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

Classificação de distúrbios na rede elétrica usando redes neurais e wavelets

Post dispatch analysis of signals obtained from digital disturbances registers provide important information to identify and classify disturbances in systems, looking for a more efficient management of the supply. In order to enhance the task of identifying and classifying the disturbances - providing an automatic assessment - techniques of digital signal processing can be helpful. The Wavelet Transform has become a very efficient tool for the analysis of voltage or current signals, obtained immediately after disturbance s occurrences in the network. This work presents a methodology based on the Discrete Wavelet Transform to implement this process. It uses a comparison between distribution curves of signals energy, with and without disturbance. This is done for different resolution levels of its decomposition in order to obtain descriptors that permit its classification, using artificial neural networks

Detecção e classificação de modos de operação do bombeio mecânico via cartas dinamométricas

The precision and the fast identification of abnormalities of bottom hole are essential to prevent damage and increase production in the oil industry. This work presents a study about a new automatic approach to the detection and the classification of operation mode in the Sucker-rod Pumping through dynamometric cards of bottom hole. The main idea is the recognition of the well production status through the image processing of the bottom s hole dynamometric card (Boundary Descriptors) and statistics and similarity mathematics tools, like Fourier Descriptor, Principal Components Analysis (PCA) and Euclidean Distance. In order to validate the proposal, the Sucker-Rod Pumping system real data are used

Um sistema inteligente de classificação de sinais de EEG para Interface Cérebro-Computador

The Brain-Computer Interfaces (BCI) have as main purpose to establish a communication path with the central nervous system (CNS) independently from the standard pathway (nervous, muscles), aiming to control a device. The main objective of the current research is to develop an off-line BCI that separates the different EEG patterns resulting from strictly mental tasks performed by an experimental subject, comparing the effectiveness of different signal-preprocessing approaches. We also tested different classification approaches: all versus all, one versus one and a hierarchic classification approach. No preprocessing techniques were found able to improve the system performance. Furthermore, the hierarchic approach proved to be capable to produce results above the expected by literature

Classificação de padrões através de um comitê de máquinas aprimorado por aprendizagem por reforço

Reinforcement learning is a machine learning technique that, although finding a large number of applications, maybe is yet to reach its full potential. One of the inadequately tested possibilities is the use of reinforcement learning in combination with other methods for the solution of pattern classification problems. It is well documented in the literature the problems that support vector machine ensembles face in terms of generalization capacity. Algorithms such as Adaboost do not deal appropriately with the imbalances that arise in those situations. Several alternatives have been proposed, with varying degrees of success. This dissertation presents a new approach to building committees of support vector machines. The presented algorithm combines Adaboost algorithm with a layer of reinforcement learning to adjust committee parameters in order to avoid that imbalances on the committee components affect the generalization performance of the final hypothesis. Comparisons were made with ensembles using and not using the reinforcement learning layer, testing benchmark data sets widely known in area of pattern classification

Classificação Automática de Modulação Digital com uso de Correntropia para Ambientes de Rádio Cognitivo

Modern wireless systems employ adaptive techniques to provide high throughput while observing desired coverage, Quality of Service (QoS) and capacity. An alternative to further enhance data rate is to apply cognitive radio concepts, where a system is able to exploit unused spectrum on existing licensed bands by sensing the spectrum and opportunistically access unused portions. Techniques like Automatic Modulation Classification (AMC) could help or be vital for such scenarios. Usually, AMC implementations rely on some form of signal pre-processing, which may introduce a high computational cost or make assumptions about the received signal which may not hold (e.g. Gaussianity of noise). This work proposes a new method to perform AMC which uses a similarity measure from the Information Theoretic Learning (ITL) framework, known as correntropy coefficient. It is capable of extracting similarity measurements over a pair of random processes using higher order statistics, yielding in better similarity estimations than by using e.g. correlation coefficient. Experiments carried out by means of computer simulation show that the technique proposed in this paper presents a high rate success in classification of digital modulation, even in the presence of additive white gaussian noise (AWGN)

Algoritmos genéticos aplicados a um comitê de LS-SVM em problemas de classificação

