propagation artificial neural network (CP-ANN) methods, indicated successful classification but KNN produced 100% classification rate. Thus, the fruit were discriminated on the basis of their places of origin.

Functional neuroanatomy of auditory mismatch processing : an event-related fMRI study of duration-deviant oddballs

This study was designed to identify the neural networks underlying automatic auditory deviance detection in 10 healthy subjects using functional magnetic resonance imaging. We measured blood oxygenation level-dependent contrasts derived from the comparison of blocks of stimuli presented as a series of standard tones (50 ms duration) alone versus blocks that contained rare duration-deviant tones (100 ms) that were interspersed among a series of frequent standard tones while subjects were watching a silent movie. Possible effects of scanner noise were assessed by a “no tone” condition. In line with previous positron emission tomography and EEG source modeling studies, we found temporal lobe and prefrontal cortical activation that was associated with auditory duration mismatch processing. Data were also analyzed employing an event-related hemodynamic response model, which confirmed activation in response to duration-deviant tones bilaterally in the superior temporal gyrus and prefrontally in the right inferior and middle frontal gyri. In line with previous electrophysiological reports, mismatch activation of these brain regions was significantly correlated with age. These findings suggest a close relationship of the event-related hemodynamic response pattern with the corresponding electrophysiological activity underlying the event-related “mismatch negativity” potential, a putative measure of auditory sensory memory.

Wide-band spectral power fluctuations characterize the response of simulated cortical networks to increasing stimulus intensity

The hippocampus is an anatomically distinct region of the medial temporal lobe that plays a critical role in the formation of declarative memories. Here we show that a computer simulation of simple compartmental cells organized with basic hippocampal connectivity is capable of producing stimulus intensity sensitive wide-band fluctuations of spectral power similar to that seen in real EEG. While previous computational models have been designed to assess the viability of the putative mechanisms of memory storage and retrieval, they have generally been too abstract to allow comparison with empirical data. Furthermore, while the anatomical connectivity and organization of the hippocampus is well defined, many questions regarding the mechanisms that mediate large-scale synaptic integration remain unanswered. For this reason we focus less on the specifics of changing synaptic weights and more on the population dynamics. Spectral power in four distinct frequency bands were derived from simulated field potentials of the computational model and found to depend on the intensity of a random input. The majority of power occurred in the lowest frequency band (3-6 Hz) and was greatest to the lowest intensity stimulus condition (1% maximal stimulus). In contrast, higher frequency bands ranging from 7-45 Hz show an increase in power directly related with an increase in stimulus intensity. This trend continues up to a stimulus level of 15% to 20% of the maximal input, above which power falls dramatically. These results suggest that the relative power of intrinsic network oscillations are dependent upon the level of activation and that above threshold levels all frequencies are damped, perhaps due to over activation of inhibitory interneurons.

Memory and Epileptogenesis in Complex Biological and Simulated Systems

Oscillations of neural activity may bind widespread cortical areas into a neural representation that encodes disparate aspects of an event. In order to test this theory we have turned to data collected from complex partial epilepsy (CPE) patients with chronically implanted depth electrodes. Data from regions critical to word and face information processing was analyzed using spectral coherence measurements. Similar analyses of intracranial EEG (iEEG) during seizure episodes display HippoCampal Formation (HCF)—NeoCortical (NC) spectral coherence patterns that are characteristic of specific seizure stages (Klopp et al. 1996). We are now building a computational memory model to examine whether spatio-temporal patterns of human iEEG spectral coherence emerge in a computer simulation of HCF cellular distribution, membrane physiology and synaptic connectivity. Once the model is reasonably scaled it will be used as a tool to explore neural parameters that are critical to memory formation and epileptogenesis.

Supervised latent dirichlet allocation models for efficient activity representation

Local spatio-temporal features with a Bag-of-visual words model is a popular approach used in human action recognition. Bag-of-features methods suffer from several challenges such as extracting appropriate appearance and motion features from videos, converting extracted features appropriate for classification and designing a suitable classification framework. In this paper we address the problem of efficiently representing the extracted features for classification to improve the overall performance. We introduce two generative supervised topic models, maximum entropy discrimination LDA (MedLDA) and class- specific simplex LDA (css-LDA), to encode the raw features suitable for discriminative SVM based classification. Unsupervised LDA models disconnect topic discovery from the classification task, hence yield poor results compared to the baseline Bag-of-words framework. On the other hand supervised LDA techniques learn the topic structure by considering the class labels and improve the recognition accuracy significantly. MedLDA maximizes likelihood and within class margins using max-margin techniques and yields a sparse highly discriminative topic structure; while in css-LDA separate class specific topics are learned instead of common set of topics across the entire dataset. In our representation first topics are learned and then each video is represented as a topic proportion vector, i.e. it can be comparable to a histogram of topics. Finally SVM classification is done on the learned topic proportion vector. We demonstrate the efficiency of the above two representation techniques through the experiments carried out in two popular datasets. Experimental results demonstrate significantly improved performance compared to the baseline Bag-of-features framework which uses kmeans to construct histogram of words from the feature vectors.

