We present techniques for joint modeling of binomial and rank response data using the Bayesian paradigm for inference. The motivating application consists of results from a series of assessments on several primate species. Among 20 assessments representing 6 paradigms, 6 assessments are considered to produce a rank response and the remaining 14 are considered to have a binomial response. In order to model each of the 20 assessments simultaneously, we use the popular technique of data augmentation so that the observed responses are based on latent variables. The modeling uses Bayesian techniques for modeling the latent variables using random effects models. Competing models are specified in a consistent fashion which easily allows comparisons across assessments and across models. Non-local priors are readily admitted to enable more effective testing of random effects should Bayes factors be used for model comparison. The model is also extended to allow assessment-specific conditional error variances for the latent variables. Due to potential difficulties in calculating Bayes factors, discrepancy measures based on pivotal quantities are adapted to test for the presence of random effects and for the need to allow assessment-specific conditional error variances. In order to facilitate implementation, we describe in detail the joint prior distribution and a Markov chain Monte Carlo (MCMC) algorithm for posterior sampling. Results from the primate intelligence data are presented to illustrate the methodology. The results indicate substantial paradigm-specific differences between species. These differences are supported by the discrepancy measures as well as model posterior summaries. Furthermore, the results suggest that meaningful and parsimonious inferences can be made using the proposed techniques and that the discrepancy measures can effectively differentiate between necessary and unnecessary random effects. The contributions should be particularly useful when binomial and rank data are to be jointly analyzed in a parsimonious fashion.

Thesis (M.S.)--University of Michigan, 1962.

Thesis (Doctoral)--Umeå University, Sweden, 1995. / Added t.p. with thesis statement inserted. Includes bibliographical references.

Thesis (M.S.)--University of Michigan, 1962.

Thesis (Doctoral)--Umeå University, Sweden, 1995. / Added t.p. with thesis statement inserted. Includes bibliographical references.

Made available in DSpace on 2015-03-03T11:52:29Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-02-03Bitstream added on 2015-03-03T12:06:37Z : No. of bitstreams: 1 000800427.pdf: 8452943 bytes, checksum: c40ffd64f5262a9010864dacab63d5d7 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A proposta deste estudo foi avaliar a capacidade osteocondutora da forma sólida do copolímero de PLA/PGA em um modelo experimental de enxerto em seio maxilar . Vinte coelhos machos, branco, da raça Nova Zelândia, cada um pesando cerca de 3,0 kg, foram utilizados e foram divididos em dois grupos de acordo com o material de preenchimento do seio, como segue: ósseo autógeno particulado e copolímero PLA/PGA . A área aumentada diferiu entre os grupos após 3, 7 e 15 dias (p = 0,004) . No entanto, os valores tornaram-se semelhante aos 40 dias (p = 0,458). Depois de 3 e 7 dias, a porcentagem óssea foi estatisticamente significativa entre osso autógeno e PLA / PGA copolímero (P = 0,004 e P = 0,004). Após 15 e 40 dias os valores foram semelhantes nos dois períodos (P = 0,087 e P = 0,087 ). Imunohistoquímica confirma os resultados sobre os dados histomorfométricos. Em conclusão, o copolímero de PLA/PGA parece ser adequado como material reabsorvível capaz de induzir o crescimento do osso em defeitos ósseos. Esta observação sugere que o material tem propriedades osteocondutoras também apropriado para aplicação em cirurgia maxilo-facial / The proposition of this study was evaluate the osteoconductive capability of the solid form of PLA/PGA copolymer in an experimental model of maxillary sinus grafting. Twenty male white New Zealand rabbits, each weighing about 3.0kg, were used and were divided into two groups, according to the sinus filling material, as follows: autogenous bone chips and PLA/PGA copolymers. Augmented area differ between the groups after 3, 7 and 15 days (P=0.004). However, the values became similar on day 40 (P=0.458). After 3 and 7 days the percentage of bone was statistically significant between autogenous bone and PLA/PGA copolymer (P=0.004 and P=0.004). After 15 and 40 days the values were comparable in the two periods (P=0.087 and P=0.087). Immunohistochemistry confirms the results on the histomorphometric data. In conclusion, PLA/PGA copolymer seems to be suitable as resorbable material able to induce bone growth in bone defects. This observation suggests that the material have osteoconductive properties also suitable for application in maxillofacial surgery