The pattern classification is one of the machine learning subareas that has the most outstanding. Among the various approaches to solve pattern classification problems, the Support Vector Machines (SVM) receive great emphasis, due to its ease of use and good generalization performance. The Least Squares formulation of SVM (LS-SVM) finds the solution by solving a set of linear equations instead of quadratic programming implemented in SVM. The LS-SVMs provide some free parameters that have to be correctly chosen to achieve satisfactory results in a given task. Despite the LS-SVMs having high performance, lots of tools have been developed to improve them, mainly the development of new classifying methods and the employment of ensembles, in other words, a combination of several classifiers. In this work, our proposal is to use an ensemble and a Genetic Algorithm (GA), search algorithm based on the evolution of species, to enhance the LSSVM classification. In the construction of this ensemble, we use a random selection of attributes of the original problem, which it splits the original problem into smaller ones where each classifier will act. So, we apply a genetic algorithm to find effective values of the LS-SVM parameters and also to find a weight vector, measuring the importance of each machine in the final classification. Finally, the final classification is obtained by a linear combination of the decision values of the LS-SVMs with the weight vector. We used several classification problems, taken as benchmarks to evaluate the performance of the algorithm and compared the results with other classifiers

Medição da densidade óssea através de ondas eletromagnéticas e classificação por MVS

This work holds the purpose of presenting an auxiliary way of bone density measurement through the attenuation of electromagnetic waves. In order to do so, an arrangement of two microstrip antennas with rectangular configuration has been used, operating in a frequency of 2,49 GHz, and fed by a microstrip line on a substrate of fiberglass with permissiveness of 4.4 and height of 0,9 cm. Simulations were done with silica, bone meal, silica and gypsum blocks samples to prove the variation on the attenuation level of different combinations. Because of their good reproduction of the human beings anomaly aspects, samples of bovine bone were used. They were subjected to weighing, measurement and microwave radiation. The samples had their masses altered after mischaracterization and the process was repeated. The obtained data were inserted in a neural network and its training was proceeded with the best results gathered by correct classification on 100% of the samples. It comes to the conclusion that through only one non-ionizing wave in the 2,49 GHz zone it is possible to evaluate the attenuation level in the bone tissue, and that with the appliance of neural network fed with obtained characteristics in the experiment it is possible to classify a sample as having low or high bone density

Arquiteturas eficientes para sensoriamento espectral e classificação automática de modulações usando características cicloestacionárias

The increasing demand for high performance wireless communication systems has shown the inefficiency of the current model of fixed allocation of the radio spectrum. In this context, cognitive radio appears as a more efficient alternative, by providing opportunistic spectrum access, with the maximum bandwidth possible. To ensure these requirements, it is necessary that the transmitter identify opportunities for transmission and the receiver recognizes the parameters defined for the communication signal. The techniques that use cyclostationary analysis can be applied to problems in either spectrum sensing and modulation classification, even in low signal-to-noise ratio (SNR) environments. However, despite the robustness, one of the main disadvantages of cyclostationarity is the high computational cost for calculating its functions. This work proposes efficient architectures for obtaining cyclostationary features to be employed in either spectrum sensing and automatic modulation classification (AMC). In the context of spectrum sensing, a parallelized algorithm for extracting cyclostationary features of communication signals is presented. The performance of this features extractor parallelization is evaluated by speedup and parallel eficiency metrics. The architecture for spectrum sensing is analyzed for several configuration of false alarm probability, SNR levels and observation time for BPSK and QPSK modulations. In the context of AMC, the reduced alpha-profile is proposed as as a cyclostationary signature calculated for a reduced cyclic frequencies set. This signature is validated by a modulation classification architecture based on pattern matching. The architecture for AMC is investigated for correct classification rates of AM, BPSK, QPSK, MSK and FSK modulations, considering several scenarios of observation length and SNR levels. The numerical results of performance obtained in this work show the eficiency of the proposed architectures

Recuperação e purificação de proteínas do soro de queijo tipo coalho usando cromatografia de troca iônica e interação hidrofóbica em leito na forma expandida

