Classificador máquina de suporte vetorial com análise de Fourier aplicada em dados de EEG e EMG

The classifier support vector machine is used in several problems in various areas of knowledge. Basically the method used in this classier is to end the hyperplane that maximizes the distance between the groups, to increase the generalization of the classifier. In this work, we treated some problems of binary classification of data obtained by electroencephalography (EEG) and electromyography (EMG) using Support Vector Machine with some complementary techniques, such as: Principal Component Analysis to identify the active regions of the brain, the periodogram method which is obtained by Fourier analysis to help discriminate between groups and Simple Moving Average to eliminate some of the existing noise in the data. It was developed two functions in the software R, for the realization of training tasks and classification. Also, it was proposed two weights systems and a summarized measure to help on deciding in classification of groups. The application of these techniques, weights and the summarized measure in the classier, showed quite satisfactory results, where the best results were an average rate of 95.31% to visual stimuli data, 100% of correct classification for epilepsy data and rates of 91.22% and 96.89% to object motion data for two subjects.

Dispersões sólidas de sinvastatina: preparação, caracterização, no estado sólido utilizando técnicas emergentes e estudo de estabilidade

This thesis aimed to assess the increase in solubility of simvastatin (SINV) with solid dispersions using techniques such as kneading (MA), co-solvent evaporation (ES), melting carrier (FC) and spray dryer (SD). Soluplus (SOL), PEG 6000 (PEG), PVP K-30 (PVP) e sodium lauryl sulphate (LSS) were used as carriers. The solid dispersions containing PEG [PEG-2(SD)], Soluplus [SOL-2(MA)] and sodium lauryl sulphate [LSS-2(ES)] were presented with a greater increase in solubility (5.02, 5.60 and 5.43 times respectively); analyses by ANOVA between the three groups did not present significant difference (p<0.05). In the phase solubility study, the calculation of the Gibbs free energy (ΔG) revealed that the spontaneity of solubilisation of SINV occurred in the order SOL>PEG >PVP 75%>LSS, always 80%. The phase diagrams of PEG and LSS presented solubilization stoichiometry of type 1:1 (type AL). The diagrams with PVP and SOL tend to 1:2 stoichiometry (type AL + AP). The stability coefficients (Ks) of the phase diagrams revealed that the most stable reactions occurred with LSS and PVP. The solid dispersions were characterized by Fourier transform infrared (FTIR), differential scanning calorimetry (DSC), scanning electron microscopy (SEM), particle size distribution (PSD), near-infrared spectroscopy imaging (NIR-CI) and X-ray diffraction of the powder using the Topas software (PDRX-TOPAS). The solid dispersion PEG-2(SD) presented the greatest homogeneity and the lowest degree of crystallinity (18.2%). The accelerated stability study revealed that the solid dispersions are less stable than SINV, with PEG-2(SD) being the least stable, confirmed by FTIR and DSC. The analyses by PDRX-TOPAS revealed the amorphous character of the dispersions and the mechanism of increasing solubility

Análise e classificação de imagens de lesões da pele por atributos de cor, forma e textura utilizando máquina de vetor de suporte

The skin cancer is the most common of all cancers and the increase of its incidence must, in part, caused by the behavior of the people in relation to the exposition to the sun. In Brazil, the non-melanoma skin cancer is the most incident in the majority of the regions. The dermatoscopy and videodermatoscopy are the main types of examinations for the diagnosis of dermatological illnesses of the skin. The field that involves the use of computational tools to help or follow medical diagnosis in dermatological injuries is seen as very recent. Some methods had been proposed for automatic classification of pathology of the skin using images. The present work has the objective to present a new intelligent methodology for analysis and classification of skin cancer images, based on the techniques of digital processing of images for extraction of color characteristics, forms and texture, using Wavelet Packet Transform (WPT) and learning techniques called Support Vector Machine (SVM). The Wavelet Packet Transform is applied for extraction of texture characteristics in the images. The WPT consists of a set of base functions that represents the image in different bands of frequency, each one with distinct resolutions corresponding to each scale. Moreover, the characteristics of color of the injury are also computed that are dependants of a visual context, influenced for the existing colors in its surround, and the attributes of form through the Fourier describers. The Support Vector Machine is used for the classification task, which is based on the minimization principles of the structural risk, coming from the statistical learning theory. The SVM has the objective to construct optimum hyperplanes that represent the separation between classes. The generated hyperplane is determined by a subset of the classes, called support vectors. For the used database in this work, the results had revealed a good performance getting a global rightness of 92,73% for melanoma, and 86% for non-melanoma and benign injuries. The extracted describers and the SVM classifier became a method capable to recognize and to classify the analyzed skin injuries

Sistema inteligente para diagnóstico de patologias na laringe utilizando máquinas de vetor de suporte

The human voice is an important communication tool and any disorder of the voice can have profound implications for social and professional life of an individual. Techniques of digital signal processing have been used by acoustic analysis of vocal disorders caused by pathologies in the larynx, due to its simplicity and noninvasive nature. This work deals with the acoustic analysis of voice signals affected by pathologies in the larynx, specifically, edema, and nodules on the vocal folds. The purpose of this work is to develop a classification system of voices to help pre-diagnosis of pathologies in the larynx, as well as monitoring pharmacological treatments and after surgery. Linear Prediction Coefficients (LPC), Mel Frequency cepstral coefficients (MFCC) and the coefficients obtained through the Wavelet Packet Transform (WPT) are applied to extract relevant characteristics of the voice signal. For the classification task is used the Support Vector Machine (SVM), which aims to build optimal hyperplanes that maximize the margin of separation between the classes involved. The hyperplane generated is determined by the support vectors, which are subsets of points in these classes. According to the database used in this work, the results showed a good performance, with a hit rate of 98.46% for classification of normal and pathological voices in general, and 98.75% in the classification of diseases together: edema and nodules

