Bakgrund: Personer över 80 år har ökat med nästan 22 % under åren 1992 – 2005. Under samma tid har vårdplatserna inom hela akutsjukvården nästan halverats. Detta har lett till ökat vårdbehov och större belastning på framför allt akutmottagningar, medicinkliniker, primärvården och kommunen. Tidigare studier har visat att inadekvata akutmedicinska inläggningar sker ofta runtom i Europa. Syfte: Syftet med studien var att studera inskrivning och utskrivning av patienter vid akutmedicinska avdelningar på ett sjukhus i Sverige. Metod: Studien genomfördes under 2007 på en akutmedicinsk klinik i Sydöstra Sverige. 411 akutmedicinska patientinläggningar bedömdes med Appropriateness Evaluation Protocol (AEP). Lika många hade möjlighet att besvara en enkätfråga, riktad till patienter, sjuksköterskor och läkare, om bedömning av utskrivningen. Resultat: Studien visar att 28,7% av patientinläggningarna inte uppfyllde AEP’s kriterier. De flesta patienter hade bröstsmärta som inskrivningsorsak oavsett adekvat eller inadekvat inläggning. Nära 30 % av patienterna, sjuksköterskorna och läkarna bedömde möjlig utskrivning ett dygn tidigare. Slutsats: Studien visar att inadekvata akutmedicinska inläggningar, enligt instrumentet AEP utgör en betydande del av beläggningen på en medicinklinik som är potentiellt påverkbar. Studien visar också att beläggningen kan påverkas genom tidigarelagd utskrivning enligt bedömningar av såväl patienter som sjuksköterskor och läkare. Resultatet talar för att en bättre kommunikation mellan patienter, sjuksköterskor och läkare skulle kunna optimera längden för akuta medicinska vårdtillfällen. / Background: Persons aged over 80 has increased by almost 22% during 1992-2005. In that time, bed occupancy throughout the emergency medical services has almost decreased to the half. This has led to increased care needs and greater burden, especially on emergency wards, medical clinics, primary care and community. Previous studies have shown that inadequate emergency medical admissions are frequently around Europe. Aim: The aim was to study admission and discharge of patients in acute medical wards at a hospital in Sweden. Method: The study was conducted in 2007 at an acute medical clinic in Southeast Sweden. 411 acute medical patient admissions were assessed with the Appropriateness Evaluation Protocol (AEP). Just as many were able to answer a survey question, directed to patients, nurses and doctors, on the assessment of discharge. Results: The study shows that 28.7% of patients' admissions did not meet AEP's criteria. Most patients had chest pain that enrollment cause any adequate or inadequate curing. Almost 30% of patients, nurses and doctors thought that discharge of the patient was possible one day earlier. Conclusion: The study shows that inadequate emergency medical admissions, with the instrument AEP, are an important part of the hospital bed occupancy of a medical clinic that is potentially impressionable. The study also shows that the hospital bed occupancy can be influenced by earlier discharge according to assessments by both patients, nurses and doctors.

