Spectral-domain optical coherence tomography (SD-OCT) is a non-invasive imaging modality that allows for the quantitative study of retinal structures. SD-OCT has begun to find widespread use in the diagnosis and management of various ocular diseases. While commercial scanners provide limited analysis of a small number of retinal layers, the automated segmentation of retinal layers and other structures within these volumetric images is quite a challenging problem, especially in the presence of disease-induced changes. The incorporation of a priori information, ranging from qualitative assessments of the data to automatically learned features, can significantly improve the performance of automated methods. Here, a combined machine learning-based approach and graph-theoretic approach is presented for the automated segmentation of retinal structures in SD-OCT images. Machine-learning based approaches are used to learn textural features from a training set, which are then incorporated into the graph- theoretic approach. The impact of the learned features on the final segmentation accuracy of the graph-theoretic approach is carefully evaluated so as to avoid incorporating learned components that do not improve the method. The adaptability of this versatile combination of a machine-learning and graph-theoretic approach is demonstrated through the segmentation of retinal surfaces in images obtained from humans, mice and canines. In addition to this framework, a novel formulation of the graph-theoretic approach is described whereby surfaces with a disruption can be segmented. By incorporating the boundary of the "hole" into the feasibility definition of the set of surfaces, the final result consists of not only the surfaces but the boundary of the hole as well. Such a formulation can be used to model the neural canal opening (NCO) in SD-OCT images, which appears as a 3-D planar hole disrupting the surfaces in its vicinity. A machine-learning based approach was also used here to learn descriptive features of the NCO. Thus, the major contributions of this work include 1) a method for the automated correction of axial artifacts in SD-OCT images, 2) a combined machine-learning and graph-theoretic framework for the segmentation of retinal surfaces in SD-OCT images (applied to humans, mice and canines), 3) a novel formulation of the graph-theoretic approach for the segmentation of multiple surfaces and their shared hole (applied to the segmentation of the neural canal opening), and 4) the investigation of textural markers that could precede structural and functional change in degenerative retinal diseases.

Glaucoma is the second leading cause of blindness worldwide. The clinical standard for monitoring the functional deficits in the retina that are caused by glaucoma is the visual field test. In addition to monitoring the functional loss, evaluating the disease-related structural changes in the human retina also helps with diagnosis and management of this progressive disease. The characteristic changes of retinal structures such as the optic nerve head (ONH) are monitored utilizing imaging modalities such as color (stereo) fundus photography and, more recently, spectral-domain optical coherence tomography (SD-OCT). With the inherent subjectivity and time required for manually segmenting retinal structures, there has been a great interest in automated approaches. Since both fundus and SD-OCT images are often acquired for the assessment of glaucoma, the automated segmentation approaches can benefit from combining the multimodal complementary information from both sources. The goal of the current work is to automatically segment the retinal structures and extract the proper parameters of the optic nerve head related to the diagnosis and management of glaucoma. The structural parameters include the cup-to-disc ratio (CDR) which is a 2D parameter and is obtainable from both fundus and SD-OCT modalities. Bruch's membrane opening-minimum rim width (BMO-MRW) is a recent 3D structural parameter that is obtainable from the SD-OCT modality only. We propose to use the complementary information from both fundus and SD-OCT modalities in order to enhance the segmentation of structures of interest. In order to enable combining information from different modalities, a feature-based registration method is proposed for aligning the fundus and OCT images. In addition, our goal is to incorporate the machine-learning techniques into the graph-theoretic approach that is used for segmenting the structures of interest. Thus, the major contributions of this work include: 1) use of complementary information from SD-OCT and fundus images for segmenting the optic disc and cup boundaries in both modalities, 2) identifying the extent that accounting for the presence of externally oblique border tissue and retinal vessels in rim-width-based parameters affects structure-structure correlations, 3) designing a feature-based registration approach for registering multimodal images of the retina, and 4) developing a multimodal graph-based approach to segment the optic nerve head (ONH) structures such as Internal Limiting Membrane (ILM) surface and Bruch's membrane surface's opening.

For coupled systems of differential equations on networks, a graph-theoretic approach to the construction of Lyapunov functions is systematically developed in this thesis. Kirchhoffs Matrix-Tree Theorem in graph theory plays an essential role in the approachs development. The approach is successfully applied to several coupled systems well-known in the literature to demonstrate its applicability and effectiveness. / Applied Mathematics

No description available.