Expanded Bed Adsorption plays an important role in the downstream processing mainly for reducing costs as well as steps besides could handling cells homogenates or fermentation broth. In this work Expanded Bed Adsorption was used to recover and purify whey proteins from coalho cheese manufacture using Streamline DEAE and Streamline SP both ionic resins as well as a hydrophobic resin Streamline Phenyl. A column of 2.6 cm inner diameter with 30 cm in height was coupled to a peristaltic pump. Hydrodynamics study was carried out with the three resins using Tris-HCl buffer in concentration of 30, 50 and 70 mM, with pH ranging from 7.0 to 8.0. In this case, assays of the expansion degree as well as Residence Time Distribution (RTD) were carried out. For the recovery and purification steps, a whey sample of 200 mL, was submitted to a column with 25mL of resin previously equilibrated with Tris/HCl (50 mM, pH 7.0) using a expanded bed. After washing, elution was carried out according the technique used. For ionic adsorption elution was carried out using 100 mL of Tris/HCl (50 mM, pH 7.0 in 1M NaCl). For Hydrophobyc interaction elution was carried out using Tris/HCl (50 mM, pH 7.0). Adsorption runs were carried out using the three resins as well as theirs combination. Results showed that for hydrodynamics studies a linear fit was observed for the three resins with a correlation coefficient (R2) about 0.9. In this case, Streamline Phenyl showed highest expansion degree reaching an expansion degree (H0/H) of 2.2. Bed porosity was of 0.7 when both resins Streamline DEAE and Streamline SP were used with StremLine Phenyl showing the highest bed porosity about 0.75. The number of theorical plates were 109, 41.5 and 17.8 and the axial dipersion coefficient (Daxial) were 0.5, 1.4 and 3.7 x 10-6 m2/s, for Streamline DEAE, Streamline SP and Streamline Phenyl, respectively. Whey proteins were adsorved fastly for the three resins with equilibrium reached in 10 minutes. Breakthrough curves showed that most of proteins stays in flowthrough as well as washing steps with 84, 77 and 96%, for Streamline DEAE, Streamline SP and Streamline Phenyl, respectively. It was observed protein peaks during elution for the three resins used. According to these peaks were identified 6 protein bands that could probably be albumin (69 KDa), lactoferrin (76 KDa), lactoperoxidase (89 KDa), β-lactoglobulin (18,3 KDa) e α-lactoalbumin (14 KDa), as well as the dimer of beta-lactoglobulin. The combined system compound for the elution of Streamline DEAE applied to the Streamline SP showed the best purification of whey proteins, mainly of the α-lactoalbumina

Sistemas microemulsionados aplicados à remoção da cor de efluente têxtil

Effluent color resulting from textile dyeing processes has been one of the biggest environmental problems faced by the textile industry. In particular, reactive dyes are highly resistant to conventional wastewater treatment methods. New technologies have been contemplated, some of which have been applied in industrial treatment plants, but color removal has not been efficiently attained. Since microemulsion systems provide good results in heavy metals and proteins extraction processes, their use in dyes extraction has been suggested and investigated. In this work, a real textile wastewater from an exhaustion dyebath has been treated, which contains the following reactive dyes: Procion Yellow H-E4R (CI Reactive Yellow 84), Procion Blue H-ERD (CI Reactive Blue 160) and Procion Red H-E3B (CI Reactive Red 120), in addition to auxiliary compounds normally found in dyeing processes with reactive dyes. The dyes Remazol Blue RR and Remazol Turquoise Blue G (Reactive Blue 21) have also been examined in view of the presence of heavy metals in these molecules. The microemulsion system comprised dodecyl ammonium chloride (as a cationic surfactant), water or wastewater as aqueous phase, kerosene as oil phase, and one of the following alcohols as cosurfactant: isoamyl alcohol, n-butyl alcohol and n-octyl alcohol. The pseudo-ternary diagrams were constructed in order to define Winsor s equilibrium regions. The influence of parameters such as pH, C/S (cosurfactant/surfactant) ratio, distribution coefficient, initial dye concentration, salinity, temperature, phases relative amounts, loading capacity of the microemulsion phase and dye reextraction rate has also been investigated. An experimental planning (Scheffé Net) was used to optimize the extraction process. The removal of color and metals reached levels as high as 99%

