The estimation of probability density functions (PDF) of intensity values plays an important role in medical image analysis. Non-parametric PDF estimation methods have the advantage of generality in their application. The two most popular estimators in image analysis methods to perform the non-parametric PDF estimation task are the histogram and the kernel density estimator. But these popular estimators crucially need to be ‘tuned’ by setting a number of parameters and may be either computationally inefficient or need a large amount of training data. In this thesis, we critically analyse and further develop a recently proposed non-parametric PDF estimation method for signals, called the NP windows method. We propose three new algorithms to compute PDF estimates using the NP windows method. One of these algorithms, called the log-basis algorithm, provides an easier and faster way to compute the NP windows estimate, and allows us to compare the NP windows method with the two existing popular estimators. Results show that the NP windows method is fast and can estimate PDFs with a significantly smaller amount of training data. Moreover, it does not require any additional parameter settings. To demonstrate utility of the NP windows method in image analysis we consider its application to image segmentation. To do this, we first describe the distribution of intensity values in the image with a mixture of non-parametric distributions. We estimate these distributions using the NP windows method. We then use this novel mixture model to evolve curves with the well-known level set framework for image segmentation. We also take into account the partial volume effect that assumes importance in medical image analysis methods. In the final part of the thesis, we apply our non-parametric mixture model (NPMM) based level set segmentation framework to segment colorectal MR images. The segmentation of colorectal MR images is made challenging due to sparsity and ambiguity of features, presence of various artifacts, and complex anatomy of the region. We propose to use the monogenic signal (local energy, phase, and orientation) to overcome the first difficulty, and the NPMM to overcome the remaining two. Results are improved substantially on those that have been reported previously. We also present various ways to visualise clinically useful information obtained with our segmentations in a 3-dimensional manner.

As unidades de beneficiamento de minério de ouro buscam cada vez mais uma produção de baixo custo e maximização dos ganhos financeiros. A caracterização tecnológica está inserida em uma abordagem multidisciplinar que permite agregar conhecimento, alternativas de otimização e redução nos custos de operação. Inserida como uma ferramenta na caracterização tecnológica, a análise de imagens automatizada tem importante papel no setor mineral principalmente pela rapidez das análises, robustez estatística e confiabilidade nos resultados. A técnica pode ser realizada por meio de imagens adquiridas em microscópio eletrônico de varredura, associada a microanálises químicas sendo utilizada em diversas etapas de um empreendimento mineiro. Este estudo tem como objetivo a caraterização tecnológica de minério de ouro da Mina Morro do Ouro, Minas Gerais na qual foi utilizado a técnica de análise de imagens automatizada por MLA em um conjunto de 88 amostras. Foi possível identificar que 90% do ouro está na fração acima de 0,020 mm; o quartzo e mica representam cerca de 80% da massa total do minério; os sulfetos apresentam diâmetro de círculo equivalente entre 80 e 100 ?m e são representados por pirita e arsenopirita, com pirrotita, calcopirita, esfalerita e galena subordinada. Também foi possível observar que o ouro está majoritariamente associado à pirita e arsenopirita e com o aumento de teor de arsênio, cresce a parcela de ouro associado à arsenopirita. As medianas das distribuições de tamanho dos grãos de ouro apresentam um valor médio de 19 ?m. Verificou-se que a composição dos grãos de ouro é bastante diversa, em média 77% de ouro e 23% de prata. Para material abaixo de 0,50 mm observa-se uma parcela expressiva de perímetro exposto dos grãos de ouro (média 73%); o ouro incluso (21% do total dos grãos de ouro) está associado a pirita e arsenopirita, sendo que em 14 das 88 amostras este valor pode superar 40% do total de ouro contido. A ferramenta da análise de imagens automatizada mostrou-se bastante eficiente definindo características particulares o que fornece de forma objetiva subsídios para os trabalhos de planejamento de mina e processamento mineral. / Evolution in gold ore processing units are increasingly focused towards efficiency seeking routines, leading to a low-cost and maximal financial gain. The mineral characterization, inserted