Automatic UAV forced landing site detection using machine learning

The commercialization of aerial image processing is highly dependent on the platforms such as UAVs (Unmanned Aerial Vehicles). However, the lack of an automated UAV forced landing site detection system has been identified as one of the main impediments to allow UAV flight over populated areas in civilian airspace. This article proposes a UAV forced landing site detection system that is based on machine learning approaches including the Gaussian Mixture Model and the Support Vector Machine. A range of learning parameters are analysed including the number of Guassian mixtures, support vector kernels including linear, radial basis function Kernel (RBF) and polynormial kernel (poly), and the order of RBF kernel and polynormial kernel. Moreover, a modified footprint operator is employed during feature extraction to better describe the geometric characteristics of the local area surrounding a pixel. The performance of the presented system is compared to a baseline UAV forced landing site detection system which uses edge features and an Artificial Neural Network (ANN) region type classifier. Experiments conducted on aerial image datasets captured over typical urban environments reveal improved landing site detection can be achieved with an SVM classifier with an RBF kernel using a combination of colour and texture features. Compared to the baseline system, the proposed system provides significant improvement in term of the chance to detect a safe landing area, and the performance is more stable than the baseline in the presence of changes to the UAV altitude.

Application of anomaly technique in wind turbine bearing fault detection

Bearing faults are the most common cause of wind turbine failures. Unavailability and maintenance cost of wind turbines are becoming critically important, with their fast growing in electric networks. Early fault detection can reduce outage time and costs. This paper proposes Anomaly Detection (AD) machine learning algorithms for fault diagnosis of wind turbine bearings. The application of this method on a real data set was conducted and is presented in this paper. For validation and comparison purposes, a set of baseline results are produced using the popular one-class SVM methods to examine the ability of the proposed technique in detecting incipient faults.

Neuroimaging and genetics: Exploring, searching, and finding

This issue on the genetics of brain imaging phenotypes is a celebration of the happy marriage between two of science's highly interesting fields: neuroscience and genetics. The articles collected here are ample evidence that a good deal of synergy exists in this marriage. A wide selection of papers is presented that provide many different perspectives on how genes cause variation in brain structure and function, which in turn influence behavioral phenotypes (including psychopathology). They are examples of the many different methodologies in contemporary genetics and neuroscience research. Genetic methodology includes genome-wide association (GWA), candidate-gene association, and twin studies. Sources of data on brain phenotypes include cortical gray matter (GM) structural/volumetric measures from magnetic resonance imaging (MRI); white matter (WM) measures from diffusion tensor imaging (DTI), such as fractional anisotropy; functional- (activity-) based measures from electroencephalography (EEG), and functional MRI (fMRI). Together, they reflect a combination of scientific fields that have seen great technological advances, whether it is the single-nucleotide polymorphism (SNP) array in genetics, the increasingly high-resolution MRI imaging, or high angular resolution diffusion imaging technique for measuring WM connective properties.

Combining HPLC-DAD and ICP-MS data for improved analysis of complex samples: Classification of the root samples from Cortex moutan

A combined data matrix consisting of high performance liquid chromatography–diode array detector (HPLC–DAD) and inductively coupled plasma-mass spectrometry (ICP-MS) measurements of samples from the plant roots of the Cortex moutan (CM), produced much better classification and prediction results in comparison with those obtained from either of the individual data sets. The HPLC peaks (organic components) of the CM samples, and the ICP-MS measurements (trace metal elements) were investigated with the use of principal component analysis (PCA) and the linear discriminant analysis (LDA) methods of data analysis; essentially, qualitative results suggested that discrimination of the CM samples from three different provinces was possible with the combined matrix producing best results. Another three methods, K-nearest neighbor (KNN), back-propagation artificial neural network (BP-ANN) and least squares support vector machines (LS-SVM) were applied for the classification and prediction of the samples. Again, the combined data matrix analyzed by the KNN method produced best results (100% correct; prediction set data). Additionally, multiple linear regression (MLR) was utilized to explore any relationship between the organic constituents and the metal elements of the CM samples; the extracted linear regression equations showed that the essential metals as well as some metallic pollutants were related to the organic compounds on the basis of their concentrations

A data mining approach for fault diagnosis: An application of anomaly detection algorithm

Rolling-element bearing failures are the most frequent problems in rotating machinery, which can be catastrophic and cause major downtime. Hence, providing advance failure warning and precise fault detection in such components are pivotal and cost-effective. The vast majority of past research has focused on signal processing and spectral analysis for fault diagnostics in rotating components. In this study, a data mining approach using a machine learning technique called anomaly detection (AD) is presented. This method employs classification techniques to discriminate between defect examples. Two features, kurtosis and Non-Gaussianity Score (NGS), are extracted to develop anomaly detection algorithms. The performance of the developed algorithms was examined through real data from a test to failure bearing. Finally, the application of anomaly detection is compared with one of the popular methods called Support Vector Machine (SVM) to investigate the sensitivity and accuracy of this approach and its ability to detect the anomalies in early stages.