Expressão imuno-histoquímica das proteínas MMP-9, VEGF e FVW em lesões centrais e periféricas de células gigantes

Central giant cell lesion (CGCL) and peripheral giant cell lesion (PGCL) of the jaws have a distinct clinical behavior, although they share histopathologic features. It is still unclear whether these clinical differences are supported by a distinct pattern of immunoexpression of markers for multinucleated giant cells (GC) and mononuclear cells (MC). The purpose of this study was to compare the immunohistochemical expression of VEGF, MMP-9 in CG and MC and measure the vascularization by vWF to check whether there are differences in expression of these biomarkers between CGCL and PGCL. Paraffin wax blocks of 20 cases of LCCG and 20 LPCG were retrieved. MMP-9 immunoreactivity was greater in the CM of PGCL compared to VEGF (p<0.05). VEGF expression was greater in the CM of CGCL compared to PGCL (p<0.05) and it was greater in the overall expression of CGCL compared to PGCL (p<0.05). Vascularity was quantified by microvascular counting (MVC). MVC was greater in the PGCL compared CGCL (p<0.05). MMP-9 showed a greater tendency of expression in CGCL, though was not significant (p>0.05). We tested correlation between the proteins studied in each group and found a significant negative correlation between VEGF and vWF in CGCL (p<0.05). These results suggest that there are differences in the expression of VEGF in CM and overall expression between the lesions, although no statistically significant difference in the overall expression of the MMP-9. Then, there was a trend in increased expression of MMP-9 and VEGF in CGCL, possibly by the involvement of both proteins in osteoclastogenesis. Additionally, the results of this study indicate a higher degree of vascularization in PGCL compared to CGCL, fact that can be directly linked to the reactive nature of the PGCL, where the inflammatory process with its rich angiogenesis contributes significantly to these findings.

Expressão imuno-hitoquímica das proteínas RANK, RANKL e OPG em cistos radiculares e cistos dentígeros

Receptor ativador nuclear κappa B (RANK), ligante do receptor ativador nuclear κappa B (RANKL) e osteoprotegerina (OPG) são membros da família do fator de necrose tumoral relacionados com o metabolismo ósseo. A formação, diferenciação e atividade dos osteoclastos são reguladas por estas três proteínas. RANK é um receptor transmembrana presente em diversos tipos celulares, principalmente em células de linhagem macrofágica, linfócitos, células dendríticas e fibroblastos e quando ativado pelo seu ligante, RANKL, promove a diferenciação e ativação de células osteoclásticas responsáveis pelo processo de reabsorção óssea. A OPG impede a ligação RANK/RANKL atuando como um receptor inibitório para a atividade osteolítica. O objetivo deste estudo foi comparar a expressão imuno-histoquímica destes biomarcadores em cistos radiculares (n=20) e cistos dentígeros (n=20). A expressão imuno-histoquímica destes marcadores foi avaliada no epitélio e na cápsula dos cistos por escores e percentuais médios de imunomarcação. Para o epitélio, a análise semi-quantitativa revelou um padrão similar dos escores de imunomarcação de RANK, RANKL e OPG nas lesões, não havendo diferença estatística significante (p=0.589, p=0.688, p=0.709, respectivamente). Para a cápsula cística a análise quantitativa, mostrou diferença estatística significante entre os percentuais médios de imunomarcação do RANK e RANKL (p=0,001 e p=0,005, respectivamente) nos cistos. A correlação dos escores de imunomarcação de RANKL e OPG no epitélio do CR e do CD revelou diferença estatística significante (p=0,029, p=0,003, respectivamente). No epitélio dos CRs e dos CDs observou-se uma maior imunoexpressão da OPG comparada a do RANKL. Os resultados apontam a presença de RANK, RANKL e OPG nos cistos radiculares e cistos dentígeros, sugerindo a atuação destas proteínas no desenvolvimento e expansão das lesões no osso adjacente

