Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2016-04-05T14:16:18Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Shayane_FINAL.pdf: 7778172 bytes, checksum: bef887b2265bc2ffe53c75c2c275d796 (MD5) / Made available in DSpace on 2016-04-05T14:16:18Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Shayane_FINAL.pdf: 7778172 bytes, checksum: bef887b2265bc2ffe53c75c2c275d796 (MD5) Previous issue date: 2015-08-21 / IF Sertão - PE / Os sistemas de Descoberta de Conhecimentos em Bases de Dados (mais conhecidos como sistemas KDD) e métodos de Aprendizagem de Máquinas preveem situações, agrupam e reconhecem padrões, entre outras tarefas que são demandas de um mundo no qual a maioria dos serviços está sendo oferecido por meio virtual. Apesar dessas aplicações se preocuparem em gerar informações de fácil interpretação, rápidas e confiáveis, as extensas bases de dados utilizadas dificultam o alcance de precisão unida a um baixo custo computacional. Para resolver esse problema, as bases de dados podem ser reduzidas com o objetivo de diminuir o tempo de processamento e facilitar o seu armazenamento, bem como, guardar apenas informações suficientes e relevantes para a extração do conhecimento. Nesse contexto, Métodos de Seleção de Instâncias (MSIs) têm sido propostos para reduzir e filtrar as bases de dados, selecionando ou criando novas instâncias que melhor as descrevam. Todavia, aqui se aplica o Teorema do No Free Lunch, ou seja, a performance dos MSIs varia conforme a base e nenhum dos métodos sempre sobrepõe seu desempenho aos demais. Por isso, esta dissertação propõe uma arquitetura para selecionar o “melhor” MSI para uma dada base de dados (mais adequado emrelação à precisão), chamadaMeta-CISM (Metalearning for Choosing Instance SelectionMethod). Estratégias de meta-aprendizagem são utilizadas para treinar um meta-classificador que aprende sobre o relacionamento entre a taxa de acerto de MSIs e a estrutura das bases. O Meta-CISM utiliza ainda reamostragem e métodos de seleção de atributos para melhorar o desempenho do meta-classificador. A proposta foi avaliada com os MSIs: C-pruner, DROP3, IB3, ICF e ENN-CNN. Os métodos de reamostragem utilizados foram: Bagging e Combination (método proposto neste trabalho). Foram utilizados como métodos de seleção de atributos: Relief-F, CFS, Chi Square Feature Evaluation e Consistency-Based Subset Evaluation. Cinco classificadores contribuíram para rotular as meta-instâncias: C4.5, PART, MLP-BP, SMO e KNN. Uma MLP-BP treinou o meta-classificador. Os experimentos foram realizados com dezesseis bases de dados públicas. O método proposto (Meta-CISM) foi melhor que todos os MSIs estudados, na maioria dos experimentos realizados. Visto que eficientemente seleciona um dos três melhores MSIs em mais de 85% dos casos, a abordagemé adequada para ser automaticamente utilizada na fase de pré-processamento das base de dados. / The systems for Knowledge Discovery in Databases (better known as KDD systems) andMachine Learning methods predict situations, recognize and group (cluster) patterns, among other tasks that are demands of a world in which the most of the services is being offered by virtual ways. Although these applications are concerned in generate fast, reliable and easy to interpret information, extensive databases used for such applications make difficult achieving accuracy with a low computational cost. To solve this problem, the databases can be reduced aiming to decrease the processing time and facilitating its storage, as well as, to save only sufficient and relevant information for the knowledge extraction. In this context, Instances SelectionMethods (ISMs) have been proposed to reduce and filter databases, selecting or creating new instances that best describe them. Nevertheless, No Free Lunch Theorem is applied, that is, the ISMs performance varies according to the base and none of the methods always overcomes their performance over others. Therefore, this work proposes an architecture to select the "best"ISM for a given database (best suited in relation to accuracy), called Meta-CISM (Meta-learning for Choosing Instance SelectionMethod). Meta-learning strategies are used to train a meta-classifier that learns about the relationship between the accuracy rate of ISMs and the bases structures. TheMeta-CISM still uses resampling and feature selection methods to improve the meta-classifier performance. The proposal was evaluated with the ISMs: C-pruner, DROP3, IB3, ICF and ENN-CNN. Resampling methods used were: Bagging and Combination (method proposed in this work). The Feature SelectionMethods used were: Relief-F, CFS, Chi Square Feature Evaluation e Consistency-Based Subset Evaluation. Five classifiers contributed to label the meta-instances: C4.5, PART, MLP-BP, SMO e KNN. The meta-classifier was trained by a MLP-BP. Experiments were carried with sixteen public databases. The proposed method (Meta-CISM) was better than all ISMs studied in the most of the experiments performed. Since that efficiently selects one of the three best ISMs in more than 85% of cases, the approach is suitable to be automatically used in the pre-processing of the databases.