Análise de superfícies seletivas de freqüência para aplicações em ondas milimétricas

This work consists in the development of a theoretical and numerical analysis for frequency selective surfaces (FSS) structures with conducting patch elements, such as rectangular patches, thin dipoles and cross dipoles, on anisotropic dielectric substrates. The analysis is developed for millimeter wave band applications. The analytical formulation is developed in the spectral domain, by using a rigorous technique known as equivalent transmission line method, or immitance approach. The numerical analysis is completed through the use of the Galerkin's technique in the Fourier transform domain, using entire-domain basis functions. In the last decades, several sophisticated analytical techniques have been developed for FSS structure applications. Within these applications, it can be emphasized the use of FSS structures on reflecting antennas and bandpass radomes. In the analysis, the scattered fields of the FSS geometry are related to the surface induced currents on the conducting patches. After the formulation of the scattering problem, the numerical solution is obtained by using the moment method. The choice of the basis functions plays a very important role in the numerical efficiency of the numerical method, once they should provide a very good approximation to the real current distributions on the FSS analyzed structure. Thereafter, the dyadic Green's function components are obtained in order to evaluate the basis functions unknown coefficients. To accomplish that, the Galerkin's numerical technique is used. Completing the formulation, the incident fields are determined through the incident potential, and as a consequence the FSS transmission and reflection characteristics are determined, as function of the resonant frequency and structural parameters. The main objective of this work was to analyze FSS structures with conducting patch elements, such as thin dipoles, cross dipoles and rectangular patches, on anisotropic dielectric substrates, for high frequency applications. Therefore, numerical results for the FSS structure main characteristics were obtained in the millimeter wave bando Some of these FSS characteristics are the resonant

Análise de superfícies seletivas de freqüência através do método dos potenciais vetorias de Hertz

This work presents a theoretical and numerical analysis of Frequency Selective Surfaces (FSS) with elements as rectangular patch, thin dipole and crossed dipole mounted on uniaxial anisotropic dielectric substrate layers for orientations of the optical axis along x, y and z directions. The analysis of these structures is accomplished by combination of the Hertz vector potentials method and the Galerkin's technique, in the Fourier transform-domain, using entire¬domain basis functions. This study consists in the use of one more technique for analysis of FSS on anisotropic dielectric substrate. And presents as the main contribution the introduction of one more project parameter to determinate the transmission and reflection characteristics of periodic structures, from the use of anisotropic dielectric with orientations of the crystal optical axis along x, y and z directions. To validate this analysis, the numerical results of this work are compared to those obtained by other authors, for FSS structures on anisotropic and isotropic dielectric substrates. Also are compared experimental results and the numerical correspondent ones for the FSS isotropic case. The technique proposed in this work is accurate and efficient. ln a second moment, curves are presented for the transmission and reflection characteristics of the FSS structures using conducting patch elements mounted on uniaxial anisotropic dielectric substrate layers with optical axis oriented along x, y and z directions. From analysis of these curves, the performance of the considered FSS structures as function of the optical axis orientation is described

Análise sísmica usando transformada de Curvelet

Oil prospecting is one of most complex and important features of oil industry Direct prospecting methods like drilling well logs are very expensive, in consequence indirect methods are preferred. Among the indirect prospecting techniques the seismic imaging is a relevant method. Seismic method is based on artificial seismic waves that are generated, go through the geologic medium suffering diffraction and reflexion and return to the surface where they are recorded and analyzed to construct seismograms. However, the seismogram contains not only actual geologic information, but also noise, and one of the main components of the noise is the ground roll. Noise attenuation is essential for a good geologic interpretation of the seismogram. It is common to study seismograms by using time-frequency transformations that map the seismic signal into a frequency space where it is easier to remove or attenuate noise. After that, data is reconstructed in the original space in such a way that geologic structures are shown in more detail. In addition, the curvelet transform is a new and effective spectral transformation that have been used in the analysis of complex data. In this work, we employ the curvelet transform to represent geologic data using basis functions that are directional in space. This particular basis can represent more effectively two dimensional objects with contours and lines. The curvelet analysis maps real space into frequencies scales and angular sectors in such way that we can distinguish in detail the sub-spaces where is the noise and remove the coefficients corresponding to the undesired data. In this work we develop and apply the denoising analysis to remove the ground roll of seismograms. We apply this technique to a artificial seismogram and to a real one. In both cases we obtain a good noise attenuation

Métodos numéricos para resolução de equações diferenciais ordinárias lineares baseados em interpolação por spline

In this work we have elaborated a spline-based method of solution of inicial value problems involving ordinary differential equations, with emphasis on linear equations. The method can be seen as an alternative for the traditional solvers such as Runge-Kutta, and avoids root calculations in the linear time invariant case. The method is then applied on a central problem of control theory, namely, the step response problem for linear EDOs with possibly varying coefficients, where root calculations do not apply. We have implemented an efficient algorithm which uses exclusively matrix-vector operations. The working interval (till the settling time) was determined through a calculation of the least stable mode using a modified power method. Several variants of the method have been compared by simulation. For general linear problems with fine grid, the proposed method compares favorably with the Euler method. In the time invariant case, where the alternative is root calculation, we have indications that the proposed method is competitive for equations of sifficiently high order.