into a multidisciplinary approach enables a knowledge adding dynamics that results in optimization alternatives and consequent operating cost reduction. Being one of commonly used tools in mineral characterization, automated image analysis plays an important role in the mining sector mainly due to its rapid analysis, statistical robustness and reliability of the results. The technique can be performed using images acquired in a scanning electron microscope, usually associated with chemical microanalysis having applications in different stages of a mining venture. The following study aims to produce a mineral characterization using an automated image analysis technique by MLA in a set of 88 samples from the Morro do Ouro gold mine, located in Minas Gerais State, Brazil. Results showed that 90% of gold content is above the 0,02mm size fraction; quartz and mica are the mayor mineral phases identified representing roughly 80% of the total; sulfides have equivalent circle diameter ranging from 80 to 100 ?m present mainly as pyrite and arsenopyrite. It was also noted that gold is mainly associated with pyrite and arsenopyrite and that there is a correlation between the higher share of gold associated with arsenopyrite as a function of higher contents in arsenic. The median size distributions of gold grains have an average of 19 ?m. It was found that the composition of the gold is highly diverse, on average 77% gold and 23% silver. For material below 0.50 mm, was observed that a significant portion of the exposed gold grains circumference (average 73%); the included gold (21% of total grains of gold) is associated with pyrite and arsenopyrite, and that in 14 of 88 samples this value may exceed 40% of the gold contained. The automated image analysis tool proved to be very efficient defining particular features, which provides an objective subsidy for mine planning work and mineral processing.

A análise de imagens automatizada tem atualmente importante papel no setor mineral, devido à rapidez das análises e à confiabilidade dos resultados. A técnica pode ser realizada por meio de imagens adquiridas em microscópio eletrônico de varredura, geralmente associada a microanálises químicas por EDS. Estas analises podem ser realizadas em diversas etapas de um empreendimento mineiro: no auxílio à exploração, lavra (blendagens), processos de beneficiamento (associação mineral), estudos de gênese e controles de mineralização. Essa técnica permite a análise estatística de grande número de partículas/dados, fornecendo informações de composição mineralógica, química mineral, partição de elementos, associações minerais incluindo liberação de fases de interesse, entre outros. Este estudo investiga a aplicabilidade da análise de imagens automatizada em amostras de minério de ferro, com base em diferentes procedimentos de preparação de amostras e rotinas de análise avaliadas por meio do sistema MLA. Um minério de ferro foi escolhido por se tratar de um minério já bastante estudado e bem documentado na bibliografia e que mostra, em geral, mineralogia e texturas simples e repetitivas facilitando a análise dos resultados obtidos com os diferentes procedimentos e rotinas aplicadas. Além disso, o minério de ferro apresenta fases minerais com significativas diferenças de peso específico, o que pode acarretar em maiores dificuldades em termos de amostragem e preparação de amostras. / The automated image analysis has an important role in the mineral industry due to the quickly analysis and the results reliability. The technique can be applied using images from a scanning electron microscope associated usually coupled with chemical microanalysis from an EDS system. These analyses are carried out in many stages of a mining project helping mineral exploration, mining (blending), mineral processing (mineral association), genetic studies and mineralization control. This technique allows a statistic analysis of a large numbers particles and data, providing information of mineralogical composition, mineral chemistry, elemental distribution, mineral associations, phase and spectra liberation and others. This study sets the applicability of automated image analysis of an iron ore sample with diverse preparation procedures and analysis routines, which were evaluated by the MLA system. An iron ore sample was chosen because it is a familiar and well-documented ore in the literature, that usually shows simple textures and mineralogy which makes easier the interpretation of results by a diversity of preparation procedures and analysis routines. Furthermore, the iron ore shows mineral phases with significant differences in specific gravity resulting in more difficulties in sampling and preparation procedures due to potential density segregation.

