A novel method for automated classification of epileptiform activity in the human electroencephalogram-based on independent component analysis.

Diagnosis of several neurological disorders is based on the detection of typical pathological patterns in the electroencephalogram (EEG). This is a time-consuming task requiring significant training and experience. Automatic detection of these EEG patterns would greatly assist in quantitative analysis and interpretation. We present a method, which allows automatic detection of epileptiform events and discrimination of them from eye blinks, and is based on features derived using a novel application of independent component analysis. The algorithm was trained and cross validated using seven EEGs with epileptiform activity. For epileptiform events with compensation for eyeblinks, the sensitivity was 65 +/- 22% at a specificity of 86 +/- 7% (mean +/- SD). With feature extraction by PCA or classification of raw data, specificity reduced to 76 and 74%, respectively, for the same sensitivity. On exactly the same data, the commercially available software Reveal had a maximum sensitivity of 30% and concurrent specificity of 77%. Our algorithm performed well at detecting epileptiform events in this preliminary test and offers a flexible tool that is intended to be generalized to the simultaneous classification of many waveforms in the EEG.

8

Objective: To assess the factorial validity of the Portuguese version of the Maslach Burnout Inventory - Human Services Survey (MBI-HSS). Methods: Between November 2010 and November 2011 a Portuguese version of the MBI-HSS was applied to 151 Portuguese family doctors (55% women, median age 54 years). The factorial structure of the MBI-HSS was examined by principal component analysis (PCA) and confirmatory factor analysis (CFA). Internal consistency estimates of the MBI-HSS were determined with Cronbach's alpha. Results: The fit of the hypothesized three-factor model to the data was superior to the alternative two-factor and four-factor models. CFA supported MBI-HSS as an acceptable measure to evaluate burnout and deletion of items 12 and 16 improved the goodness of fit of the model. In PCA, the three-factor model explained 50.58% of the variance and the four-factor model did not lead to understandable components. Item 12 was also found to be problematic in PCA. The Cronbach's alpha was satisfactory for emotional exhaustion (alpha=0.90), lack of personal accomplishment (alpha=0.73), and depersonalization (alpha=0.64). Conclusion: The Portuguese version of the MBI-HSS was found to be reliable to measure burnout among Portuguese medical doctors. We also recommend the deletion of items 12 and 16 from the MBI-HSS.

Principal curves and principal oriented points

Principal curves have been defined Hastie and Stuetzle (JASA, 1989) assmooth curves passing through the middle of a multidimensional dataset. They are nonlinear generalizations of the first principalcomponent, a characterization of which is the basis for the principalcurves definition.In this paper we propose an alternative approach based on a differentproperty of principal components. Consider a point in the space wherea multivariate normal is defined and, for each hyperplane containingthat point, compute the total variance of the normal distributionconditioned to belong to that hyperplane. Choose now the hyperplaneminimizing this conditional total variance and look for thecorresponding conditional mean. The first principal component of theoriginal distribution passes by this conditional mean and it isorthogonal to that hyperplane. This property is easily generalized todata sets with nonlinear structure. Repeating the search from differentstarting points, many points analogous to conditional means are found.We call them principal oriented points. When a one-dimensional curveruns the set of these special points it is called principal curve oforiented points. Successive principal curves are recursively definedfrom a generalization of the total variance.

Analysis of matched matrices

We consider the joint visualization of two matrices which have common rowsand columns, for example multivariate data observed at two time pointsor split accord-ing to a dichotomous variable. Methods of interest includeprincipal components analysis for interval-scaled data, or correspondenceanalysis for frequency data or ratio-scaled variables on commensuratescales. A simple result in matrix algebra shows that by setting up thematrices in a particular block format, matrix sum and difference componentscan be visualized. The case when we have more than two matrices is alsodiscussed and the methodology is applied to data from the InternationalSocial Survey Program.