Nas últimas décadas, diversos estudos acadêmicos investigaram práticas de gestão da Segurança e Saúde no Trabalho (SST) de forma a identificar fatores que contribuem para o sucesso de programas de prevenção de acidentes no setor da construção civil. A maior parte desses estudos foi baseada na realização de surveys e objetivaram identificar correlações entre as taxas de acidentes de cada empresa e a existência ou não de práticas, permitindo apontar os conjuntos de práticas que têm maior impacto na SST. Em que pesem as contribuições destes estudos, principalmente no que diz respeito à identificação e disseminação de práticas associadas à prevenção de acidentes, quatro principais críticas podem ser feitas: (a) não há um conceito definido de o que são boas práticas e critérios para categorizá-las; (b) escopo limitado a grandes empresas de países desenvolvidos; (c) a falta de um referencial teórico adequado; e (d) a abordagem metodológica adotada, cujo enfoque é descritivo. O objetivo principal desta pesquisa é propor um protocolo para caracterização e avaliação do grau de implementação de práticas de gestão da SST, através da utilização de múltiplas fontes de evidência, visando a obter uma compreensão mais detalhada destas práticas, que levem em consideração o contexto de aplicação. Em relação ao método, foi adotada a abordagem da pesquisa construtiva, escolhida por se tratar do desenvolvimento de um protocolo para a solução de um problema com relevância prática e teórica. Inicialmente, foi realizada uma revisão da literatura acerca das práticas de gestão da SST e do referencial teórico da Engenharia de Resiliência (ER). O protocolo foi proposto com base na revisão da literatura e também em uma survey realizada com uma amostra de empresas de construção. A ferramenta foi desenvolvida, aprimorada e testada em 5 estudos de caso em canteiros de obras, realizados de forma sequencial. Além disso, a mesma foi aplicada em 3 obras na Espanha, com o objetivo de avaliar a sua aplicabilidade em outro contexto. Os resultados deste conjunto de obras foram processados e analisados, de forma a exemplificar os dados produzidos. Por fim, com base na avaliação do protocolo desenvolvido, foram propostas diretrizes para a sua implantação. As principais contribuições da pesquisa são referentes ao protocolo de avaliação desenvolvido, que possui um enfoque prescritivo. Por um lado, este possibilita a coleta sistemática de dados sobre práticas de SST que pode ser utilizado na avaliação do sistema de gestão da SST das empresas, possibilitando a realização de benchmarking externo e interno. Por outro lado, a ferramenta pode contribuir para o registro e a disseminação das práticas de gestão de SST no setor da construção. / Several academic studies have investigated Occupational Safety and Health (OSH) managerial practices in order to identify factors that contribute for the success of accident prevention programs in the construction industry. Most of those studies were based on surveys and aimed to identify correlations between accident rates and the existence or non-existence of such practices, establishing sets of practices that have the greatest impact on OSH. Despite the contributions of those studies, especially regarding the identification and dissemination of practices associated with accident prevention, four main drawbacks can be pointed out: (a) the lack of definition for best practices and of criteria for categorizing them, (b) scope limited to large companies from developed countries, (c) the lack of theoretical foundation, and (d) the descriptive focus of the methodological approach adopted. The main objective of this research work is to devise a protocol for characterizing and evaluating the degree of implementation of OSH managerial practices based on multiple sources of evidence, aiming to get a more detailed understanding of those practices, and take into account the context of application. Regarding the method, the constructive research approach was adopted, since this investigation proposes the development of a protocol for solving a problem with practical and theoretical relevance. Initially, a literature review on OSH managerial best practices and on the Resilience Engineering (RE) theoretical framework was undertaken. The protocol has been proposed based on both a literature review and also on a survey conducted with a sample of construction companies. The tool was developed and tested in five case studies in construction sites carried out sequentially. In addition, it was also applied in three construction sites in Spain in order to evaluate its applicability in another context. The results of those construction sites were processed and analyzed to illustrate the data produced. Finally, based on the evaluation of the protocol, some guidelines for implementing it were proposed. The main contributions of this research are related to the protocol, which has a prescriptive approach. From one hand, it enables the systematic collection of data on OSH practices that can be used for assessing the OSH management system of construction companies, making it possible to carry out internal and external benchmarking. From the other hand, with the tool may contribute for the documentation and dissemination of OSH best practices in the construction sector.

