Cette thèse porte sur la segmentation d'images OCT cutanées. Cette modalité d'imagerie permet de visualiser les structures superficielles avec une profondeur de l'ordre du millimètre. En dermatologie, elle permet d'explorer l'épiderme et sa jonction avec le derme. Cependant, les images OCT sont sévèrement affectées par le bruit speckle. Ce phénomène conjugué à la complexité inhérente aux structures de la peau rend l'interprétation des images difficile même pour des experts. L'approche classique consiste à filtrer le speckle avant de faire des traitements de segmentation. A l'inverse, dans cette thèse nous exploitons exclusivement le bruit comme information pour segmenter. Notre approche repose sur la modélisation statistique du speckle. La segmentation se fait par classification des paramètres de ce modèle probabiliste. Ainsi, - On montre que le speckle ne suit pas une loi Rayleigh, comme cela est établi analytiquement. - On ajuste plusieurs lois de probabilité à l'amplitude OCT; et on montre que celle-ci est distribuée selon la loi Gamma généralisée. - On établit que les paramètres de cette loi discriminent statistiquement les couches de la peau. - On conçoit une méthode de segmentation par classification des paramètres locaux du speckle. Les nombreuses expérimentations faites sur plusieurs corpus d'images in-vivo confirment la validité de notre approche. / This thesis deals with the segmentation of skin OCT images. This modality provides the means to visualize superficial structures down to a millimeter depth. In dermatology, it is used to examine the epidermis and its junction with the dermis. However, OCT images are severely affected by the speckle noise. This random phenomenon added to the complexity of human skin structures makes the visual interpretation of images very difficult. Classical image processing techniques filter this noise prior to any segmentation step. In this thesis, we rely exclusively on the properties of the speckle to perform segmentation. Our approach is based on the statistical modeling of the speckle. Images are segmented by classifying parameters of the statistical model. Therefore, - We show that speckle does not follow the Rayleigh distribution, as developed analytically in the literature. - We fit various probability laws to model OCT signal amplitude ; we show that Generalized Gamma has the best goodness of fit. - We establish that statistical parameters of this distribution discriminate skin layers with good variability. - We develop a segmentation method based on the classification of local statistical parameters. The various experimental results with a number of in-vivo images reported in the thesis confirm the validity of our approach

Discrete-time survival analysis as an educational research technique has focused on analysing and interpretating parameter estimates. The purpose of this study was to examine the effects of certain data characteristics on the hazard estimates and goodness of fit statistics. Fifty-four simulated data sets were crossed with four conditions in a 2 (time period) by 3 (distribution of Y = 1) by 3 (distribution of Y = 0) by 3 (sample size) design.

In radiation therapy, it is imperative to deliver high doses of radiation to the tumor while reducing radiation to the healthy tissue. Respiratory motion is the most significant source of errors during treatment. Therefore, it is essential to accurately model respiratory motion for precise and effective radiation delivery. Many approaches exist to account for respiratory motion, such as controlled breath hold and respiratory gating, and they have been relatively successful. They still present many drawbacks. Thus, research has been expanded to tumor tracking. The overall goal of 4D-CT is to predict tumor motion in real time, and this work attempts to move in that direction. The following work addresses both the temporal and the spatial aspects of four-dimensional CT reconstruction. The aims of the paper are to (1) estimate the temporal parameters of 4D models for anatomy deformation using a novel neural network approach and (2) to use intelligently chosen non-uniform, non-separable splines to improve the spatial resolution of the deformation models in image registration.

This thesis deals with the estimation of model parameters of the interacting segments process in plane. The motivation is application on the system of stress fibers in human mesenchymal stem cells, which are detected by fluorescent microscopy. The model of segments is defined as a spatial Gibbs point process with marks. We use two methods for parameter estimation: moment method and Takacs-Fiksel method. Further, we implement algorithm for these estimation methods in software Mathematica. Also we are able to simulate the model structure by Markov Chain Monte Carlo, using birth-death process. Numerical results are presented for real and simulated data. Match of model and data is considered by descriptive statistics. Powered by TCPDF (www.tcpdf.org)