Submitted by Maike Costa (maiksebas@gmail.com) on 2016-04-29T13:19:13Z No. of bitstreams: 1 arquivo total.pdf: 1827584 bytes, checksum: 8e2c4699b20e87ffce14b44d0467564c (MD5) / Made available in DSpace on 2016-04-29T13:19:13Z (GMT). No. of bitstreams: 1 arquivo total.pdf: 1827584 bytes, checksum: 8e2c4699b20e87ffce14b44d0467564c (MD5) Previous issue date: 2016-02-02 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / To reduce waste and increase the generation of value for customers, more and more construction companies have sought to apply the Lean Construction. However, there is no unanimous formalization of the implementation process and companies lack information about their real capacity to go through a lean transition. As stressed by the literature, one of the key factors for the success of any lean implementation is to assess the readiness of the organization. This paper assumes that this assessment should consider the technical, human and cultural dimensions and their interrelationships. Seeking ways to evaluate, an important tool based on graph theory and matrix algebra, identified in the literature as Graph Theoretic Approach (GTA), has been applied in different contexts, with desirable properties such as the ability to model interactions between criteria and generating hierarchical models for solving complex decision-making problems. Thus, this study aims to develop a readiness assessment method for implementation of Lean Construction using the GTA as a tool for its operationalization. As steps involved in building the method, it was carried out a literature review, in which it was possible to identify the readiness factors and sub-factors considered in the evaluation, and the parameterization of the model, through the establishment of inter-relations between the selected factors and sub-factors. The developed method was tested through the application in construction companies. As a result from the application, it was obtained the classification of two companies which showed insufficient and partial readiness levels. From the case studies, it was concluded that the proposed method is suitable for its purpose, meeting the criteria of feasibility, usability and utility. The method results allow to provide a diagnostic of the current situation of construction companies, which serves as a driver for change initiatives towards the Lean Construction. / Visando reduzir desperdícios e aumentar a geração de valor para seus clientes finais, cada vez mais empresas construtoras tem buscado aplicar a Construção Enxuta. O que se constata é que não há uma formalização unânime do processo de implementação e que as empresas carecem de informações sobre sua real capacidade para passar por um processo de transição enxuta. Como ressalta a literatura, um dos fatores chave para o sucesso de qualquer implementação enxuta consiste em avaliar a prontidão da organização. Esta pesquisa parte do pressuposto que essa avaliação deve considerar as dimensões de ordem técnica, humana e cultural e suas inter-relações. Na busca de meios para a avaliação, uma importante ferramenta baseada na teoria dos grafos e álgebra matricial, identificada na literatura como Graph Theoretic Approach (GTA), vem sendo aplicada em diversos contextos, apresentando propriedades desejáveis como a capacidade de modelar interações de critérios e de gerar modelos hierárquicos para resolução de problemas de tomada de decisões complexas. Desta forma, este trabalho tem como objetivo desenvolver um método de avaliação de prontidão para implementação da Construção Enxuta utilizando a GTA como ferramenta para sua operacionalização. Como etapas inerentes à construção do método foi realizada uma revisão bibliográfica, na qual foi possível identificar os fatores e os subfatores de prontidão considerados na avaliação, e a parametrização do modelo, por meio do estabelecimento das inter-relações entre os fatores e os subfatores selecionados. O método desenvolvido foi testado por meio da aplicação em empresas construtoras. Como resultado da aplicação, obteve-se a classificação de duas empresas avaliadas as quais apresentaram níveis insuficientes e parciais de prontidão. A partir dos estudos de caso, foi possível concluir que o método proposto é adequado aos fins a que se destina, atendendo aos critérios de viabilidade, usabilidade e utilidade. Os seus resultados permitem fornecer um diagnóstico da situação atual das construtoras, o qual serve de balizador para iniciativas de mudança em direção à Construção Enxuta.