Partial epilepsy syndrome in a Gypsy family linked to 5q31.3-q32

PURPOSE The restricted genetic diversity and homogeneous molecular basis of Mendelian disorders in isolated founder populations have rarely been explored in epilepsy research. Our long-term goal is to explore the genetic basis of epilepsies in one such population, the Gypsies. The aim of this report is the clinical and genetic characterization of a Gypsy family with a partial epilepsy syndrome. METHODS Clinical information was collected using semistructured interviews with affected subjects and informants. At least one interictal electroencephalography (EEG) recording was performed for each patient and previous data obtained from records. Neuroimaging included structural magnetic resonance imaging (MRI). Linkage and haplotype analysis was performed using the Illumina IVb Linkage Panel, supplemented with highly informative microsatellites in linked regions and Affymetrix SNP 5.0 array data. RESULTS We observed an early-onset partial epilepsy syndrome with seizure semiology strongly suggestive of temporal lobe epilepsy (TLE), with mild intellectual deficit co-occurring in a large proportion of the patients. Psychiatric morbidity was common in the extended pedigree but did not cosegregate with epilepsy. Linkage analysis definitively excluded previously reported loci, and identified a novel locus on 5q31.3-q32 with an logarithm of the odds (LOD) score of 3 corresponding to the expected maximum in this family. DISCUSSION The syndrome can be classified as familial temporal lobe epilepsy (FTLE) or possibly a new syndrome with mild intellectual deficit. The linked 5q region does not contain any ion channel-encoding genes and is thus likely to contribute new knowledge about epilepsy pathogenesis. Identification of the mutation in this family and in additional patients will define the full phenotypic spectrum.

Comparison of decision tree, support vector machines, and Bayesian network approaches for classification of falls in Parkinson’s disease

Being able to accurately predict the risk of falling is crucial in patients with Parkinson’s dis- ease (PD). This is due to the unfavorable effect of falls, which can lower the quality of life as well as directly impact on survival. Three methods considered for predicting falls are decision trees (DT), Bayesian networks (BN), and support vector machines (SVM). Data on a 1-year prospective study conducted at IHBI, Australia, for 51 people with PD are used. Data processing are conducted using rpart and e1071 packages in R for DT and SVM, con- secutively; and Bayes Server 5.5 for the BN. The results show that BN and SVM produce consistently higher accuracy over the 12 months evaluation time points (average sensitivity and specificity > 92%) than DT (average sensitivity 88%, average specificity 72%). DT is prone to imbalanced data so needs to adjust for the misclassification cost. However, DT provides a straightforward, interpretable result and thus is appealing for helping to identify important items related to falls and to generate fallers’ profiles.

Learning with symmetric label noise: The importance of being unhinged

Convex potential minimisation is the de facto approach to binary classification. However, Long and Servedio [2008] proved that under symmetric label noise (SLN), minimisation of any convex potential over a linear function class can result in classification performance equivalent to random guessing. This ostensibly shows that convex losses are not SLN-robust. In this paper, we propose a convex, classification-calibrated loss and prove that it is SLN-robust. The loss avoids the Long and Servedio [2008] result by virtue of being negatively unbounded. The loss is a modification of the hinge loss, where one does not clamp at zero; hence, we call it the unhinged loss. We show that the optimal unhinged solution is equivalent to that of a strongly regularised SVM, and is the limiting solution for any convex potential; this implies that strong l2 regularisation makes most standard learners SLN-robust. Experiments confirm the unhinged loss’ SLN-robustness.

A flexible hierarchical approach for facial age estimation based on multiple features

Age estimation from facial images is increasingly receiving attention to solve age-based access control, age-adaptive targeted marketing, amongst other applications. Since even humans can be induced in error due to the complex biological processes involved, finding a robust method remains a research challenge today. In this paper, we propose a new framework for the integration of Active Appearance Models (AAM), Local Binary Patterns (LBP), Gabor wavelets (GW) and Local Phase Quantization (LPQ) in order to obtain a highly discriminative feature representation which is able to model shape, appearance, wrinkles and skin spots. In addition, this paper proposes a novel flexible hierarchical age estimation approach consisting of a multi-class Support Vector Machine (SVM) to classify a subject into an age group followed by a Support Vector Regression (SVR) to estimate a specific age. The errors that may happen in the classification step, caused by the hard boundaries between age classes, are compensated in the specific age estimation by a flexible overlapping of the age ranges. The performance of the proposed approach was evaluated on FG-NET Aging and MORPH Album 2 datasets and a mean absolute error (MAE) of 4.50 and 5.86 years was achieved respectively. The robustness of the proposed approach was also evaluated on a merge of both datasets and a MAE of 5.20 years was achieved. Furthermore, we have also compared the age estimation made by humans with the proposed approach and it has shown that the machine outperforms humans. The proposed approach is competitive with current state-of-the-art and it provides an additional robustness to blur, lighting and expression variance brought about by the local phase features.