Induction motors (IM) find widespread use in modern industry and for this reason they have been subject to a significant amount of research interest in recent times. One particular aspect of this research is the fault detection and diagnosis (FDD) of induction motors for use in a condition based maintenance (CBM) strategy; by effectively tracking the condition of the motor, maintenance action need only be carried out when necessary. This type of maintenance strategy minimises maintenance costs and unplanned downtime. The benefits of an effective FDD for IM is clear and there have been numerous studies in this area but few which consider the problem in a practical sense with the aim of developing a single system that can be used to monitor motor condition under a range of different conditions, with different motor specifications and loads. This thesis aims to address some of these problems by developing a general FDD system for induction motor. The solution of this problem involved the development and testing of a new approach; the adaptive mixed-residual approach (AMRA). The main aim of the AMRA system is to avoid the vast majority of unplanned failures of the machine and therefore as opposed to tackling a single induction motor fault, the system is developed to detect all four of the most statistically prevalent induction motor fault types; rotor fault, stator fault, air-gap fault and bearing fault. The mixed-residual fault detection algorithm is used to detect these fault types which includes a combination of spectral and model-based techniques coupled with particle swarm optimisation (PSO) for automatic identification of motor parameters. The AMRA residuals are analysed by a fuzzy-logic classifier and the system requires only current and voltage inputs to operate. Validation results indicate that the system performs well under a range of load torques and different coupling methods proving it to have significant potential for use in industrial applications.

Orientador: Sandra Aidar de Queiroz / Coorientador: William Koury Filho / Coorientador: Marcos Eli Buzanskas / Banca: Roberto Carvalheiro / Banca: Patrícia Tholon / Resumo: Neste estudo, foram estimadas herdabilidades e correlações genéticas, fenotípicas e ambientais para as características estrutura, precocidade e musculosidade à desmama (ED, PD, MD) e ao sobreano (ES, PS e MS), peso padronizado aos 210 e 450 dias de idade (P210 e P450, respectivamente), idade ao primeiro parto (IPP), probabilidade de parto precoce (3P) e habilidade de permanência no rebanho (STAY) em bovinos da raça Nelore participantes de programa de melhoramento genético. O conjunto de dados utilizado foi composto por registros de 37.826 animais e pedigree contendo 88.213 animais. O método dos quadrados mínimos foi utilizado para definição de efeitos fixos considerados nos modelos de estimação de parâmetros. Foram testados quatro modelos de estimação, em análises uni-característica por meio do método da máxima verossimilhança restrita, sob modelo animal. Para as características mensuradas ao desmame, verificou-se a necessidade de inclusão dos efeitos genético materno e de ambiente permanente materno no modelo de estimação. Para as características medidas ao sobreano e características reprodutivas, o modelo foi composto pelo efeito aleatório genético direto de animal e residual. As estimativas dos parâmetros genéticos, ambientais e fenotípicos foram obtidas em análises uni e bi-características, por meio de análise bayesiana. Estimativas de herdabilidade média, obtidas por análises bi-característica, foram iguais a 0,28, 0,30, 0,27 e 0,28, para ED, PD, MD e P210, respectivamente. Para as características mensuradas ao sobreano, as estimativas médias de herdabilidade variaram entre 0,40 e 0,37 (ES), 0,44 e 0,42 (PS), 0,39 e 0,37 (MS) e 0,50 e 0,48 (P450). As características medidas em diferentes idades (desmame e sobreano) apresentaram altas correlações genéticas, sendo iguais a 0,96 entre ED e ES, PD e PS e P210 e P450; e 0,94 entre MD e MS. As herdabilidades das características reprodutivas foram... / Abstract: The aim of this study was to estimate the heritability and the genetic, phenotypic and environmental correlations for visual scores (structure, precocity and musculature at weaning and yearling), body weight at 210 and 450 days of age, age at first calving, heifer pregnancy and stayability in Nelore beef cattle participating in a breeding program. The data set consisted on information of 37,826 animals and 88,213 animals in the pedigree. The least square method was used to define the fixed effects for parameters estimation. For traits measured at weaning, four estimation models were tested using single-trait analysis, by means of the restricted maximum likelihood method, using the animal model. Was considered additive and residual random effects (Model 1); additive, maternal and residual random effects (Model 2); additive, maternal permanent environment and residual random effects (Model 3); and additive, maternal, maternal permanent environment and residual random effects (Model 4). Model 1 was used for yearling and reproductive traits. The genetic parameters were determined by single and two-trait analyses using Bayesian methodology. Estimates of average heritability obtained in two-trait analyses, were equal to 0.28, 0.30, 0.27, and 0.28, for structure, precocity and musculature at weaning and body weight at 210 days of age. For yearling traits, the average heritability estimates varied from 0.40 to 0.37 (structure), 0.44 to 0.42 (precocity), 0.39 to 0.37 (musculature), and 0.50 to 0.48 (body weight at 450 days of age). The traits measured at different ages (weaning and yearling) had high genetic correlation: 0.96 for structure, precocity and weight, and 0.94 for musculature. The mean heritability estimate for IPP, 3P and STAY were equal to 0.18, 0.38 and 0.20, respectively. The genetic correlations between visual scores and reproductive traits were favorable, indicating that when selecting animals of greater weight and higher ... / Mestre