Submitted by ANA KARLA PEREIRA RODRIGUES (anakarla_@hotmail.com) on 2017-10-03T11:36:30Z No. of bitstreams: 1 arquivototal.pdf: 3049650 bytes, checksum: 3c12ba8f608227bcf644c23dbded37d2 (MD5) / Made available in DSpace on 2017-10-03T11:36:30Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 3049650 bytes, checksum: 3c12ba8f608227bcf644c23dbded37d2 (MD5) Previous issue date: 2017-02-22 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / The formation of business arrangements became a strategy adopted by companies to acquire combined benefits that could not be generated individually. Thus, collaboration emerges as a fundamental characteristic in setting and maintaining interorganizational relationships. Local productive arrangements (APLs), known in the international literature as clusters or industrial districts, are widely discussed in several aspects, for instance, their taxonomy. However, there are research gaps that demand new investigations, such as the collaboration between firms in this type of arrangement. There are papers that discuss the benefits that companies can achieve from collaborative interorganizational relationships, but, in addition, it is necessary to create mechanisms to assess the collaboration level in order to assist managers in decision-making. Therefore, this work aims to develop a model to assess the collaboration in companies participating in APLs, testing it through case studies in five companies of a furniture APL in the state of Paraíba. The proposed model used a support tool named Graph Theoretic Approach (GTA), which was combined with the Delphi technique applied to experts in order to define the interrelationships between attributes. The attributes from literature were classified as assessment factors and sub-factors, by the following hierarchical relationships: governance factor (constituted by sub-factors public policies, institutions support and risks and rewards sharing); geographical proximity factor (constituted by sub-factors tangible resources sharing, information sharing and knowledge sharing); and trust factor (constituted by sub-factors informal links, long-term relationship and reputation). Through qualitative-quantitative analysis, it was identified that, in general, the evaluated companies are in the average level of collaboration, according to the classification scale that was developed. The application of the model contributes to encouraging companies to develop collaborative practices for factors with lower levels. As a result, the proposed model can be considered appropriate for its purposes, allowing a general evaluation of the collaboration in companies that work in APLs. / A formação de arranjos empresarias tornou-se uma estratégia adotada pelas empresas para adquirirem benefícios conjuntos que não poderiam ser gerados individualmente. Neste sentido, a colaboração surge como uma característica fundamental no estabelecimento e na manutenção dos relacionamentos interorganizacionais. Os arranjos produtivos locais (APLs), conhecidos na literatura internacional como clusters ou industrial districts, são amplamente discutidos na academia em relação a diversos aspectos, como, por exemplo, sua taxonomia. No entanto, existem lacunas de pesquisa que apontam para novas investigações, como é o caso da colaboração entre empresas nesse tipo de arranjo. Há pesquisas que discutem sobre os benefícios que as empresas podem obter a partir das relações interorganizacionais colaborativas, mas, além disso, é necessário criar mecanismos de avaliação da colaboração que possam auxiliar os gestores na tomada de decisão. Dessa forma, este trabalho tem como principal objetivo desenvolver um modelo para avaliar a colaboração em empresas participantes de APLs, aplicando-o por meio de estudos de caso em cinco empresas do APL de móveis do Estado da Paraíba. Para o desenvolvimento do modelo, utilizou-se uma ferramenta denominada de graph theoretic approach (GTA), cuja definição das inter-relações entre os atributos de avaliação se deu por meio da aplicação da técnica Delphi, com pesquisadores especialistas sobre a temática. Os atributos levantados na literatura foram denominados de fatores e subfatores de avaliação, com as seguintes relações hierárquicas: fator governança (composto pelos subfatores suporte de políticas públicas, suporte de instituições de apoio e compartilhamento de riscos e recompensas); fator proximidade geográfica (composto pelos subfatores compartilhamento de recursos tangíveis, compartilhamento de informações e compartilhamento de conhecimento); e fator confiança (composto pelos subfatores vínculos informais, relacionamento de longo prazo e reputação). Através da análise quali-quantitativa sobre o tema investigado, verificou-se que, em geral, as empresas avaliadas encontram-se no nível médio de colaboração, de acordo com a escala de classificação adotada. A aplicação do modelo contribui para incentivar as empresas a desenvolverem práticas colaborativas para os fatores com menores índices. Foi possível concluir que o modelo proposto é adequado para os seus devidos fins, estabelecendo uma avaliação geral da colaboração em empresas que atuam em APLs.