Synaptic vesicle recycling is the reuse of synaptic membrane material and proteins after vesicles have been exocytosed at the pre-synaptic terminal of a neuronal synapse. The discovery of the mechanisms by which recycling operates is a subject of active research. Within small mammalian central nervous system nerve terminals, two studied mechanisms of recovery are clathrin-mediated endocytosis and activity-dependent bulk endocytosis. Research into the comparative kinetics and mechanisms underlying these endocytosis mechanisms commonly involves time-series fluorescence microscopy of in vitro cultures. Synaptic proteins are tagged with fluorescent markers, or the synaptic vesicles are labelled with fluorescent dye. The change in fluorescence levels of individual synapses over time in response to stimuli is used to understand synaptic activity. The image analysis of these time-series images frequently requires substantial manual effort to extract the changing synaptic fluorescence intensity levels over time. This work focusses on two closely interlinked areas, the development of improved automated image analysis tools to facilitate the analysis of microscopy image data, and computational simulations to leverage the data obtained from these experiments to gain mechanistic insight into the underlying processes involved in synaptic vesicle recycling. The imaged properties of synapses within the time-series images are characterised, in terms of synapse movement during the course of an experiment. This characterisation highlights the properties which risk adding error to the extracted fluorescence intensity data, as analysis generally requires segmentation of regions of interest with fixed size and location. Where possible, protocols to optimise the manual selection of synapses in the image are suggested. The manual selection of synapses within time-series images is a common but time consuming and difficult task. It requires considerable skill on the part of the researcher to select synapses from noisy images without introducing error or bias. Automated tools for either general image segmentation or for segmentation of synapse-like puncta do exist, but have mixed results when applied to time-series experiments. This work introduces the use of knowledge of the experiment protocol into the segmentation process. The selection of synapses as they respond to known stimuli is compared against other current segmentation methods, and tools to perform this segmentation are provided. This use of synapse activity improves the quality of the segmented set of synapses over existing segmentation tools. Finally, this work builds a number of computational models, to allow published individual data points to be aggregated into a coherent view of overall synaptic vesicle recycling. The first is FM-Sim, a stochastic hybrid model of overall synapse recycling as is expected to occur during the course of an experiment. This closed system model handles the processes of exocytosis and endocytosis. It uses Bayesian inference to fit model parameters to experimental data. In particular, it uses the experimental protocol to separate the mechanisms and rates that may contribute to the observed experimental data. The second is a mathematical model of one aspect of synaptic vesicle recycling of particular interest - homoeostasis of plasma membrane integrity on the presynaptic terminal. This model provides bounds on efficiency of the studied endocytosis mechanisms at recovery of plasma membrane area during and after neuronal stimulus. Both the image analysis and the computational simulations demonstrated in this work provide useful tools and insights into current research of synaptic vesicle recycling and the role of activity-dependent bulk endocytosis. In particular, the utility of adding time-dependent experimental protocol knowledge to both the image analysis tools and the computational simulations is shown.

As unidades de beneficiamento de minério de ouro buscam cada vez mais uma produção de baixo custo e maximização dos ganhos financeiros. A caracterização tecnológica está inserida em uma abordagem multidisciplinar que permite agregar conhecimento, alternativas de otimização e redução nos custos de operação. Inserida como uma ferramenta na caracterização tecnológica, a análise de imagens automatizada tem importante papel no setor mineral principalmente pela rapidez das análises, robustez estatística e confiabilidade nos resultados. A técnica pode ser realizada por meio de imagens adquiridas em microscópio eletrônico de varredura, associada a microanálises químicas sendo utilizada em diversas etapas de um empreendimento mineiro. Este estudo tem como objetivo a caraterização tecnológica de minério de ouro da Mina Morro do Ouro, Minas Gerais na qual foi utilizado a técnica de análise de imagens automatizada por MLA em um conjunto de 88 amostras. Foi possível identificar que 90% do ouro está na fração acima de 0,020 mm; o quartzo e mica representam cerca de 80% da massa total do minério; os sulfetos apresentam diâmetro de círculo equivalente entre 80 e 100 ?m e são representados por pirita e arsenopirita, com pirrotita, calcopirita, esfalerita e galena subordinada. Também foi possível observar que o ouro está majoritariamente associado à pirita e arsenopirita e com o aumento de teor de arsênio, cresce a parcela de ouro associado à arsenopirita. As medianas das distribuições de tamanho dos grãos de ouro apresentam um valor médio de 19 ?m. Verificou-se que a composição dos grãos de ouro