Nas últimas décadas, diversos estudos acadêmicos investigaram práticas de gestão da Segurança e Saúde no Trabalho (SST) de forma a identificar fatores que contribuem para o sucesso de programas de prevenção de acidentes no setor da construção civil. A maior parte desses estudos foi baseada na realização de surveys e objetivaram identificar correlações entre as taxas de acidentes de cada empresa e a existência ou não de práticas, permitindo apontar os conjuntos de práticas que têm maior impacto na SST. Em que pesem as contribuições destes estudos, principalmente no que diz respeito à identificação e disseminação de práticas associadas à prevenção de acidentes, quatro principais críticas podem ser feitas: (a) não há um conceito definido de o que são boas práticas e critérios para categorizá-las; (b) escopo limitado a grandes empresas de países desenvolvidos; (c) a falta de um referencial teórico adequado; e (d) a abordagem metodológica adotada, cujo enfoque é descritivo. O objetivo principal desta pesquisa é propor um protocolo para caracterização e avaliação do grau de implementação de práticas de gestão da SST, através da utilização de múltiplas fontes de evidência, visando a obter uma compreensão mais detalhada destas práticas, que levem em consideração o contexto de aplicação. Em relação ao método, foi adotada a abordagem da pesquisa construtiva, escolhida por se tratar do desenvolvimento de um protocolo para a solução de um problema com relevância prática e teórica. Inicialmente, foi realizada uma revisão da literatura acerca das práticas de gestão da SST e do referencial teórico da Engenharia de Resiliência (ER). O protocolo foi proposto com base na revisão da literatura e também em uma survey realizada com uma amostra de empresas de construção. A ferramenta foi desenvolvida, aprimorada e testada em 5 estudos de caso em canteiros de obras, realizados de forma sequencial. Além disso, a mesma foi aplicada em 3 obras na Espanha, com o objetivo de avaliar a sua aplicabilidade em outro contexto. Os resultados deste conjunto de obras foram processados e analisados, de forma a exemplificar os dados produzidos. Por fim, com base na avaliação do protocolo desenvolvido, foram propostas diretrizes para a sua implantação. As principais contribuições da pesquisa são referentes ao protocolo de avaliação desenvolvido, que possui um enfoque prescritivo. Por um lado, este possibilita a coleta sistemática de dados sobre práticas de SST que pode ser utilizado na avaliação do sistema de gestão da SST das empresas, possibilitando a realização de benchmarking externo e interno. Por outro lado, a ferramenta pode contribuir para o registro e a disseminação das práticas de gestão de SST no setor da construção. / Several academic studies have investigated Occupational Safety and Health (OSH) managerial practices in order to identify factors that contribute for the success of accident prevention programs in the construction industry. Most of those studies were based on surveys and aimed to identify correlations between accident rates and the existence or non-existence of such practices, establishing sets of practices that have the greatest impact on OSH. Despite the contributions of those studies, especially regarding the identification and dissemination of practices associated with accident prevention, four main drawbacks can be pointed out: (a) the lack of definition for best practices and of criteria for categorizing them, (b) scope limited to large companies from developed countries, (c) the lack of theoretical foundation, and (d) the descriptive focus of the methodological approach adopted. The main objective of this research work is to devise a protocol for characterizing and evaluating the degree of implementation of OSH managerial practices based on multiple sources of evidence, aiming to get a more detailed understanding of those practices, and take into account the context of application. Regarding the method, the constructive research approach was adopted, since this investigation proposes the development of a protocol for solving a problem with practical and theoretical relevance. Initially, a literature review on OSH managerial best practices and on the Resilience Engineering (RE) theoretical framework was undertaken. The protocol has been proposed based on both a literature review and also on a survey conducted with a sample of construction companies. The tool was developed and tested in five case studies in construction sites carried out sequentially. In addition, it was also applied in three construction sites in Spain in order to evaluate its applicability in another context. The results of those construction sites were processed and analyzed to illustrate the data produced. Finally, based on the evaluation of the protocol, some guidelines for implementing it were proposed. The main contributions of this research are related to the protocol, which has a prescriptive approach. From one hand, it enables the systematic collection of data on OSH practices that can be used for assessing the OSH management system of construction companies, making it possible to carry out internal and external benchmarking. From the other hand, with the tool may contribute for the documentation and dissemination of OSH best practices in the construction sector.

