Introdução: A hanseníase é a principal causa infecciosa de neuropatia periférica e consequentes incapacidades em todo o mundo. Seu diagnóstico, na maioria das vezes é simples, especialmente quando as clássicas lesões cutâneas estão presentes. No entanto, alguns pacientes apresentam apenas envolvimento neural (forma neural pura - PNL) transformando o seu diagnóstico em um grande desafio. Nesses casos, mesmo quando essa possibilidade é aventada, sua confirmação pode ser extremamente difícil e muitos pacientes só serão corretamente diagnosticados tardiamente, quando uma neuropatia grave e irreversível já está estabelecida. Objetivos: Analisar as características de uma série de pacientes com diagnóstico definitivo ou provável de PNL seguidos no setor de doenças neuromusculares e dermatologia no HCRP em um período de 12 anos e reconhecer o padrão de apresentação mais frequente, suas manifestações clínicas e o padrão eletroneuromiográfico. Métodos: Estudo retrospectivo, observacional, cujos critérios de inclusão foram: evidência clínica de comprometimento de nervos periféricos na ausência de lesões de pele. O diagnóstico definitivo foi estabelecido quando o Mycobacterium leprae foi identificado na biópsia de nervo, e provável quando um quadro clínico sugestivo foi associado a pelo menos um dos seguintes: anti PGL1 positivo, padrão sugestivo na biópsia (neurite granulomatosa epitelióide, infiltrado linfomomononuclear, fibrose) e/ou padrão eletroneuromiográfico sugestivo. Para avaliar a importância da duração da doença na apresentação clínica, foram considerados dois grupos de acordo com o tempo da doença, 12 meses ou menos (grupo 1) e mais de 12 meses (grupo 2). Foram comparados os sinais, os sintomas, a gravidade da doença e o padrão da EMG para delinear o quadro de apresentação. Resultados: Dos 34 pacientes incluídos no estudo, 7 tinham diagnóstico definitivo e 24 diagnóstico provável. Os sintomas de início mais frequentes foram alterações sensitivas (91,2%), em 70,6% dos casos iniciaram nos membros superiores, sendo o nervo ulnar o local mais frequente. O padrão de distribuição intradérmico exclusivo foi observado apenas no grupo 1. A alteração da sensibilidade vibratória (p=0,07), a presença de alterações motoras (p=0,03) e hipo ou areflexia em 1 ou mais nervos (p=0,03) foram mais frequentemente observadas no grupo 2. Os nervos sensitivos mais frequentemente envolvidos foram o ulnar e fibular superficial. O nervo motor mais frequentemente afetado foi o ulnar. O padrão eletroneuromiográfico mais frequente foi de uma neuropatia sensitivo motora assimétrica com reduções focais da velocidade de condução e franco predomínio sensitivo. Conclusões: A PNL se apresenta invariavelmente de maneira assimétrica e com franco predomínio sensitivo. Na maioria das vezes o início ocorre nos membros superiores, especificamente no território do nervo ulnar. Há uma predisposição ao acometimento das fibras finas nos estágios iniciais e com a evolução da doença as fibras grossas passam a também serem afetadas. Os nervos sensitivos mais frequentemente envolvidos são o ulnar seguido pelo fibular superficial. / Backgrounds: Leprosy is the main infectious cause of peripheral neuropathy and disabilities in the world. Its diagnosis is straightforward when the classical skin lesions are present. However, some patients present only neural involvement (pure neural form-PNL) turning its diagnosis on a great challenge. Additionally, even when this possibility is suspected, confirmation may be extremely difficult and many patients are only correctly diagnosed late on the clinical course of the disease when a severe and irreversible neuropathy is already established. Objectives: To review the characteristics of a series of PNL patients followed in our institution in the last 12 years and recognize the clinical manifestations. Methods: Inclusion criteria: Clinical evidence of peripheral nerve impairment with no skin lesions. PNL diagnose were classified as definitive when the Mycobacterium leprae was identified under nerve biopsy, and probable when a suggestive clinical picture was associated to at least one of the following: positive anti PGL1, suggestive pattern biopsy represented by the presence of epithelioid granulomatous neuritis, mononuclear cell endoneuritis and fibrose and/or an EMG pattern showing a predominantly sensory mononeuritis multiplex pattern. Exclusion criteria: Two patients were excluded because of associated diabetes mellitus, one because had CMT1A and another had HNPP. To evaluate the importance of disease duration in clinical presentation, we considered two groups according to the time course, first that disease duration of 12 or fewer months (group 1) and those with disease duration over 12 months (group 2). Results: We reviewed 34 patients with PNL, including 7 with a definite and 24 with probable diagnosis. The most common onset symptoms were sensory (91.2 %), in 70.6 % of cases symptoms started in the upper limbs, the ulnar nerve being the most frequent site. Intradermal pattern was observed only in group 1. Vibration was altered more frequent in group 2 (p=0.07), the presence of motor abnormalities (p = 0.03) and deep tendon reflexes reduced or absent in 1 or more nerves (p = 0.03) were more frequently observed in group 2. Sensory nerves most frequently involved were the ulnar and superficial peroneal. The motor nerve most often affected was the ulnar. The most frequent EMG pattern was an asymmetrical sensory and motor neuropathy with focal slowing of conduction velocity. Conclusions: PNL is an asymmetrical sensory or sensory motor neuropathy. Upper limbs are most frequent affected with frequent ulnar nerve territory involvement. Small fibers seem to be affected at early stages. Larger fibers are affected with disease progression. It is unclear whether the PNL represents a stage prior to the appearance of typical skin lesions or whether it represents a different and more aggressive leprosy type. Phenotype characterization from early signs and symptoms its a powerful tool to PNL early diagnosis.