Análise comparativa da imunoexpressão das proteínas hmlh1 e hmsh2 em carcinomas epidermóides de lábio inferior e queilites actínicas com graus variados de displasia

Lip squamous cell carcinoma (SCC) may develop from a premalignant condition, actinic cheilitis (AC) in 95% of the cases. Both premalignant and neoplastic lip diseases are caused mainly by chronic exposure to the ultraviolet component of solar radiation, especially UVB. This exposure causes disruption of the cell cycle and damage to DNA repair systems, like mismatch repair, altering proteins repair as hMLH1 and hMSH2. This research aimed to investigate the immunohistochemical expression of hMLH1 and hMSH2 proteins in lower lip SCCs and ACs, providing additional information about carcinogenesis of the lower lip. The sample consisted 40 cases of ACs and 40 cases of lower lip SCCs. Histological sections of 3 μm were submitted to immunoperoxidase method, for immunohistochemical analysis of lesions were counted in 1000 cells (positive and negative), data were evaluated both in absolute numbers and percentage of immunostained cells, the latter by assigning scores. Associations of the variables and comparative analysis of biomarker expression were performed by Fisher s exact and Pearson s chi-square, "t" student, one-way ANOVA, Mann- Whitney e Kruskal-Wallis tests. The level of significance was 5%. It was found that, in lower lip SCC, the mean of the proteins was higher in female patients (hMLH1= 369,80 + 223,98; hMHS2 = 534,80 + 343,62), less than 50 years old (hMLH1 = 285,50 + 190,65; hMHS2 = 540,00 + 274,79) and classified as low-grade malignancy (hMLH1 = 264,59 + 179,21; hMHS2 = 519,32 + 302,58), in these data only to sex, for hMLH1 protein, was statistically significant (p=0.034). Comparing the different lesions, we observed that for both hMLH1 and hMSH2 protein, the average of positive epithelial cells decreased as the lesion was graded at later stages. The ACs classified without dysplasia or mild dysplasia had the highest average of immunostained cells (hMLH1 = 721.23 + 88.116; hMHS2 = 781.50 + 156.93). The ACs classified as moderate or severe dysplasia had intermediate values (hMLH1 = 532,86 + 197,72; hMHS2 = 611,14 + 172,48) and SSCs of the lower lip had the lowest averages (hMLH1 = 255,03 + 199,47; hMHS2 = 518,38 + 265,68). There was a statistically significant difference between groups (p<0.001). In conclusion, our data support the hypothesis that changes in immunoexpression of these proteins is related to the process of carcinogenesis of the lower lip

Expressão Imuno-histoquímica das proteínas GLUT-1 e HIF-1α em lesões vasculares de mucosa oral

The correct histological diagnosis of vascular lesions in the oral mucosa is critical, especially in defining the treatment and prognosis, as some vascular lesions show spontaneous involution and others do not show such behavior. This study analyzed the expression immunohistochemistry of human glucose transporter protein (GLUT-1), in oral benign vascular tumors and to reclassify such lesions according to with his immunoexpression. In addition, we evaluated the immunohistochemical expression of hypoxia-inducible factor 1 alpha (HIF-1α), the main transcription factor involved in cellular adaptation to hypoxia. We analyzed 60 cases of benign oral vascular tumors: 30 cases with histological diagnosis of HEM and 30 cases of oral pyogenic granuloma (PG). The results of this research showed that of the 30 lesions initially classified as HEM, only 7 showed immuno-positivity for GLUT-1, remaining with the initial diagnosis. The remaining 23 were reclassified as vascular malformation (VM) (13 cases) and PG (10 cases). All cases in the sample with an initial diagnosis of PG were negative for GLUT-1, demonstrating the accuracy of histological diagnosis of these lesions. Concerning to the immunoexpression of HIF-1α, the Mann-Whitney test revealed a statistically significant difference between the cases of GP and MV (p = 0.002), where the median of GP (m=78) was higher than the MV (m=53). Based on these results, this study showed that a histological diagnosis alone is not always sufficient for the correct diagnosis of oral HEM and that HIF-1α participates in the pathogenesis of vascular lesions