Orientador: Lucia Galvão de Albuquerque / Coorientador: Gerardo Alves Fernandes Júnior / Banca: Henrique Nunes de Oliveira / Banca: Raphael Bermal Costa / Resumo: A habilidade de permanência (HP), que pode ser definida como a probabilidade de uma vaca parir numa determinada idade, dado que ela teve esta oportunidade, é uma característica reprodutiva de grande importância em bovinos de corte, estando diretamente relacionada à produtividade do rebanho. O objetivo deste trabalho foi estimar parâmetros genéticos e identificar possíveis regiões do genoma que estejam associadas com a expressão fenotípica da HP de vacas da raça Nelore. Os componentes de variância foram estimados por inferência bayesiana utilizando um modelo de limiar no qual foram considerados os efeitos sistemáticos de grupos de contemporâneos e precocidade sexual, e os efeitos aleatórios de animal e resíduo.Os efeitos dos SNPs foram estimados por meio da metodologia ssGBLUP, sendo utilizados na análise 2838 animais genotipados com o painel de alta densidade da Illumina (Bovine HD Assay Illumina, San Diego, CA, USA). A variância explicada por janelas formadas por 200 SNPs consecutivos foi utilizada na identificação das regiões de maior efeito sobre a expressão da característica HP. A herdabilidade encontrada utilizando a matriz A (pedigree) foi de 0,11±0,01 e utilizando a matriz H( matriz de parentesco que combina a informação de pedigree e dos SNPs) de 0,14±0,01. Foram encontrados 147 genes candidatos para a característica habilidade de permanência em regiões dos cromossomos 1, 2, 5, 6, 9, 20 e no cromossomo sexual X. Novas regiões candidatas para a característica habilida... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Stayability, which can be defined as the probability of a cow calving to a certain age since it has this opportunity, is a reproduction trait of great importance in beef cattle and is directly related to the productivity of the herd. The objective of this study was to estimate variance components through single-step G-BLUP methodology, and use the solutions of the SNPs effects to identify possible regions of the genome that are associated with the phenotypic expression of stayability in Nellore cows. The variance components were estimated by Bayesian inference using a threshold model in which were considered the fixed effects of precocity and contemporary groups, and the random effects of residue and animal. The effects of SNP were estimated by ssGBLUP methodology being used for the analysis 2838 animals genotyped with the Illumina panel of high density ( HD Assay Bovine Illumina, San Diego, CA, USA ). A Manhattan plot containing the variance explained by the windows formed by 200 SNPs consecutive was used to identify regions of greatest effect on the expression of stayability trait. The heritability found using the A matrix was 0.11 ± 0.01 and using the H matrix was 0.14 ± 0.01. 147 potential genes for stayability were found in regions of chromosomes 1,2,5,6,9,20 and sexual chromosome X. New candidate regions for stayability were detected and most of them are related to reproductive, immune and nervous system. / Mestre

