Nesta tese apresentamos uma metodologia de mapeamento capaz de gerar representações em termos de grafos para sistemas biológicos de conectividade complexa. Tais sistemas são inicialmente armazenados na forma de imagens digitais e em seguida submetidos a um pré-processamento com objetivo de padronizar as imagens. As imagens pré-processadas são então utilizadas para gerar modelos tridimensionais dos sistemas de interesse. Um algoritmo de propagação de rótulos é utilizado para extrair os esqueletos dos modelos volumétricos e estes esqueletos são por fim, representados por um grafo, composto por vértices e arestas. Os vértices e arestas desse grafo armazenam propriedades do sistema original, como posição, comprimento e diâmetro, bem como as características topológicas de tais sistemas. Finalmente, os grafos resultantes são estudados através da teoria das redes complexas, dentro de um contexto específico para cada sistema. Nossos procedimentos foram aplicados com sucesso a diferentes sistemas biológicos, como artérias caríotidas, árvores arteriais, estruturas mitocondriais e poros em amostras de solo. / In the present work, we developed a mapping methodology able to build a graph representation for biological branched systems. Initially, such systems are stored as digital images and then they undergo a pre-processing in order to standardize the images. The pre-processed imagens are used to build tridimensional models of the interested systems. A label-propagation algorithm is used to extract the skeleton from the volumetric models and these skeletons are then represented by a graph, composed by nodes and edges. The nodes and edges of these graphs store properties of the original system, such as spatial position, lengths and diameter, as well as the topological features of such systems. Finally, the graphs are studied by using the complex networks theory within a specific context for each system. Our procedures were apllied sucefully to different biological systems, such as carotid artery, arterial trees, mitocondrial structure and pores in soil samples.

Nesta tese apresentamos uma metodologia de mapeamento capaz de gerar representações em termos de grafos para sistemas biológicos de conectividade complexa. Tais sistemas são inicialmente armazenados na forma de imagens digitais e em seguida submetidos a um pré-processamento com objetivo de padronizar as imagens. As imagens pré-processadas são então utilizadas para gerar modelos tridimensionais dos sistemas de interesse. Um algoritmo de propagação de rótulos é utilizado para extrair os esqueletos dos modelos volumétricos e estes esqueletos são por fim, representados por um grafo, composto por vértices e arestas. Os vértices e arestas desse grafo armazenam propriedades do sistema original, como posição, comprimento e diâmetro, bem como as características topológicas de tais sistemas. Finalmente, os grafos resultantes são estudados através da teoria das redes complexas, dentro de um contexto específico para cada sistema. Nossos procedimentos foram aplicados com sucesso a diferentes sistemas biológicos, como artérias caríotidas, árvores arteriais, estruturas mitocondriais e poros em amostras de solo. / In the present work, we developed a mapping methodology able to build a graph representation for biological branched systems. Initially, such systems are stored as digital images and then they undergo a pre-processing in order to standardize the images. The pre-processed imagens are used to build tridimensional models of the interested systems. A label-propagation algorithm is used to extract the skeleton from the volumetric models and these skeletons are then represented by a graph, composed by nodes and edges. The nodes and edges of these graphs store properties of the original system, such as spatial position, lengths and diameter, as well as the topological features of such systems. Finally, the graphs are studied by using the complex networks theory within a specific context for each system. Our procedures were apllied sucefully to different biological systems, such as carotid artery, arterial trees, mitocondrial structure and pores in soil samples.

There is no legislation in Sweden that forces municipalities to implement democracy-developing measures. Municipalities and regions are thus given a great deal of personal responsibility to implement this type of measure. Previous research shows that there are great variations in the number of democray-developing measuers between Sweden's municipalities. This means that inequality is introduced between residents in different municipalities in terms of the impact on local policy.The ambition of this study was thus to find out why this variation prevails. The question for the study is • What factors can explain the variation in the prevalence of democracy-developing work in Swedish municipalities and what proportion of the variation can these factors explain?By quantitatively testing three different factors that could have an impact on the number of democracy-developing measures, an answer to the question of the study can be obtained. The three factors were economy, political ideology and its stability and geographical network. After compiling statistics from these three factors in relation to the number of democracy-developing measures in the Swedish municipalities, results from the statistics program SPSS could be analyzed. This shows that these three factors together account for an explanatory degree between 8.7 to 33.4 percent of the variance.