é bastante diversa, em média 77% de ouro e 23% de prata. Para material abaixo de 0,50 mm observa-se uma parcela expressiva de perímetro exposto dos grãos de ouro (média 73%); o ouro incluso (21% do total dos grãos de ouro) está associado a pirita e arsenopirita, sendo que em 14 das 88 amostras este valor pode superar 40% do total de ouro contido. A ferramenta da análise de imagens automatizada mostrou-se bastante eficiente definindo características particulares o que fornece de forma objetiva subsídios para os trabalhos de planejamento de mina e processamento mineral. / Evolution in gold ore processing units are increasingly focused towards efficiency seeking routines, leading to a low-cost and maximal financial gain. The mineral characterization, inserted into a multidisciplinary approach enables a knowledge adding dynamics that results in optimization alternatives and consequent operating cost reduction. Being one of commonly used tools in mineral characterization, automated image analysis plays an important role in the mining sector mainly due to its rapid analysis, statistical robustness and reliability of the results. The technique can be performed using images acquired in a scanning electron microscope, usually associated with chemical microanalysis having applications in different stages of a mining venture. The following study aims to produce a mineral characterization using an automated image analysis technique by MLA in a set of 88 samples from the Morro do Ouro gold mine, located in Minas Gerais State, Brazil. Results showed that 90% of gold content is above the 0,02mm size fraction; quartz and mica are the mayor mineral phases identified representing roughly 80% of the total; sulfides have equivalent circle diameter ranging from 80 to 100 ?m present mainly as pyrite and arsenopyrite. It was also noted that gold is mainly associated with pyrite and arsenopyrite and that there is a correlation between the higher share of gold associated with arsenopyrite as a function of higher contents in arsenic. The median size distributions of gold grains have an average of 19 ?m. It was found that the composition of the gold is highly diverse, on average 77% gold and 23% silver. For material below 0.50 mm, was observed that a significant portion of the exposed gold grains circumference (average 73%); the included gold (21% of total grains of gold) is associated with pyrite and arsenopyrite, and that in 14 of 88 samples this value may exceed 40% of the gold contained. The automated image analysis tool proved to be very efficient defining particular features, which provides an objective subsidy for mine planning work and mineral processing.

A análise de imagens automatizada tem atualmente importante papel no setor mineral, devido à rapidez das análises e à confiabilidade dos resultados. A técnica pode ser realizada por meio de imagens adquiridas em microscópio eletrônico de varredura, geralmente associada a microanálises químicas por EDS. Estas analises podem ser realizadas em diversas etapas de um empreendimento mineiro: no auxílio à exploração, lavra (blendagens), processos de beneficiamento (associação mineral), estudos de gênese e controles de mineralização. Essa técnica permite a análise estatística de grande número de partículas/dados, fornecendo informações de composição mineralógica, química mineral, partição de elementos, associações minerais incluindo liberação de fases de interesse, entre outros. Este estudo investiga a aplicabilidade da análise de imagens automatizada em amostras de minério de ferro, com base em diferentes procedimentos de preparação de amostras e rotinas de análise avaliadas por meio do sistema MLA. Um minério de ferro foi escolhido por se tratar de um minério já bastante estudado e bem documentado na bibliografia e que mostra, em geral, mineralogia e texturas simples e repetitivas facilitando a análise dos resultados obtidos com os diferentes procedimentos e rotinas aplicadas. Além disso, o minério de ferro apresenta fases minerais com significativas diferenças de peso específico, o que pode acarretar em maiores dificuldades em termos de amostragem e preparação de amostras. / The automated image analysis has an important role in the mineral industry due to the quickly analysis and the results reliability. The technique can be applied using images from a scanning electron microscope associated usually coupled with chemical microanalysis from an EDS system. These analyses are carried out in many stages of a mining project helping mineral exploration, mining (blending), mineral processing (mineral association), genetic studies and mineralization control. This technique allows a statistic analysis of a large numbers particles and data, providing information of mineralogical composition, mineral chemistry, elemental distribution, mineral associations, phase and spectra liberation and others. This study sets the applicability of automated image analysis of an iron ore sample with diverse preparation procedures and analysis routines, which were evaluated by the MLA system. An iron ore sample was chosen because it is a familiar and well-documented ore in the literature, that usually shows simple textures and mineralogy which makes easier the interpretation of results by a diversity of preparation procedures and analysis routines. Furthermore, the iron ore shows mineral phases with significant differences in specific gravity resulting in more difficulties in sampling and preparation procedures due to potential density segregation.