Nas últimas décadas, diversos estudos acadêmicos investigaram práticas de gestão da Segurança e Saúde no Trabalho (SST) de forma a identificar fatores que contribuem para o sucesso de programas de prevenção de acidentes no setor da construção civil. A maior parte desses estudos foi baseada na realização de surveys e objetivaram identificar correlações entre as taxas de acidentes de cada empresa e a existência ou não de práticas, permitindo apontar os conjuntos de práticas que têm maior impacto na SST. Em que pesem as contribuições destes estudos, principalmente no que diz respeito à identificação e disseminação de práticas associadas à prevenção de acidentes, quatro principais críticas podem ser feitas: (a) não há um conceito definido de o que são boas práticas e critérios para categorizá-las; (b) escopo limitado a grandes empresas de países desenvolvidos; (c) a falta de um referencial teórico adequado; e (d) a abordagem metodológica adotada, cujo enfoque é descritivo. O objetivo principal desta pesquisa é propor um protocolo para caracterização e avaliação do grau de implementação de práticas de gestão da SST, através da utilização de múltiplas fontes de evidência, visando a obter uma compreensão mais detalhada destas práticas, que levem em consideração o contexto de aplicação. Em relação ao método, foi adotada a abordagem da pesquisa construtiva, escolhida por se tratar do desenvolvimento de um protocolo para a solução de um problema com relevância prática e teórica. Inicialmente, foi realizada uma revisão da literatura acerca das práticas de gestão da SST e do referencial teórico da Engenharia de Resiliência (ER). O protocolo foi proposto com base na revisão da literatura e também em uma survey realizada com uma amostra de empresas de construção. A ferramenta foi desenvolvida, aprimorada e testada em 5 estudos de caso em canteiros de obras, realizados de forma sequencial. Além disso, a mesma foi aplicada em 3 obras na Espanha, com o objetivo de avaliar a sua aplicabilidade em outro contexto. Os resultados deste conjunto de obras foram processados e analisados, de forma a exemplificar os dados produzidos. Por fim, com base na avaliação do protocolo desenvolvido, foram propostas diretrizes para a sua implantação. As principais contribuições da pesquisa são referentes ao protocolo de avaliação desenvolvido, que possui um enfoque prescritivo. Por um lado, este possibilita a coleta sistemática de dados sobre práticas de SST que pode ser utilizado na avaliação do sistema de gestão da SST das empresas, possibilitando a realização de benchmarking externo e interno. Por outro lado, a ferramenta pode contribuir para o registro e a disseminação das práticas de gestão de SST no setor da construção. / Several academic studies have investigated Occupational Safety and Health (OSH) managerial practices in order to identify factors that contribute for the success of accident prevention programs in the construction industry. Most of those studies were based on surveys and aimed to identify correlations between accident rates and the existence or non-existence of such practices, establishing sets of practices that have the greatest impact on OSH. Despite the contributions of those studies, especially regarding the identification and dissemination of practices associated with accident prevention, four main drawbacks can be pointed out: (a) the lack of definition for best practices and of criteria for categorizing them, (b) scope limited to large companies from developed countries, (c) the lack of theoretical foundation, and (d) the descriptive focus of the methodological approach adopted. The main objective of this research work is to devise a protocol for characterizing and evaluating the degree of implementation of OSH managerial practices based on multiple sources of evidence, aiming to get a more detailed understanding of those practices, and take into account the context of application. Regarding the method, the constructive research approach was adopted, since this investigation proposes the development of a protocol for solving a problem with practical and theoretical relevance. Initially, a literature review on OSH managerial best practices and on the Resilience Engineering (RE) theoretical framework was undertaken. The protocol has been proposed based on both a literature review and also on a survey conducted with a sample of construction companies. The tool was developed and tested in five case studies in construction sites carried out sequentially. In addition, it was also applied in three construction sites in Spain in order to evaluate its applicability in another context. The results of those construction sites were processed and analyzed to illustrate the data produced. Finally, based on the evaluation of the protocol, some guidelines for implementing it were proposed. The main contributions of this research are related to the protocol, which has a prescriptive approach. From one hand, it enables the systematic collection of data on OSH practices that can be used for assessing the OSH management system of construction companies, making it possible to carry out internal and external benchmarking. From the other hand, with the tool may contribute for the documentation and dissemination of OSH best practices in the construction sector.