Made available in DSpace on 2014-06-12T18:08:18Z (GMT). No. of bitstreams: 3 arquivo8047_1.pdf: 1992919 bytes, checksum: 57d424bed65e6b9de5a2871e338a68ec (MD5) arquivo8047_2.pdf: 10283231 bytes, checksum: f15b859b779fd47ee1dedfaf5d8331b5 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2002 / Neste trabalho, estudamos o modelo de rede neural para áreas do neocórtex cerebral humano, proposto por L. Ingber (1992) para explicar a atividade neocortical de duração entre centenas de milisegundos e vários segundos. O modelo permite a simulação de grandes regiões neocorticais, pelo fato de resumir em cada nó centenas de graus de liberdade correspondentes a vários neurônios. O conjunto de suas unidades básicas, chamadas mesocolunas, é modelado através de um autômato celular que evolui no tempo de acordo com regras locais estocásticas, determinadas por interações de tipo não-linear entre os nós através de conexões limitadas (de alcance finito) entre eles, uma constatação decorrente das pesquisas biológicas experimentais

Master of Science / Department of Geology / Matthew W. Totten / The Meramec member of Mississippian age is a proficient oil and gas producing formation within the midcontinent region of the United States. It is produced in Kansas, Oklahoma, and Texas. In Kansas, 12% of the state’s petroleum production comes from Mississippian-aged rocks. Bindley Field, located in central west Kansas, has produced 3,669,283 barrels of oil from one facies within the M2 interval of the Meramec formation. This facies is a grain-supported echinoderm/bryozoan dolostone, of variable thickness. Its sporadic occurrence in the subsurface has made exploring Bindley Field and the surrounding area difficult. The challenge in finding oil in this area is in locating a producible zone of this productive facies. Previously, Bindley Field has been the subject of detailed reservoir characterization studies (Ebanks et al., 1977; Johnson, 1990; Johnson, 1994). These studies helped to contribute to a better understanding of Meramecian stratigraphy in Kansas. The Meramec was divided into four major depositional sequences, with some of those sequences nonexistent in the subsurface, due to aerial exposure and erosion post-deposition. The Meramecian units were further separated into parasequence-scale chronostratigraphic units based on marine flooding events. The primary producing interval in Bindley Field is the Meramec 2 interval which consists of seven lithotypes, and is recognized to have six, meter-scale depositional cycles (Johnson, 1990). As production from this interval increased, more information became available about controls on reservoir quality. There are still areas, however, where core data do not exist, and predicting the productive facies remains challenging. The aim of this study is to create a workflow for evaluating the subsurface using regional core and log data from Bindley Field to create a model of the subsurface distribution of the reservoir facies, which could be extended to data poor areas. Geophysical logs (neutron, gamma ray, guard) along with an artificial neural network (ANN), was used to create an accurate prediction of producing intervals within the subsurface. Values are derived from wire line log data and used to develop the ANN definition of facies distribution within Bindley Field. The ANN model was examined for accuracy and precision using core description and well cuttings from wells within Bindley Field and the surrounding area. Correlations were found between the subsurface geometry of the study area, and the production of oil and gas within the study area. An ANN model with an accuracy of 72% was achieved and applied to wells surrounding the Bindley Field, where reservoir intervals have not been as extensively studied. A total of 87 wells in Bindley Field and the surrounding 50 square mile area where applied to the ANN model. The model predicted that the productive facies thickens gradually to the northwest of Bindley Field. Cross sections as well as an isopach map were created using the prediction data from the ANN. Finally, an analysis for the accuracy of the ANN and the predicted facies was created. The productive facies yielded an accuracy value of 77%.