Leukocyte differential count in peripheral blood smear has, ever since the method was developed more than 100 years ago, been one of the most frequently used diagnostics tool in the routine hematology laboratory. The manual differential count of leukocytes using a microscope is still the standard method in most small and medium sized laboratories. Even though the method does not require any expensive instruments it comes at a high cost due to it being labor intensive and time consuming. In recent years the rapid technical advancements has led to the development of automatic or semi-automatic methods in which the leukocytes are differentiated. In this study a method comparison was made between manual leukocyte differential counts, CellaVision DM1200 Vet and Advia 2120 when analyzing 106 fresh, feline blood samples. The general agreement between results was good, especially for the most common leukocytes, such as neutrophils and lymphocytes. Results for eosinophils and monocytes had moderate agreement. The confidence intervals were generally wider when CellaVision DM1200 Vet was compared with Advia 2120, than when CellaVision DM1200 Vet was compared to the manual differential count. Despite the fact that Advia 2120 and CellaVision DM1200 Vet are both faster and often show comparable results to the manual differential count, the light microscopy will remain the gold standard for difficult samples, where there is suspicion of inflammation (band neutrophils), intracellular microorganisms, reactive lymphocytes or if the sample contains a high degree of smudge cells or artifacts.

Little is known about the generation of Leishmania morphology and the function of morphology in trypanosomatids, despite every species having characteristic cell shapes and undergoing changes in morphology between life cycle stages. To address this I analysed morphogenesis of the cell body and flagellum through the cell cycle of the Leishmania insect (promastigote) life cycle stage using a novel method for determining cell cycle stage from cell size and DNA content. This showed cell body morphology is generated by growth and then remodelling of cell shape around mitosis and cytokinesis. Mathematical modelling of flagellum growth indicated flagellum length continues to increase over multiple cell cycles and does not reach a defined length. I also observed little link between the cell cycle and flagellum length regulation during differentiation to the mammalian macrophage-inhabiting (amastigote) life cycle stage. Analysis of motility showed the diverse flagellar lengths of promastigote Leishmania cells bestow different swimming abilities, and the capacity of Leishmania promastigotes for highly directional swimming differs sharply from trypomastigote Trypanosoma brucei. This difference did not arise from altered flagellar beating therefore appeared to be linked to morphology. Together these indicate the mechanisms of cell body morphogenesis, flagellum length regulation, life cycle stage differentiation and the swimming abilities of the cells the morphogenetic processes generate differ significantly between Leishmania and T. brucei. These insights motivated the programming of automated micrograph analysis tools based on a new DNA staining method to support similar future morphometric analyses. This is the first comprehensive comparison of morphogenesis and function of morphology in a promastigote and a trypomastigote and, by considering these new insights in the context of existing molecular biology and the morphological diversity across many trypanosomatid species, give insight into basic Leishmania biology, the shared molecular mechanisms underlying morphogenesis and the potential functions of the diverse morphologies which are seen in different trypanosomatid species and life cycle stages.