Orientador: Paula Schmidt Azevedo Gaiolla / Resumo: Introdução: A atual mudança no perfil epidemiológico, com aumento da expectativa de vida mundial e no Brasil, contribui para o crescimento da população de idosos e consequentemente de ampliação do espectro das comorbidades destes. Adicionalmente, independentemente da idade, existe aumento de doenças crônicas associadas principalmente aos hábitos modernos. Esses fatores elevam os riscos de complicações durante e após as cirurgias. Por outro lado, nota-se ainda avanço no que diz respeito às propostas e técnicas anestésicas e cirúrgicas, cada vez menos invasivas e mais resolutivas, permitindo aos idosos e portadores de doenças crônicas, serem submetidos a diferentes procedimentos. Portanto, é crucial que seja realizada criteriosa avaliação perioperatória individualizada, visando prever, intervir e minimizar efeitos indesejados associados aos procedimentos cirúrgicos e aos fatores de risco individuais de cada paciente. Objetivo: O presente trabalho propõe a criação de protocolo, de ficha de atendimento e de Manual para guiar a avaliação e manejo perioperatório dos pacientes do Hospital das Clínicas da Faculdade de Medicina de Botucatu. Metodologia: O método utilizado na pesquisa foi a busca avançada em inglês no banco de dados da Pubmed, Cochrane, Lilacs e Scielo dos termos: “Preoperative risk screening”; “Preoperative Cardiovascular risk stratification”, “Preoperative pulmonary risk stratification”, “perioperative management of the cirrhotic patient”, “perioperative renal care”,... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: The current change in the epidemiological profile, with an increase in life expectancy worldwide and in Brazil, contributes to the growth of the elderly population and, consequently, to a broader spectrum of their comorbidities. In addition, regardless of age, there is an increase in chronic diseases associated with modern habits. These factors raise the risk of complications during and after surgeries. On the other hand, there is still progress in regard to proposals and anesthetic and surgical techniques, which are becoming less invasive and more resolutive, allowing the elderly and patients with chronic diseases to undergo different procedures. Therefore, it is crucial that a careful individualized perioperative evaluation is performed, in order to predict, intervene and minimize unwanted effects associated with surgical procedures and individual risk factors of each patient. Objective: This paper proposes the creation of a protocol, an information sheet and a manual to guide the evaluation and perioperative management of patients at the Hospital das Clínicas, Botucatu Medical School. Methodology: The method used in the research was the advanced search in English in the database of Pubmed, Cochrane, Lilacs and Scielo of the terms: "Preoperative risk screening"; "Preoperative Cardiovascular risk stratification", "Preoperative pulmonary risk stratification", "perioperative management of the cirrhotic patient", "perioperative renal care", arranged in isolation a... (Complete abstract click electronic access below) / Mestre

Le nombre croissant d'approches de détection de texte proposé dans la littérature exige une évaluation rigoureuse de la performance. Un protocole d'évaluation repose sur trois éléments: une vérité terrain fiable, une stratégie d'appariement et enfin un ensemble de métriques. Peu de protocoles existent et ces protocoles manquent souvent de précision. Dans cette thèse, nous proposons un nouveau protocole d'évaluation qui résout la plupart des problèmes rencontrés dans les méthodes d'évaluation actuelles. Ce travail est axé sur trois contributions principales : tout d’abord, nous introduisons une représentation complexe de la vérité terrain qui ne contraint pas les détecteurs de texte à adopter un niveau de granularité de détection spécifique ou une représentation d'annotation ; d’autre part, nous proposons un ensemble de règles capables d'évaluer tous types de scénario qui peuvent se produire entre les objets de la vérité terrain et les détections correspondantes ; et enfin, nous montrons comment nous pouvons analyser un ensemble de résultats de détection, non seulement à travers un ensemble de mesures, mais aussi à travers une représentation visuelle intuitive. Un défi fréquent pour de nombreux systèmes de détection de texte est d'aborder la variété des caractéristiques de texte dans des images naturelles ou d’origine numérique pour lesquels les OCR actuels ne sont pas bien adaptées. Par exemple, des textes en perspective sont fréquemment présents dans les images réelles. Dans cette thèse, nous proposons également une procédure de rectification capable de corriger des textes hautement déformés, évalué sur un ensemble de données difficiles. / The growing number