Weighted Euclidean biplots

We construct a weighted Euclidean distance that approximates any distance or dissimilarity measure between individuals that is based on a rectangular cases-by-variables data matrix. In contrast to regular multidimensional scaling methods for dissimilarity data, the method leads to biplots of individuals and variables while preserving all the good properties of dimension-reduction methods that are based on the singular-value decomposition. The main benefits are the decomposition of variance into components along principal axes, which provide the numerical diagnostics known as contributions, and the estimation of nonnegative weights for each variable. The idea is inspired by the distance functions used in correspondence analysis and in principal component analysis of standardized data, where the normalizations inherent in the distances can be considered as differential weighting of the variables. In weighted Euclidean biplots we allow these weights to be unknown parameters, which are estimated from the data to maximize the fit to the chosen distances or dissimilarities. These weights are estimated using a majorization algorithm. Once this extra weight-estimation step is accomplished, the procedure follows the classical path in decomposing the matrix and displaying its rows and columns in biplots.

Multivariate analysis and geostatistics of the fertility of a humic rhodic hapludox under coffee cultivation

The spatial variability of soil and plant properties exerts great influence on the yeld of agricultural crops. This study analyzed the spatial variability of the fertility of a Humic Rhodic Hapludox with Arabic coffee, using principal component analysis, cluster analysis and geostatistics in combination. The experiment was carried out in an area under Coffea arabica L., variety Catucai 20/15 - 479. The soil was sampled at a depth 0.20 m, at 50 points of a sampling grid. The following chemical properties were determined: P, K+, Ca2+, Mg2+, Na+, S, Al3+, pH, H + Al, SB, t, T, V, m, OM, Na saturation index (SSI), remaining phosphorus (P-rem), and micronutrients (Zn, Fe, Mn, Cu and B). The data were analyzed with descriptive statistics, followed by principal component and cluster analyses. Geostatistics were used to check and quantify the degree of spatial dependence of properties, represented by principal components. The principal component analysis allowed a dimensional reduction of the problem, providing interpretable components, with little information loss. Despite the characteristic information loss of principal component analysis, the combination of this technique with geostatistical analysis was efficient for the quantification and determination of the structure of spatial dependence of soil fertility. In general, the availability of soil mineral nutrients was low and the levels of acidity and exchangeable Al were high.

Data-driven clustering : new methods and applications

Abstract : This work is concerned with the development and application of novel unsupervised learning methods, having in mind two target applications: the analysis of forensic case data and the classification of remote sensing images. First, a method based on a symbolic optimization of the inter-sample distance measure is proposed to improve the flexibility of spectral clustering algorithms, and applied to the problem of forensic case data. This distance is optimized using a loss function related to the preservation of neighborhood structure between the input space and the space of principal components, and solutions are found using genetic programming. Results are compared to a variety of state-of--the-art clustering algorithms. Subsequently, a new large-scale clustering method based on a joint optimization of feature extraction and classification is proposed and applied to various databases, including two hyperspectral remote sensing images. The algorithm makes uses of a functional model (e.g., a neural network) for clustering which is trained by stochastic gradient descent. Results indicate that such a technique can easily scale to huge databases, can avoid the so-called out-of-sample problem, and can compete with or even outperform existing clustering algorithms on both artificial data and real remote sensing images. This is verified on small databases as well as very large problems. Résumé : Ce travail de recherche porte sur le développement et l'application de méthodes d'apprentissage dites non supervisées. Les applications visées par ces méthodes sont l'analyse de données forensiques et la classification d'images hyperspectrales en télédétection. Dans un premier temps, une méthodologie de classification non supervisée fondée sur l'optimisation symbolique d'une mesure de distance inter-échantillons est proposée. Cette mesure est obtenue en optimisant une fonction de coût reliée à la préservation de la structure de voisinage d'un point entre l'espace des variables initiales et l'espace des composantes principales. Cette méthode est appliquée à l'analyse de données forensiques et comparée à un éventail de méthodes déjà existantes. En second lieu, une méthode fondée sur une optimisation conjointe des tâches de sélection de variables et de classification est implémentée dans un réseau de neurones et appliquée à diverses bases de données, dont deux images hyperspectrales. Le réseau de neurones est entraîné à l'aide d'un algorithme de gradient stochastique, ce qui rend cette technique applicable à des images de très haute résolution. Les résultats de l'application de cette dernière montrent que l'utilisation d'une telle technique permet de classifier de très grandes bases de données sans difficulté et donne des résultats avantageusement comparables aux méthodes existantes.

Ambulatory assessment of 3D ground reaction force using plantar pressure distribution.