Historically, artificial neural networks have been loosely defined as biologically inspired computational models. When deciding what sort of network to use for a given task there are two things that need to be considered. The first is the representational power of the given network, that is what class of problems can be solved by this network? Given a set of problems to be solved by neural networks, a network that can solve any of these problems is called a universal approximator. The second problem is the ability to find a desired network given an initial network via a learning rule. Here we are interested in the question of universal approximation. A general definition of artificial neural networks is provided along with definitions for different kinds of universal approximation. We then prove that the recurrent temporal restricted Boltzmann machine (RTRBM) satisfies a general type of universal approximation for stochastic processes, an extention of previous results for the simple RBM. We conclude by examining the potential use of such temporal artificial neural networks in the biological process of perception. / Graduate

Please read the abstract in the section 00front of this document / Dissertation (MSc (Computer Science))--University of Pretoria, 2007. / Computer Science / unrestricted

Robotic simulators for use in Evolutionary Robotics (ER) have certain challenges associated with the complexity of their construction and the accuracy of predictions made by these simulators. Such robotic simulators are often based on physics models, which have been shown to produce accurate results. However, the construction of physics-based simulators can be complex and time-consuming. Alternative simulation schemes construct robotic simulators from empirically-collected data. Such empirical simulators, however, also have associated challenges, such as that some of these simulators do not generalize well on the data from which they are constructed, as these models employ simple interpolation on said data. As a result of the identified challenges in existing robotic simulators for use in ER, this project investigates the potential use of Artificial Neural Networks, henceforth simply referred to as Neural Networks (NNs), as alternative robotic simulators. In contrast to physics models, NN-based simulators can be constructed without needing an explicit mathematical model of the system being modeled, which can simplify simulator development. Furthermore, the generalization capabilities of NNs suggest that NNs could generalize well on data from which these simulators are constructed. These generalization abilities of NNs, along with NNs’ noise tolerance, suggest that NNs could be well-suited to application in robotics simulation. Investigating whether NNs can be effectively used as robotic simulators in ER is thus the endeavour of this work. Since not much research has been done in employing NNs as robotic simulators, many aspects of the experimental framework on which this dissertation reports needed to be carefully decided upon. Two robot morphologies were selected on which the NN simulators created in this work were based, namely a differentially steered robot and an inverted pendulum robot. Motion tracking and robotic sensor logging were used to acquire data from which the NN simulators were constructed. Furthermore, custom code was written for almost all aspects of the study, namely data acquisition for NN training, the actual NN training process, the evolution of robotic controllers using the created NN simulators, as well as the onboard robotic implementations of evolved controllers. Experimental tests performed in order to determine ideal topologies for each of the NN simulators developed in this study indicated that different NN topologies can lead to large differences in training accuracy. After performing these tests, the training accuracy of the created simulators was analyzed. This analysis showed that the NN simulators generally trained well and could generalize well on data not presented during simulator construction. In order to validate the feasibility of the created NN simulators in the ER process, these simulators were subsequently used to evolve controllers in simulation, similar to controllers developed in related studies. Encouraging results were obtained, with the newly-evolved controllers allowing real-world experimental robots to exhibit obstacle avoidance and light-approaching behaviour with a reasonable degree of success. The created NN simulators furthermore allowed for the successful evolution of a complex inverted pendulum stabilization controller in simulation. It was thus clearly established that NN-based robotic simulators can be successfully employed as alternative simulation schemes in the ER process.