of text detection approaches proposed in the literature requires a rigorous performance evaluation and ranking. An evaluation protocol relies on three elements: a reliable text reference, a matching strategy and finally a set of metrics. The few existing evaluation protocols often lack accuracy either due to inconsistent matching or due to unrepresentative metrics. In this thesis we propose a new evaluation protocol that tackles most of the drawbacks faced by currently used evaluation methods. This work is focused on three main contributions: firstly, we introduce a complex text reference representation that does not constrain text detectors to adopt a specific detection granularity level or annotation representation; secondly, we propose a set of matching rules capable of evaluating any type of scenario that can occur between a text reference and a detection; and finally we show how we can analyze a set of detection results, not only through a set of metrics, but also through an intuitive visual representation. A frequent challenge for many Text Understanding Systems is to tackle the variety of text characteristics in born-digital and natural scene images for which current OCRs are not well adapted. For example, texts in perspective are frequently present in real-word images because the camera capture angle is not normal to the plane containing the text regions. Despite the ability of some detectors to accurately localize such text objects, the recognition stage fails most of the time. In this thesis we also propose a rectification procedure capable of correcting highly distorted texts evaluated on a very challenging dataset.

The field of professional learning in education has been studied and added to extensively in the last few decades. Because the importance of learning in authentic contexts through professional dialogue has become so important, high quality, school-based professional learning is vital to building capacity at the school level. Unfortunately, the literature on professional development (PD) does not provide much guidance on how to bridge theory and practice at the school level, creating a gap. With the goal of PD ultimately being to improve teacher performance and student learning, the problem with this gap is that school-level professional development is arbitrarily planned, resulting in variable outcomes. I propose the reason for this is schools lack a comprehensive framework or tool that guides the design of a quality professional learning plan. This problem was identified in Orange County Public School and this dissertation in practice aims at developing a solution that accounts for the district*s specific contextual needs. My proposed solution is the design of an integrative tool that school leaders can use to guide them through the professional development planning process. The School-based Professional Learning Design Tool incorporates the professional development standards in planning, learning, implementing, and evaluating outlined in the Florida Professional Development System Evaluation Protocol. It also guides leaders in taking an inventory of the culture and context of their school in order to plan PD that will be viable given those considerations. The components of the Tool guide teams through assessing school teacher performance and student achievement data to help identify focus groups; determining gaps in learning through root cause analysis; creating goals aligned to gaps in performance; and selecting strategies for professional learning, follow-up support, and evaluation. The development of the Tool was informed by the extant literature on professional development, organizational theory, state and national standards for professional development, and principles of design. The Tool is to be completed in four phases. Phases one and two, the focus of this paper, include the literature review, organizational assessment, design specifications, and the first iteration of the Tool. In the next phases, the goals are to solicit feedback from an expert panel review, create a complete version of the Tool, and pilot it in elementary schools. Although the development of the Tool through its final phases will refine it considerably, there are limitations that will transcend all iterations. While the Tool incorporates best practices in professional development, the lack of empirical evidence on the effectiveness of specific PD elements in the literature renders this Tool only a best guess in helping schools plan effective professional development. Another limitation is that the Tool is not prescriptive and cannot use school data to make decisions for what strategies to implement. Taking these limitations into consideration, the use of this Tool can significantly impact the quality and effectiveness of professional development in schools.