This work employs statistical and Bayesian techniques to analyze mathematical forward models with several sources of uncertainty. The forward models usually arise from phenomenological and physical phenomena and are expressed through regression-based models or partial differential equations (PDEs) associated with uncertain parameters and input data. One of the critical challenges in real-world applications is to quantify uncertainties of the unknown parameters using observations. To this purpose, methods based on the likelihood function, and Bayesian techniques constitute the two main statistical inferential approaches considered here. Two problems are studied in this thesis. The first problem is the prediction of fatigue life of metallic specimens. The second part is related to inverse problems in linear PDEs. Both problems require the inference of unknown parameters given certain measurements. We first estimate the parameters by means of the maximum likelihood approach. Next, we seek a more comprehensive Bayesian inference using analytical asymptotic approximations or computational techniques. In the fatigue life prediction, there are several plausible probabilistic stress-lifetime (S-N) models. These models are calibrated given uniaxial fatigue experiments. To generate accurate fatigue life predictions, competing S-N models are ranked according to several classical information-based measures. A different set of predictive information criteria is then used to compare the candidate Bayesian models. Moreover, we propose a spatial stochastic model to generalize S-N models to fatigue crack initiation in general geometries. The model is based on a spatial Poisson process with an intensity function that combines the S-N curves with an averaged effective stress that is computed from the solution of the linear elasticity equations.

Orientador: Vera Fernanda Martins Hossepian de Lima / Coorientador: Danísio Prado Munari / Banca: João Ademir de Oliveira / Banca: Sabrina Luzia Caetano / Resumo: As estimativas de parâmetros genéticos das características relacionadas à produção de ovos auxiliam no processo de seleção de linhagem de aves poedeiras. O objetivo deste trabalho foi estimar os parâmetros genéticos para as características de desempenho e reprodutivas, para fornecer suporte ao programa de melhoramento de poedeiras. As características reprodutivas estudadas foram registros de fertilidade (FERT), eclodibilidade (ECLO) e taxa de nascimento (NASC) medidas às 60 semanas de idade, e de desempenho foram peso corporal às 16 e 60 semanas de idade (PC16 e PC60), idade à maturidade sexual (IMS), relação alturalargura, peso e densidade dos ovos às 28, 36 e 40 semanas de idade (REL28, REL36, REL40, POVO28, POVO36, POVO40, DENS28, DENS36 e DENS40, respectivamente). Foram utilizados dados de 1894 animais, provenientes de três gerações de uma linhagem denominada CC, desenvolvida pelo Programa de Melhoramento Genético de Aves da Embrapa Suínos e Aves, em Concórdia, SC. Análises multicaracterísticas foram realizadas pela metodologia de inferência bayesiana por meio do software GIBBS2F90 para cada conjunto de características (desempenho e reprodutivas). O modelo estatístico considerou os efeitos aleatórios genético aditivo e residuais e o efeito fixo de geração. A variação das médias à posteriori das estimativas de herdabilidade para as características reprodutivas foi de 0,14±0,003 (ECLO) à 0,22±0,005 (FERT) e de 0,07±0,001 (REL28) à 0,42±0,001 (POVO40) para as características de desempenho. As médias à posteriori das estimativas de correlações genéticas entre as características reprodutivas variaram de 0,18±0,026 (FERT e ECLO) até 0,79±0,007 (FERT e NASC). Para as características de desempenho, as médias à posteriori das estimativas de correlação genética variaram de - 0,49±0,001 (POVO36 e DENS36) à 0,75±0,003 (DENS28 e DENS36). Em geral, as características poderiam ser... / Abstract: Genetic parameter estimates for traits associated to egg production help in the selection process of laying hens. The aim of this study was to estimate genetic parameters for performance and reproductive traits to provide support for an egg stock breeding program. The reproductive traits studied were fertility (FERT), hatchability (ECLO), birth rate (NASC) at 60 weeks of age and the performance traits were body weight at 16 and 60 weeks of age (PC16 and PC60), age at first egg (IMS), eggs height-width ratio, weight and density at 28, 36 and 40 weeks of age (REL28, REL36, REL40, POVO28, POVO36, POVO40, DENS28, DENS36 and DENS40, respectively). Data from 1894 animals from three generations of line of laying hens, developed by the Embrapa Poultry Breeding Program, Concordia, SC, was used. Analyses multi-trait were conducted by Bayesian inference through GIBBS2F90 software. The statistical model included the fixed effects of generation and additive and residual random. The posterior means of heritability for reproductive traits was 0.14 ± 0.003 (ECLO) to 0.22 ± 0.005 (FERT) and 0.07 ± 0.001 (REL28) to 0.42 ± 0.001 (POVO40) for the performance characteristics. The posterior means of genetic correlation estimates between reproductive traits ranged from 0.18 ± 0.026 (FERT and ECLO) to 0.79 ± 0.007 (FERT and NAAT). For performance traits, the posterior means of genetic correlation ranged from -0.49 ± 0.001 (POVO36 and DENS36) to 0.75 ± 0.003 (DENS28 and DENS36). In general, the traits could be used in animal breeding programs of laying hens because they have enough genetic variability to respond to selection, except for REL28. It is concluded that the applied selection on the birth rate would be most appropriate to improve fertility and hatchability of eggs. Also, the use of POVO40 could improve egg weight or uniformity thereof. If the selection is conducted to reduce body weight at 16 weeks of age, result in increased age at first egg ... / Mestre

