Using neural networks and support vector regression to relate marchetti dilatometer test parameters and maximum shear modulus

In the last two decades, small strain shear modulus became one of the most important geotechnical parameters to characterize soil stiffness. Finite element analysis have shown that in-situ stiffness of soils and rocks is much higher than what was previously thought and that stress-strain behaviour of these materials is non-linear in most cases with small strain levels, especially in the ground around retaining walls, foundations and tunnels, typically in the order of 10−2 to 10−4 of strain. Although the best approach to estimate shear modulus seems to be based in measuring seismic wave velocities, deriving the parameter through correlations with in-situ tests is usually considered very useful for design practice.The use of Neural Networks for modeling systems has been widespread, in particular within areas where the great amount of available data and the complexity of the systems keeps the problem very unfriendly to treat following traditional data analysis methodologies. In this work, the use of Neural Networks and Support Vector Regression is proposed to estimate small strain shear modulus for sedimentary soils from the basic or intermediate parameters derived from Marchetti Dilatometer Test. The results are discussed and compared with some of the most common available methodologies for this evaluation.

Using neural networks and support vector regression to relate marchetti dilatometer test parameters and maximum shear modulus

In the last two decades, small strain shear modulus became one of the most important geotechnical parameters to characterize soil stiffness. Finite element analysis have shown that in-situ stiffness of soils and rocks is much higher than what was previously thought and that stress-strain behaviour of these materials is non-linear in most cases with small strain levels, especially in the ground around retaining walls, foundations and tunnels, typically in the order of 10−2 to 10−4 of strain. Although the best approach to estimate shear modulus seems to be based in measuring seismic wave velocities, deriving the parameter through correlations with in-situ tests is usually considered very useful for design practice.The use of Neural Networks for modeling systems has been widespread, in particular within areas where the great amount of available data and the complexity of the systems keeps the problem very unfriendly to treat following traditional data analysis methodologies. In this work, the use of Neural Networks and Support Vector Regression is proposed to estimate small strain shear modulus for sedimentary soils from the basic or intermediate parameters derived from Marchetti Dilatometer Test. The results are discussed and compared with some of the most common available methodologies for this evaluation.

Overexpressing BDNF in Neural Stem Cells using non-viral gene delivery strategies

Dissertação para obtenção do Grau de Mestre em Biotecnologia

Majoritarian delays

This paper illustrates how delayed debt stabilizations can arise in a society without any emerging conflict of interests among its members. We argue that, under a majority voting rule, the economy may generate excessive levels of government spending and larger debts over time, and that this delay is increasing in income inequality. The intuition for this result is simple: a majority of citizens may find in delaying stabilizations a way to increase government expenditures, transferring in this way resources from the richest to the poorest citizens in the economy. This process may explain the upward trend and the difficulty to reduce public expenditures, the so called "ratchet effect."

Tbx1 and Bmp2 in the development of the ear, neural crest and pharyngeal system in mice

Dissertation presented to obtain the Ph.D degree in Biology

Uncertainty, generalization, and neural representation of relevant variables for decision making

Dissertation presented to obtain the Ph.D degree in Biology, Computational Biology.

A utilização de uma rede neural artificial para previsão da incidência da malária no município de Cantá, estado de Roraima

INTRODUÇÃO: A malária é uma doença endêmica na Amazônia Legal Brasileira, apresentando riscos diferentes para cada região. O Município de Cantá, no Estado de Roraima, apresentou para todo o período estudado, um dos maiores índices parasitários anuais do Brasil, com valor sempre maior que 50. O presente estudo visa à utilização de uma rede neural artificial para previsão da incidência da malária nesse município, a fim de auxiliar os coordenadores de saúde no planejamento e gestão dos recursos. MÉTODOS: Os dados foram coletados no site do Ministério da Saúde, SIVEP - Malária entre 2003 e 2009. Estruturou-se uma rede neural artificial com três neurônios na camada de entrada, duas camadas intermediárias e uma camada de saída com um neurônio. A função de ativação foi à sigmoide. No treinamento, utilizou-se o método backpropagation, com taxa de aprendizado de 0,05 e momentum 0,01. O critério de parada foi atingir 20.000 ciclos ou uma meta de 0,001. Os dados de 2003 a 2008 foram utilizados para treinamento e validação. Comparam-se os resultados com os de um modelo de regressão logística. RESULTADOS: Os resultados para todos os períodos previstos mostraram-se que as redes neurais artificiais obtiveram um menor erro quadrático médio e erro absoluto quando comparado com o modelo de regressão para o ano de 2009. CONCLUSÕES: A rede neural artificial se mostrou adequada para um sistema de previsão de malária no município estudado, determinando com pequenos erros absolutos os valores preditivos, quando comparados ao modelo de regressão logística e aos valores reais.

Nanostructuring silicon probes via electrodeposition: characterization of electrode coatings for acute in vivo neural recordings

Understanding how the brain works will require tools capable of measuring neuron elec-trical activity at a network scale. However, considerable progress is still necessary to reliably increase the number of neurons that are recorded and identified simultaneously with existing mi-croelectrode arrays. This project aims to evaluate how different materials can modify the effi-ciency of signal transfer from the neural tissue to the electrode. Therefore, various coating materials (gold, PEDOT, tungsten oxide and carbon nano-tubes) are characterized in terms of their underlying electrochemical processes and recording ef-ficacy. Iridium electrodes (177-706 μm2) are coated using galvanostatic deposition under different charge densities. By performing electrochemical impedance spectroscopy in phosphate buffered saline it is determined that the impedance modulus at 1 kHz depends on the coating material and decreased up to a maximum of two orders of magnitude for PEDOT (from 1 MΩ to 25 kΩ). The electrodes are furthermore characterized by cyclic voltammetry showing that charge storage capacity is im-proved by one order of magnitude reaching a maximum of 84.1 mC/cm2 for the PEDOT: gold nanoparticles composite (38 times the capacity of the pristine). Neural recording of spontaneous activity within the cortex was performed in anesthetized rodents to evaluate electrode coating performance.