O ouro, assim como outros bens minerais, é uma commodity, ou seja, um produto não diferenciado, com preço determinado pelo mercado internacional, sem a interferência de seus produtores. Diante desse cenário, associado à exaustão dos depósitos minerais de maiores teores, as mineradoras vêm buscando melhores formas de aproveitamento de matérias-primas minerais mais complexas quanto à extração e ao beneficiamento. Os retornos financeiros são obtidos no estrito controle da produção com redução de custos e mitigação de perdas nas operações unitárias. A caracterização tecnológica está inserida como uma abordagem multidisciplinar e fundamental para o melhor aproveitamento dos bens minerais. Possibilita um maior conhecimento do minério e das associações minerais presentes, que auxiliará no desenvolvimento das alternativas de explotação e na otimização do processo em funcionamento. Dentre os procedimentos de caracterização das associações minerais mais acurados citam-se os sistemas automatizados de análise de imagens adquiridas por microscopia eletrônica de varredura. Permitem a avaliação qualitativa ou quantitativa de grande número de partículas quanto à composição química e mineral, partição de elementos nos minerais presentes, formas de associações e liberação entre os minerais. Este estudo se concentra na caracterização de quatro amostras mineralizadas a ouro, de diferentes regiões do Brasil, pela análise automatizada de imagens e por métodos laboratoriais de separações minerais e extração hidrometalúrgica do ouro. A combinação de procedimentos laboratoriais com a análise de imagens permitiu confrontar os resultados das recuperações potenciais, fornecendo subsídios para abordagens de processo, para obtenção da máxima recuperação do ouro e para diagnosticar as características interferentes nesses processos. / Gold and other mineral goods are considered commodities, i.e., a non-differentiated product, at a price determined by the international market, without control of its producers. Given this, coupled with the exhaustion of mineral deposits of higher content, mining companies have been trying to find best procedure alternatives of more complex raw mineral matter. The financial returns are achieved by means of strict production controls, cost reduction and mitigation of losses in unit operations. In this context, process mineralogy is a multidisciplinary and fundamental approach to the better use of mineral goods. Enables a greater knowledge of the ore and its associations, to assist in the development of process alternatives to optimize mineral beneficiation plants. SEM-based automated images analysis is one of the most precise procedures to characterize mineral associations. These systems allow the qualitative or quantitative evaluation of a large number of particles, both as to its chemical and mineralogical content, partition of elements of interest among different minerals, phase\'s associations and liberation. This study focus on the characterization of four gold-mineralized samples from different regions of Brazil, through SEM-based automated image analysis and mineral separations at laboratory scale. The combination of experimental procedures with images analysis allowed the comparison of the results of possible recoveries, providing subsidies for process approaches, to obtain the maximum recovery of gold, and to diagnose the interfering features in these processes.

Since the publication of the complete sequence of the human genome in 2003 there has been great interest in exploring the functions of the proteins encoded by the genes. To reveal the function of each and every protein, investigation of protein localization at the subcellular level has become a central focus in this research area, since the localization and function of a protein is closely related. The objective of the studies presented in this doctoral thesis was to systematically explore the human proteome at the subcellular level using bioimaging and to develop techniques for validation of the results obtained. A common imaging technique for protein detection is immunofluorescence (IF), where antibodies are used to target proteins in fixated cells. A fixation protocol suitable for large-scale IF studies was developed and optimized to work for a broad set of proteins. As the technique relies on antibodies, validation of their specificity to the target protein is crucial. A platform based on siRNA gene silencing in combination with IF was set-up to evaluate antibody specificity by quantitative image analysis before and after suppression of its target protein. As a proof of concept, the platform was then used for validation of 75 antibodies, proving it to be applicable for validation of antibodies in a systematic manner. Because of the fixation, there is a common concern about how well IF data reflects the in vivo subcellular distribution of proteins. To address this, 500 proteins were tagged with green fluorescent protein (GFP) and used to compare protein localization results between IF to those achieved using GFP tagged proteins in live cells. It was concluded that protein localization data from fixated cells satisfactory represented the situation in vivo and together exhibit a powerful approach for confirming localizations of yet uncharacterized proteins. Finally, a global analysis based on IF data of approximately 20 % of the human proteome was performed, providing a first overview of the subcellular landscape in three different cell lines. It was found that the intracellular distribution of proteins is complex, with many proteins occurring in several organelles. The results also confirmed the close relationship between protein function and localization, which in a way further strengthens the accuracy of the IF approach for detection of proteins at the subcellular level. / <p>QC 20121017</p> / The Human Protein Atlas