Master’s thesis deals with problems of maintenance of small transport aircraft with application MSG-3 methodology and maintenance plan development. It is aimed at systems and powerplant maintenance. This method is applied on L-410UVP-E20 aircraft. Other parts of this work are assessment of recent situation in maintenance of airplanes, used approaches to maintenance, methods and documentation for maintenance. Last part of this work is practical example of MSG-3 process on assign item and its interpretation.

The objective of the thesis is to design and implement in the VHDL language a simple multiprocessor supporting parallel computing. Furthemore, the author has designed and realized universal transparent generic interconnection layer with the objective to connect any given number of processor cores to shared address space using arbitrated bus. Parametrized cache has been allocated to each core in the layer. MSI protocol was used to deal with the issue of memory coherence of the implemented system. Direct and indirect synchornisation support is available to the user. In order to verify the functionality of the system, simple processor core has been designed and implemented, and its copies were connected to the interconnection layer. Various testing programmes have been used to verify the functionality of the system, which also confirmed that the acceleration of computing has been achieved successfully. Virtex6 chip has been used to test the whole system.

(EN) Presented PhD thesis was aimed at the utilization of desorption ionization mass spectrometric techniques to study different types of biological samples. An important part of this thesis was the construction of a universal platform for ambient ionization techniques (AIT) in mass spectrometry (MS) for the experiments performed in combination with desorption electrospray ionization (DESI) and desorption atmospheric pressure photoionization (DAPPI). Initially manual platform for AIT was during the development rebuild into a motorized platform operated by software via user interface of the mass spectrometer. Using the motorized platform it was possible to carry out a one- dimensional (1-D) and two dimensional (2-D) samples analysis with a defined step size of imaging. Platform designed for AIT in MS was tested in DESI-MS mode by model analyte Rhodamine B and by plant samples which were represented by Aconitum plicatum. The platform was also tested on biological samples of vernix caseosa using DAPPI-MS ion source. Model analyte rhodamine B was in DESI-MS technique used to study the basic desorption ionization processes occurring in the DESI during the sample analysis. In order to highlight the surface structures the colloidal solution of ferrite nanoparticles was added into the sprayed liquid. For...

Desorption Electrospray Ionization Mass spectrometry Imaging (DESI-MSI) is an area of great interest and a promising tool in the field of chemical imaging. It is a powerful, label-free technique, which can determine, map and visualize different molecular compounds on a sample surface. The amount of information acquired in a single DESI-MSI experiment is enormous compared to other techniques, as it can simultaneously detect different compounds with their spatial distribution on the surface. The experiment can be used to produce two-dimensional and three-dimensional images. Chapter 2 focuses on the design and optimization of the setup for performing DESI-MS imaging on various substrates. The proposed setup was tested for its lateral spatial resolution. To provide proof-of-concept of the design, preliminary tests were performed to generate images from commercial thin layer chromatographic plates and photographic paper. Chapter 3 focuses on demonstrating the compatibility of novel microfabricated Thin Layer Chromatography plates (M-TLC plates) for detection with DESI-MSI.

Cette thèse propose l’analyse spatio-temporelle d’un mouvement scientifique au moyen d’une théorie sociologique (Frickel, Gross, 2005), complétée par une approche d’analyse spatiale, combinaison qui fait l’originalité de ce travail. La géographie théorique et quantitative européenne francophone, dont l’évolution temporelle et la configuration spatiale restaient très mal connues, en est l’objet. Elle permet de mettre en perspective les rares auto-analyses qui y sont consacrées. Deux grands types de matériaux sont mobilisés : archivistique (Répertoire des géographes, revues ou listes de communications à des colloques spécialisés) et mémoriel (sur la base de 58 entretiens semi-directifs). L’analyse de ces corpus combine des méthodes quantitatives (de la théorie des graphes à l’analyse de la spécialisation des unités géographiques) et qualitatives (analyse thématique et sémantique des corpus et des entretiens des acteurs). Cette investigation a permis de proposer une périodisation du mouvement, articulée autour de trois grands moments : 1. l’émergence, de 1971-72 à 1984, impliquant un faible nombre d’acteurs et de lieux, moment où des chercheurs isolés entrent en interaction, 2. l’âge d’or, de 1985 à la fin des années 1990, marqué par l’existence d’un collectif structuré, reposant sur de nombreuses interactions entre les lieux européens francophones, qui se reproduit et essaime, 3. la généralisation du cœur de connaissances qui s’accompagne d’un déclin de l’identification au mouvement, des années 2000 à nos jours, selon de nombreuses reconfigurations, voyant la disparition de certains pôles historiques. / The aim of this thesis is to implement a spatio-temporal analysis of a scientific movement based on a sociological theory (Frickel, Gross, 2005), combined with spatial analysis. This combination constitutes the major contribution of the present work. For this analysis, the author has chosen as a subject the European French-speaking theoretical and quantitative geography, whose temporal evolution and spatial configuration remained largely unknown until now. The thesis uses two main kinds of materials: archival (Répertoire des géographes, journals, lists of communications of TQG colloquium) and memorial with 58 semi-structured interviews collected from a sample of actors of the movement. The analysis of the corpus combines quantitative methods (from graph theory to the analysis of geographical units’ specialization), and qualitative ones (thematic and semantic analysis of the corpus and interviews of actors). This investigation has allowed us to propose a periodization of the movement since its beginnings in the 1960’s organized around three main periods: 1. the emergence, from 1971-72 to 1984, involving a small number of actors and locations, and while individual researchers interact in order to give themselves a self training, 2. the Golden Age, from 1985 to the end of the 1990’s, marked by the formation of organized groups based on many interactions between French-speaking places in Europe, self reproducing and spreading. 3. The generalization of the knowledge core which is accompanied by a decline in identification with the movement, from the 2000’s to the present, according to many reconfigurations including the disappearance of certain historical centres.