Orientador: Sandra Aidar de Queiroz / Coorientador: Patrícia Tholon / Banca: Claudia Cristina Paro de Paz / Banca: Danísio Prado Munari / Abstract: Temperament is a trait of notable economic relevance, due to its association with management and productive trait, and can be used as a selection criterion of cattle, requiring identification of environmental factors that influence, as well as the verification of genetic variability and the interrelationship with the other characteristics of interest to the production system. The objective of this study was to estimate genetic, phenotypic and environmental parameters using four measures of temperament evaluated on Canchim cattle. The temperament measures used were: flight speed (TS), composite score of temperament (ECT), total score of temperament (ET) and reactivity (REAT). This study analyzed 2090 information of Canchim cattle from the Canchim farm belonging to Embrapa Pecuária Sudeste, located at São Carlos, SP. The temperament traits were analyzed using an animal model, through bi-trait analyzes. For all traits, the contemporaries group (CG) was formed by animals born in the same year and belonging to the same sex. The evaluations were carried out during the weaning, yearling and long yearling phases. The temperament measures showed heritability estimates ranging from 0,07 (0,142) (TS at weaning) to 0,36 (0,240) (TS at yearling) values of moderate to low magnitude that were influenced by management practices and environmental conditions, being important knowledge such factors for selection actions. The genetic correlation estimates were higher for ETxREAT (1.00 at long ye... (Complete abstract click electronic access below) / Resumo: As medidas de temperamento possuem notória relevância econômica devido a sua associação com características de interesse produtivo. Essas medidas podem ser utilizadas como critérios de seleção de bovinos, requerendo identificação de fatores ambientais que as influenciem, assim como a verificação de variabilidade genética e a inter-relação destas com as demais características de interesse ao sistema de produção. O objetivo deste estudo foi estimar parâmetros genéticos, fenotípicos e ambientais de quatro medidas de avaliação do temperamento de bovinos da raça Canchim. As medidas de temperamento utilizadas foram: tempo de saída (TS), escore composto de temperamento (ECT), escore total (ET) e reatividade (REAT). Neste trabalho foram analisadas 2090 informações de bovinos da raça Canchim provenientes da fazenda Canchim, pertencente à Embrapa Pecuária Sudeste, situada na cidade de São Carlos, SP. As características foram analisadas utilizando-se um modelo animal, por meio de análises bi-característica. Para todas as características, o grupo de contemporâneos (GC) foi formado por animais nascidos no mesmo ano e pertencentes ao mesmo sexo. As avaliações foram realizadas durante as fases de desmama, ano e sobreano. As medidas de temperamento apresentaram estimativas de herdabilidade que variaram de 0,07 (0,142) (TS na desmama) a 0,36 (0,240) (TS, ao sobreano). As baixas magnitudes das estimativas indicam que as características são influenciadas principalmente por práticas de manejo e... (Resumo completo, clicar acesso eletrônico abaixo) / Mestre