The drive for autonomy in manufacturing is making increasing demands on control systems, both for improved performance and for extra flexibility. This is reflected in the research and development of autonomously guided vehicles which must operate safely in ill-defined, complex and time-varying environments. Traditional control systems generally make infeasible assumptions which limit their application within this domain, and therefore current research has concentrated on Intelligent Control techniques in order to make the control systems flexible and robust. An integral part of intelligence is the ability to learn from a systems interaction with its environment, and this thesis provides a unified description of several adaptive neural and fuzzy networks. The recent resurgence of interest in these two anthropomorphic techniques has seen these algorithms widely applied within learning control systems, although a firm theoretical framework which can compare different networks and establish convergence and stability conditions has not evolved. Such results are essential if these adaptive algorithms are to be used in real-world applications where safety and correctness are prime concerns. The work described in this thesis addresses these questions by introducing a class of systems called associative memory networks, which is used to describe the similarities and differences which exist between certain fuzzy and neural algorithms. All of the networks can be implemented within a 3-layer structure, where the output is linearly dependent on a set of adjustable parameters. This allows parameter convergence to be established when a gradient descent training rule is used, and the rate of convergence can be directly related to the condition of the network's basis functions. The size, shape and position of these basis functions gives each network its own specific modelling attributes, since the learning rules are identical. Therefore it is important to study the network's internal representation as this provides information about how each network generalises (both interpolation and extrapolation), the rate of parameter convergence and the type of nonlinear functions which can be successfully modelled. Three networks are described in detail: the Albus CMAC, the is given of the Albus CMAC which illustrates its desirable features for on-line, nonlinear adaptive modelling and control: local learning and a computational cost which depends linearly on the input space dimension. The modelling capabilities of the algorithm are rigorously analysed and it is shown that they are strongly dependent on the generalisation parameter, and a set of consistency equations is derived which specify how the network generalises. The adaptive B-spline network, which embodies a piecewise polynomial representation, is also described and used for nonlinear modelling and constructing a static rule base which guides and autonomous vehicle into a parking slot. B-splines are also used for on-line, constrained trajectory generation where they approximate a set of velocity or positional subgoals. Fuzzy systems are typically ill-defined, although the approach taken in this thesis is to use algebraic rather than truncation operators and smooth fuzzy sets which means that the modelling capabilities of the fuzzy network can be determined exactly, and convergence and stability results can be derived for these algorithms. These results focus research on the learning, modelling and representational abilities of the networks by providing a common framework for their analysis. The desirable features of the networks (local learning, linearly dependent on the parameter set, fuzzy interpretation) are emphasised, and the algorithms are all evaluated on a common time series prediciton problem.

The similarity between objects is judged in a wide variety of contexts from visual search to categorization to face recognition. There is a correspondingly rich history of similarity research, including empirical work and theoretical models. However, the field lacks an account of the real time neural processing dynamics of different similarity judgment behaviors. Some accounts focus on the lower-level processes that support similarity judgments, but they do not capture a wide range of canonical behaviors, and they do not account for the moment-to-moment stability and interaction of realistic neural object representations. The goal of this dissertation is to address this need and present a broadly applicable and neurally implemented model of object similarity judgments. I accomplished this by adapting and expanding an existing neural process model of change detection to capture a set of canonical, task-general similarity judgment behaviors. Target behaviors to model were chosen by reviewing the similarity judgment literature and identifying prominent and consistent behavioral effects. I tested each behavior for task-generality across three experiments using three diverse similarity judgment tasks. The following behaviors observed across all three tasks served as modeling targets: the effect of feature value comparisons, attentional modulation of feature dimensions, sensitivity to patterns of objects encountered over time, violations of minimality and triangle equality, and a sensitivity to circular feature dimensions like color hue. The model captured each effect. The neural processes implied by capturing these behaviors are discussed, along with the broader theoretical implications of the model and possibilities for its future expansion.

Analyzing large volumes of video data is a challenging and time-consuming task. Automating this process would very valuable, especially in ecological research where massive amounts of video can be used to unlock new avenues of ecological research into the behaviour of animals in their environments. Deep Neural Networks, particularly Deep Convolutional Neural Networks, are a powerful class of models for computer vision. When combined with Recurrent Neural Networks, Deep Convolutional models can be applied to video for frame level video classification. This research studies two datasets: penguins and seals. The purpose of the research is to compare the performance of image-only CNNs, which treat each frame of a video independently, against a combined CNN-RNN approach; and to assess whether incorporating the motion information in the temporal aspect of video improves the accuracy of classifications in these two datasets. Video and image-only models offer similar out-of-sample performance on the simpler seals dataset but the video model led to moderate performance improvements on the more complex penguin action recognition dataset.

Machine translation (MT) as a field of research has known significant advances in recent years, with the increased interest for neural machine translation (NMT). By combining deep learning with translation, researchers have been able to deliver systems that perform better than most, if not all, of their predecessors. While the general consensus regarding NMT is that it renders higher-quality translations that are overall more idiomatic, researchers recognize that NMT systems still struggle to deal with certain classic difficulties, and that their performance may vary depending on their architecture. In this project, we implement a challenge-set based approach to the evaluation of examples of three main NMT architectures: convolutional neural network-based systems (CNN), recurrent neural network-based (RNN) systems, and attention-based systems, trained on the same data set for English to French translation. The challenge set focuses on a selection of lexical and syntactic difficulties (e.g., ambiguities) drawn from literature on human translation, machine translation, and writing for translation, and also includes variations in sentence lengths and structures that are recognized as sources of difficulties even for NMT systems. This set allows us to evaluate performance in multiple areas of difficulty for the systems overall, as well as to evaluate any differences between architectures’ performance. Through our challenge set, we found that our CNN-based system tends to reword sentences, sometimes shifting their meaning, while our RNN-based system seems to perform better when provided with a larger context, and our attention-based system seems to struggle the longer a sentence becomes.