This study aimed to use the plantar pressure insole for estimating the three-dimensional ground reaction force (GRF) as well as the frictional torque (T(F)) during walking. Eleven subjects, six healthy and five patients with ankle disease participated in the study while wearing pressure insoles during several walking trials on a force-plate. The plantar pressure distribution was analyzed and 10 principal components of 24 regional pressure values with the stance time percentage (STP) were considered for GRF and T(F) estimation. Both linear and non-linear approximators were used for estimating the GRF and T(F) based on two learning strategies using intra-subject and inter-subjects data. The RMS error and the correlation coefficient between the approximators and the actual patterns obtained from force-plate were calculated. Our results showed better performance for non-linear approximation especially when the STP was considered as input. The least errors were observed for vertical force (4%) and anterior-posterior force (7.3%), while the medial-lateral force (11.3%) and frictional torque (14.7%) had higher errors. The result obtained for the patients showed higher error; nevertheless, when the data of the same patient were used for learning, the results were improved and in general slight differences with healthy subjects were observed. In conclusion, this study showed that ambulatory pressure insole with data normalization, an optimal choice of inputs and a well-trained nonlinear mapping function can estimate efficiently the three-dimensional ground reaction force and frictional torque in consecutive gait cycle without requiring a force-plate.

Soil penetration resistance analysis by multivariate and geostatistical methods

The penetration resistance (PR) is a soil attribute that allows identifies areas with restrictions due to compaction, which results in mechanical impedance for root growth and reduced crop yield. The aim of this study was to characterize the PR of an agricultural soil by geostatistical and multivariate analysis. Sampling was done randomly in 90 points up to 0.60 m depth. It was determined spatial distribution models of PR, and defined areas with mechanical impedance for roots growth. The PR showed a random distribution to 0.55 and 0.60 m depth. PR in other depths analyzed showed spatial dependence, with adjustments to exponential and spherical models. The cluster analysis that considered sampling points allowed establishing areas with compaction problem identified in the maps by kriging interpolation. The analysis with main components identified three soil layers, where the middle layer showed the highest values of PR.

Integrated multivariate analysis to evaluate effects of pre-slaughter handling on pork quality

The aim of this study was to investigate the effect of pre-slaughter handling on the occurrence of PSE (Pale, Soft, and Exudative) meat in swine slaughtered at a commercial slaughterhouse located in the metropolitan region of Dourados, Mato Grosso do Sul, Brazil. Based on the database (n=1,832 carcasses), it was possible to apply the integrated multivariate analysis for the purpose of identifying, among the selected variables, those of greatest relevance to this study. Results of the Principal Component Analysis showed that the first five components explained 89.28% of total variance. In the Factor Analysis, the first factor represented the thermal stress and fatiguing conditions for swine during pre-slaughter handling. In general, this study indicated the importance of the pre-slaughter handling stages, evidencing those of greatest stress and threat to animal welfare and pork quality, which are transport time, resting period, lairage time before unloading, unloading time, and ambience.

Multivariate statistical analysis to support the minimum streamflow regionalization

ABSTRACT This study aimed to develop a methodology based on multivariate statistical analysis of principal components and cluster analysis, in order to identify the most representative variables in studies of minimum streamflow regionalization, and to optimize the identification of the hydrologically homogeneous regions for the Doce river basin. Ten variables were used, referring to the river basin climatic and morphometric characteristics. These variables were individualized for each of the 61 gauging stations. Three dependent variables that are indicative of minimum streamflow (Q7,10, Q90 and Q95). And seven independent variables that concern to climatic and morphometric characteristics of the basin (total annual rainfall – Pa; total semiannual rainfall of the dry and of the rainy season – Pss and Psc; watershed drainage area – Ad; length of the main river – Lp; total length of the rivers – Lt; and average watershed slope – SL). The results of the principal component analysis pointed out that the variable SL was the least representative for the study, and so it was discarded. The most representative independent variables were Ad and Psc. The best divisions of hydrologically homogeneous regions for the three studied flow characteristics were obtained using the Mahalanobis similarity matrix and the complete linkage clustering method. The cluster analysis enabled the identification of four hydrologically homogeneous regions in the Doce river basin.