L’opération morphologique de squelettisation transforme chaque objet d’une image en une forme linéique qui préserve la topologie de ce dernier (propriété d’homotopie). Elle est largement utilisée en biométrie mais aussi dans la reconnaissance des caractères ainsi que pour l’extraction de la microarchitecture osseuse. L’objectif de cette thèse est de développer une méthode de squelettisation appliquée directement sur les niveaux de gris de l’image, ce qui a pour large avantage de s’affranchir de prétraitement comme la binarisation. Une revue des méthodes de squelettisation en niveaux de gris permet de constater que l’amincissement est l’une des approches les plus usitées de par sa propriété d’homotopie. Cependant, cette approche est sensible au bruit de l’image et produit des squelettes sur-connectés. Un premier paramétrage de l’amincissement a été proposé dans la littérature afin d’abaisser des configurations de pixels liées au bruit. La première contribution de ce travail est de proposer un ajustement de ce paramètre basé sur une décision statistique. Il s’agit d’identifier les tests d’hypothèses correspondants aux différentes configurations d’abaissement en vue de fixer ce paramètre de façon locale. Ceci conduit à la mise en place d’une squelettisation appelée Self Contrast Controlled Thinning (SCCT) puisque robuste au bruit tout en s’adaptant automatiquement au contraste de l’image. La squelettisation SCCT est rendue accessible aux domaines d’application grâce à son implantation optimisée basée sur les files d’attente hiérarchiques. Ayant noté le peu d’efforts consacrés à l’évaluation de la squelettisation en niveaux de gris, la deuxième contribution de ce travail est de proposer un protocole visant à évaluer l’opération de squelettisation sur la base des propriétés requises à savoir la préservation de la topologie et de la géométrie. Ce protocole est déroulé sur une base d’images synthétiques et nous permet de comparer notre approche à celles de la littérature. La troisième contribution est de proposer une structuration du squelette en graphe donnant accès aux descripteurs structurels et morphométriques des objets étudiés en vue d’une exploitation du squelette par les experts des domaines d’applications. Dans le cadre du projet Voxelo coordonné par le laboratoire B2OA de l’Université Paris Diderot, cette structuration est exploitée pour extraire les descripteurs de la qualité de la microarchitecture osseuse à partir d’images RX haute résolution. / Skeletonization is an image transformation that aims to represent objects by their medial axis while preserving their topological characteristics (homotopy). It is widely used in biometrics, character recognition and also in the extraction of bone microarchitecture. The objective of this thesis is to develop a skeletonization method applied directly on image gray levels. This has the large advantage of freeing the operation from preprocessing techniques such as binarization. A review of grayscale skeletonization methods shows that the morphological thinning is one of the most used approaches for its topology preservation property. However, this approach is sensitive to image noise and produces inexploitable skeletons. A first parameterization of the thinning process has been proposed in the literature to reduce noise-related information. The first contribution of this work is to propose an adjustment of this parameter based on a statistical decision. To this end, a hypothesis test is identified for each lowering criterion in order to set the thinning parameter locally. This leads us to propose the Self Contrast Controlled Thinning method SCCT that is robust to noise and is automatically adjusted to image contrast. The SCCT is made available to application domains through its optimized implementation based on hierarchical queues. Noticing the lack of efforts to assess grayscale skeletonization, the second contribution of this work is to propose a quantitative evaluation protocol assessing skeletonization with regard to its fundamental properties that are namely the preservation of topology and geometry. This protocol is conducted over a synthetic images database and allows us to compare SCCT to approaches from the literature. The third contribution consists in structuring the skeleton into a graph that gives access to objects structural and morphometric descriptors and enables the exploitation of the skeleton by experts from various fields of application. This structuring is applied in the context of Voxelo project which is coordinated by B2OA laboratory of the University Paris Diderot. In this context, descriptors of bone microarchitecture quality are extracted from X-ray high resolution images.