Hypothesis testing has long been a formal and standardized process. Hypothesis generation, on the other hand, remains largely informal. This thesis assess whether eXplainable AI (XAI) can aid in the standardization of hypothesis generation through its utilization as a hypothesis generating tool for medical research. We produce XAI heat maps for a Convolutional Neural Network (CNN) trained to classify Microsatellite Instability (MSI) in colon and gastric cancer with four different XAI methods: Guided Backpropagation, VarGrad, Grad-CAM and Sobol Attribution. We then compare these heat maps with pathology annotations in order to look for differences to turn into new hypotheses. Our CNN successfully generates non-random XAI heat maps whilst achieving a validation accuracy of 85% and a validation AUC of 93% – as compared to others who achieve a AUC of 87%. Our results conclude that Guided Backpropagation and VarGrad are better at explaining high-level image features whereas Grad-CAM and Sobol Attribution are better at explaining low-level ones. This makes the two groups of XAI methods good complements to each other. Images of Microsatellite Insta- bility (MSI) with high differentiation are more difficult to analyse regardless of which XAI is used, probably due to exhibiting less regularity. Regardless of this drawback, our assessment is that XAI can be used as a useful hypotheses generating tool for research in medicine. Our results indicate that our CNN utilizes the same features as our basic pathology annotations when classifying MSI – with some additional features of basic pathology missing – features which we successfully are able to generate new hypotheses with.

No description available.

No description available.

<div> Advancements in cancer treatments have increased rapidly in recent years, but cures remain elusive. Surgical tumor resection is a central treatment for many solid malignancies. Residual tumor at surgical margins leads to tumor recurrence. Novel tools for assessing residual tumor at surgical margins could improve surgical outcomes by helping to maximize the extent of resection. Ambient ionization-mass spectrometry (MS) methods generate and analyze ions from minimally prepared samples in near-real-time (e.g. seconds to minutes). These methods leverage the high sensitivity and specificity of mass spectrometry for analyzing gas phase ions and generating those ions quickly and with minimal sample preparation. Recent work has shown that differential profiles of ions, corresponding to phospholipids and small metabolites, are detected from cancerous and their respective normal tissue with ambient ionization-MS methods. When properly implemented, ambient ionization-MS could be used to assess for tumor at surgical margins and provide a molecular diagnosis during surgery. </div><div><br></div><div>The research herein reports efforts in developing rapid intraoperative ambient ionization-MS methods for the molecular assessment of cancerous tissues. Touch spray (TS) ionization and desorption electrospray ionization (DESI) were utilized to analyze kidney cancer and brain cancer.</div><div><br></div><div> As a demonstration of the applicability of TS-MS to provide diagnostic information from fresh surgical tissues, TS-MS was used to rapidly analyze renal cell carcinoma and healthy renal tissue biopsies obtained from human subjects undergoing nephrectomy surgery. Differential phospholipid profiles were identified using principal component analysis (PCA), and the significant ions were characterized using multiple stages of mass spectrometry and high resolution/exact mass MS. The same TS-MS analyzed renal tissues were subsequently analyzed with DESI-MS imaging to corroborate the TS-MS results, and the significant DESI-MS ions were also characterized with MS.</div><div><br></div><div>Significant efforts were made in developing and evaluating a standalone intraoperative DESI-MS system for analyzing brain tissue biopsies during brain tumor surgery. The intraoperative DESI-MS system consists of a linear trap quadrupole mass spectrometer placed on a custom-machined cart that contains all hardware for operating the mass spectrometer. This instrument was operated in the neurosurgical suites at Indiana University School of Medicine to rapidly analyze brain tissue biopsies obtained from glioma resection surgeries. A DESI-MS library of normal brain tissue and glioma was used to statistically classify the brain tissue biopsies collected in the operating room. Multivariate statistical methodologies were employed to predict the disease state and tumor cell percentage of the samples. A DESI-MS assay for detecting 2-hydroxyglutarate (2HG), the oncometabolic product of the isocitrate dehydrogenase (IDH) mutation (a key glioma prognostic marker), was developed and applied to determine the IDH mutation status during the surgical resection. The strengths, weaknesses, and areas of future work in this field are discussed. </div><div><br></div>