Application of exploratory factor analysis to assess fish consumption in a university community

The objective of this research was to use the technique of Exploratory Factor Analysis (EFA) for the adequacy of a tool for the assessment of fish consumption and the characteristics involved in this process. Data were collected during a campaign to encourage fish consumption in Brazil with the voluntarily participation of members of a university community. An assessment instrument consisting of multiple-choice questions and a five-point Likert scale was designed and used to measure the importance of certain attributes that influence the choice and consumption of fish. This study sample was composed of of 224 individuals, the majority were women (65.6%). With regard to the frequency of fish consumption, 37.67% of the volunteers interviewed said they consume the product two or three times a month, and 29.6% once a week. The Exploratory Factor Analysis (EFA) was used to group the variables; the extraction was made using the principal components and the rotation using the Quartimax method. The results show clusters in two main constructs, quality and consumption with Cronbach Alpha coefficients of 0.75 and 0.69, respectively, indicating good internal consistency.

Factor Analysis of a Large DSGE Model

We study the workings of the factor analysis of high-dimensional data using artificial series generated from a large, multi-sector dynamic stochastic general equilibrium (DSGE) model. The objective is to use the DSGE model as a laboratory that allow us to shed some light on the practical benefits and limitations of using factor analysis techniques on economic data. We explain in what sense the artificial data can be thought of having a factor structure, study the theoretical and finite sample properties of the principal components estimates of the factor space, investigate the substantive reason(s) for the good performance of di¤usion index forecasts, and assess the quality of the factor analysis of highly dissagregated data. In all our exercises, we explain the precise relationship between the factors and the basic macroeconomic shocks postulated by the model.

Quantification de la relation séquence-activité de l’ARN par prédiction de structure tridimensionnelle