Electromyography function, disability degree, and pain in leprosy patients undergoing neural mobilization treatment

INTRODUCTION: This study aimed to evaluate the effect of the neural mobilization technique on electromyography function, disability degree, and pain in patients with leprosy. METHODS: A sample of 56 individuals with leprosy was randomized into an experimental group, composed of 29 individuals undergoing treatment with neural mobilization, and a control group of 27 individuals who underwent conventional treatment. In both groups, the lesions in the lower limbs were treated. In the treatment with neural mobilization, the procedure used was mobilization of the lumbosacral roots and sciatic nerve biased to the peroneal nerve that innervates the anterior tibial muscle, which was evaluated in the electromyography. RESULTS: Analysis of the electromyography function showed a significant increase (p<0.05) in the experimental group in both the right (Δ%=22.1, p=0.013) and the left anterior tibial muscles (Δ%=27.7, p=0.009), compared with the control group pre- and post-test. Analysis of the strength both in the movement of horizontal extension (Δ%right=11.7, p=0.003/Δ%left=27.4, p=0.002) and in the movement of back flexion (Δ%right=31.1; p=0.000/Δ%left=34.7, p=0.000) showed a significant increase (p<0.05) in both the right and the left segments when comparing the experimental group pre- and post-test. The experimental group showed a significant reduction (p=0.000) in pain perception and disability degree when the pre- and post-test were compared and when compared with the control group in the post-test. CONCLUSIONS: Leprosy patients undergoing the technique of neural mobilization had an improvement in electromyography function and muscle strength, reducing disability degree and pain.

Factors associated with neural alterations and physical disabilities in patients with leprosy in São Luis, State of Maranhão, Brazil

Introduction Leprosy is a chronic infectious disease that is caused by Mycobacterium leprae. The objective of this study was to evaluate the risk factors that are associated with neural alterations and physical disabilities in leprosy patients at the time of diagnosis. Methods A prospective cross-sectional study was conducted on 155 leprosy patients who participated in a program that aimed to eliminate leprosy from São Luis, State of Maranhão. Results Patients who were 31-45 years of age, were older than 60 years of age or had a partner were more likely to have a disability. Patients with partners were 1.14 times more likely (p = 0.025) to have disabilities of the hands. The frequency of disabilities in the feet among the patients with different clinical forms of leprosy was statistically significant. Conclusions The identification of risk factors that are associated with neural alterations and physical disabilities in leprosy patients is important for diagnosing the disease because this approach enables physicians to plan and prioritize actions for the treatment and monitoring of patients.

Predicting market movement with Artificial Neural Networks

This paper presents an application of an Artificial Neural Network (ANN) to the prediction of stock market direction in the US. Using a multilayer perceptron neural network and a backpropagation algorithm for the training process, the model aims at learning the hidden patterns in the daily movement of the S&P500 to correctly identify if the market will be in a Trend Following or Mean Reversion behavior. The ANN is able to produce a successful investment strategy which outperforms the buy and hold strategy, but presents instability in its overall results which compromises its practical application in real life investment decisions.

Using artificial neural networks to generate trading signals for crude oil, copper and gold futures

In this thesis, a feed-forward, back-propagating Artificial Neural Network using the gradient descent algorithm is developed to forecast the directional movement of daily returns for WTI, gold and copper futures. Out-of-sample back-test results vary, with some predictive abilities for copper futures but none for either WTI or gold. The best statistically significant hit rate achieved was 57% for copper with an absolute return Sharpe Ratio of 1.25 and a benchmarked Information Ratio of 2.11.

Neural correlates of face familiarity in institutionally reared children with distinctive, atypical social behavior

Although the impact of early adverse experience on neural processing of face familiarity has been studied, research has not taken into account disordered child behavior. This work compared the neural processing of familiar versus strangers' faces in 47 institutionalized children with a mean age of 54 months to determine the effects of (a) the presence versus absence of atypical social behavior and (b) inhibited versus indiscriminant atypical behavior. Results revealed a pattern of cortical hypoactivation in institutionalized children manifesting atypical social behavior and that inhibited children displayed larger neural response to a caregiver's face than to the stranger's, while indiscriminant children did not discriminate between stimuli. These findings suggest that neural correlates of face familiarity are associated with social functioning in institutionalized children.

Abstract computation in schizophrenia detection through artificial neural network based systems

Schizophrenia stands for a long-lasting state of mental uncertainty that may bring to an end the relation among behavior, thought, and emotion; that is, it may lead to unreliable perception, not suitable actions and feelings, and a sense of mental fragmentation. Indeed, its diagnosis is done over a large period of time; continuos signs of the disturbance persist for at least 6 (six) months. Once detected, the psychiatrist diagnosis is made through the clinical interview and a series of psychic tests, addressed mainly to avoid the diagnosis of other mental states or diseases. Undeniably, the main problem with identifying schizophrenia is the difficulty to distinguish its symptoms from those associated to different untidiness or roles. Therefore, this work will focus on the development of a diagnostic support system, in terms of its knowledge representation and reasoning procedures, based on a blended of Logic Programming and Artificial Neural Networks approaches to computing, taking advantage of a novel approach to knowledge representation and reasoning, which aims to solve the problems associated in the handling (i.e., to stand for and reason) of defective information.