Dans un premier temps, nous avons modélisé la structure d’une famille d’ARN avec une grammaire de graphes afin d’identifier les séquences qui en font partie. Plusieurs autres méthodes de modélisation ont été développées, telles que des grammaires stochastiques hors-contexte, des modèles de covariance, des profils de structures secondaires et des réseaux de contraintes. Ces méthodes de modélisation se basent sur la structure secondaire classique comparativement à nos grammaires de graphes qui se basent sur les motifs cycliques de nucléotides. Pour exemplifier notre modèle, nous avons utilisé la boucle E du ribosome qui contient le motif Sarcin-Ricin qui a été largement étudié depuis sa découverte par cristallographie aux rayons X au début des années 90. Nous avons construit une grammaire de graphes pour la structure du motif Sarcin-Ricin et avons dérivé toutes les séquences qui peuvent s’y replier. La pertinence biologique de ces séquences a été confirmée par une comparaison des séquences d’un alignement de plus de 800 séquences ribosomiques bactériennes. Cette comparaison a soulevée des alignements alternatifs pour quelques unes des séquences que nous avons supportés par des prédictions de structures secondaires et tertiaires. Les motifs cycliques de nucléotides ont été observés par les membres de notre laboratoire dans l'ARN dont la structure tertiaire a été résolue expérimentalement. Une étude des séquences et des structures tertiaires de chaque cycle composant la structure du Sarcin-Ricin a révélé que l'espace des séquences dépend grandement des interactions entre tous les nucléotides à proximité dans l’espace tridimensionnel, c’est-à-dire pas uniquement entre deux paires de bases adjacentes. Le nombre de séquences générées par la grammaire de graphes est plus petit que ceux des méthodes basées sur la structure secondaire classique. Cela suggère l’importance du contexte pour la relation entre la séquence et la structure, d’où l’utilisation d’une grammaire de graphes contextuelle plus expressive que les grammaires hors-contexte. Les grammaires de graphes que nous avons développées ne tiennent compte que de la structure tertiaire et négligent les interactions de groupes chimiques spécifiques avec des éléments extra-moléculaires, comme d’autres macromolécules ou ligands. Dans un deuxième temps et pour tenir compte de ces interactions, nous avons développé un modèle qui tient compte de la position des groupes chimiques à la surface des structures tertiaires. L’hypothèse étant que les groupes chimiques à des positions conservées dans des séquences prédéterminées actives, qui sont déplacés dans des séquences inactives pour une fonction précise, ont de plus grandes chances d’être impliqués dans des interactions avec des facteurs. En poursuivant avec l’exemple de la boucle E, nous avons cherché les groupes de cette boucle qui pourraient être impliqués dans des interactions avec des facteurs d'élongation. Une fois les groupes identifiés, on peut prédire par modélisation tridimensionnelle les séquences qui positionnent correctement ces groupes dans leurs structures tertiaires. Il existe quelques modèles pour adresser ce problème, telles que des descripteurs de molécules, des matrices d’adjacences de nucléotides et ceux basé sur la thermodynamique. Cependant, tous ces modèles utilisent une représentation trop simplifiée de la structure d’ARN, ce qui limite leur applicabilité. Nous avons appliqué notre modèle sur les structures tertiaires d’un ensemble de variants d’une séquence d’une instance du Sarcin-Ricin d’un ribosome bactérien. L’équipe de Wool à l’université de Chicago a déjà étudié cette instance expérimentalement en testant la viabilité de 12 variants. Ils ont déterminé 4 variants viables et 8 létaux. Nous avons utilisé cet ensemble de 12 séquences pour l’entraînement de notre modèle et nous avons déterminé un ensemble de propriétés essentielles à leur fonction biologique. Pour chaque variant de l’ensemble d’entraînement nous avons construit des modèles de structures tertiaires. Nous avons ensuite mesuré les charges partielles des atomes exposés sur la surface et encodé cette information dans des vecteurs. Nous avons utilisé l’analyse des composantes principales pour transformer les vecteurs en un ensemble de variables non corrélées, qu’on appelle les composantes principales. En utilisant la distance Euclidienne pondérée et l’algorithme du plus proche voisin, nous avons appliqué la technique du « Leave-One-Out Cross-Validation » pour choisir les meilleurs paramètres pour prédire l’activité d’une nouvelle séquence en la faisant correspondre à ces composantes principales. Finalement, nous avons confirmé le pouvoir prédictif du modèle à l’aide d’un nouvel ensemble de 8 variants dont la viabilité à été vérifiée expérimentalement dans notre laboratoire. En conclusion, les grammaires de graphes permettent de modéliser la relation entre la séquence et la structure d’un élément structural d’ARN, comme la boucle E contenant le motif Sarcin-Ricin du ribosome. Les applications vont de la correction à l’aide à l'alignement de séquences jusqu’au design de séquences ayant une structure prédéterminée. Nous avons également développé un modèle pour tenir compte des interactions spécifiques liées à une fonction biologique donnée, soit avec des facteurs environnants. Notre modèle est basé sur la conservation de l'exposition des groupes chimiques qui sont impliqués dans ces interactions. Ce modèle nous a permis de prédire l’activité biologique d’un ensemble de variants de la boucle E du ribosome qui se lie à des facteurs d'élongation.

Non invasive classification system of scoliosis curve types using least-squares support vector machines

Objective To determine scoliosis curve types using non invasive surface acquisition, without prior knowledge from X-ray data. Methods Classification of scoliosis deformities according to curve type is used in the clinical management of scoliotic patients. In this work, we propose a robust system that can determine the scoliosis curve type from non invasive acquisition of the 3D back surface of the patients. The 3D image of the surface of the trunk is divided into patches and local geometric descriptors characterizing the back surface are computed from each patch and constitute the features. We reduce the dimensionality by using principal component analysis and retain 53 components using an overlap criterion combined with the total variance in the observed variables. In this work, a multi-class classifier is built with least-squares support vector machines (LS-SVM). The original LS-SVM formulation was modified by weighting the positive and negative samples differently and a new kernel was designed in order to achieve a robust classifier. The proposed system is validated using data from 165 patients with different scoliosis curve types. The results of our non invasive classification were compared with those obtained by an expert using X-ray images. Results The average rate of successful classification was computed using a leave-one-out cross-validation procedure. The overall accuracy of the system was 95%. As for the correct classification rates per class, we obtained 96%, 84% and 97% for the thoracic, double major and lumbar/thoracolumbar curve types, respectively. Conclusion This study shows that it is possible to find a relationship between the internal deformity and the back surface deformity in scoliosis with machine learning methods. The proposed system uses non invasive surface acquisition, which is safe for the patient as it involves no radiation. Also, the design of a specific kernel improved classification performance.