Orientador: Idelmo Rangel Garcia Júnior / Banca: Wilson Roberto Poi / Banca: Ronaldo Célio Mariano / A dissertação foi escrita em inglês / Resumo: A proposta deste estudo foi avaliar a capacidade osteocondutora da forma sólida do copolímero de PLA/PGA em um modelo experimental de enxerto em seio maxilar . Vinte coelhos machos, branco, da raça Nova Zelândia, cada um pesando cerca de 3,0 kg, foram utilizados e foram divididos em dois grupos de acordo com o material de preenchimento do seio, como segue: ósseo autógeno particulado e copolímero PLA/PGA . A área aumentada diferiu entre os grupos após 3, 7 e 15 dias (p = 0,004) . No entanto, os valores tornaram-se semelhante aos 40 dias (p = 0,458). Depois de 3 e 7 dias, a porcentagem óssea foi estatisticamente significativa entre osso autógeno e PLA / PGA copolímero (P = 0,004 e P = 0,004). Após 15 e 40 dias os valores foram semelhantes nos dois períodos (P = 0,087 e P = 0,087 ). Imunohistoquímica confirma os resultados sobre os dados histomorfométricos. Em conclusão, o copolímero de PLA/PGA parece ser adequado como material reabsorvível capaz de induzir o crescimento do osso em defeitos ósseos. Esta observação sugere que o material tem propriedades osteocondutoras também apropriado para aplicação em cirurgia maxilo-facial / Abstract: The proposition of this study was evaluate the osteoconductive capability of the solid form of PLA/PGA copolymer in an experimental model of maxillary sinus grafting. Twenty male white New Zealand rabbits, each weighing about 3.0kg, were used and were divided into two groups, according to the sinus filling material, as follows: autogenous bone chips and PLA/PGA copolymers. Augmented area differ between the groups after 3, 7 and 15 days (P=0.004). However, the values became similar on day 40 (P=0.458). After 3 and 7 days the percentage of bone was statistically significant between autogenous bone and PLA/PGA copolymer (P=0.004 and P=0.004). After 15 and 40 days the values were comparable in the two periods (P=0.087 and P=0.087). Immunohistochemistry confirms the results on the histomorphometric data. In conclusion, PLA/PGA copolymer seems to be suitable as resorbable material able to induce bone growth in bone defects. This observation suggests that the material have osteoconductive properties also suitable for application in maxillofacial surgery / Mestre

La spectrométrie de masse MALDI est une technique devenue incontournable pour l’analyse des biomolécules et largement employée dans de nombreuses applications. Bien que les performances de cette source soient notoires, celles-ci restent encore limitées en particulier en termes de sensibilité ou encore de résolution spatiale pour des applications telles que l’imagerie MALDI MS. En effet, les rendements de production d’ions sont très faibles en MALDI et chutent encore lorsque la résolution spatiale est augmentée (diminution de l’aire irradiée). Ainsi, les objectifs de ces travaux de thèse ont été de développer et d’étudier deux systèmes permettant d’augmenter la sensibilité. La première partie de ces travaux a donc portée sur la réalisation d’un système permettant la désolvatation des agrégats formés dans le processus MALDI via l’utilisation d’un second faisceau laser interceptant la plume en expansion. Par utilisation d’un laser pulsé émettant à 1064 nm il a été ainsi possible de démontrer une augmentation d’un facteur 2 à 3 de l’intensité des signaux d’analyte. Dans une seconde partie, un système de masques de silicium permettant de réduire les dimensions de la zone irradiée (sans agir sur la focalisation du faisceau laser) a été développé. Les études réalisées pour différentes géométries de ces masques ont permis de démontrer d’une part l’efficacité de ces systèmes pour réduire l’aire de la zone irradiée en coupant le faisceau laser incident tout en maintenant l’intensité des signaux, et d’autre part que certaines géométries particulières permettaient d’obtenir un effet d’augmentation de la sensibilité et de la résolution spectrale. / MALDI Mass Spectrometry is an essential tool for biomolecules analysis and is largely employed in various applications. Albeit performances of this ion source are well known, there are still clear limitations in terms of sensitivity and spatial resolution for applications such as MALDI MS Imaging. Indeed, ion production yields are very low in MALDI and, moreover drop when spatial resolution is increased (decrease of irradiated area). Thus, objectives of this work were to develop and study two systems for improving sensitivity. First part of this work was dedicated to setup a system allowing desolvation of material clusters formed in the MALDI process by using a second laser to intercept the expanding plume. By using a pulsed laser emitting at 1064 nm, it was possible to demonstrate an increase of signal intensity by 2 to 3 folds. In a second part, Silicon mask systems allowing reduction of irradiated area (without involving any focusing of the laser beam) was developed. Studies of various geometries of the mask showed their efficiency to reduce irradiated area by cutting part of the laser beam while maintaining signal intensity. Moreover, for certain geometries it was demonstrated that masks could lead to an increase in sensitivity and spectral resolution.

We document how sampling from a conditional Student's t distribution is implemented in stochvol. Moreover, a simple example using EUR/CHF exchange rates illustrates how to use the augmented sampler. We conclude with results and implications. (author's abstract)

Implants such as intramedullary nails or cancellous screws are used to mechanically stabilize fractures in bone. They provide reinforcement to the bone if they find good purchase in cancellous bone. Not all implants hold enough loads for mechanical stability and pull-out or cut-out may happen in some cases. This is linked to the interface between the bone and the implant. Computer modelling techniques are used to investigate both the effects of cut-out in a femur model, and the pull-out forces of cancellous bone screws. The bone geometry was based on CT scanned cancellous bone and converted using Mimics® software. The finite element models were produced in ANSYS®. Simple bone models were used to examine a fractured femur under standard gait loading. These models were continuum models and idealised the screw to bone interface in order to ease computational demand. The models were used to investigate the ideal positions of intramedullary devices lag screws on an anterior-posterior view of the implant location. In accordance with literature, an inferior-central or central-central position was the best position of the lag screw, while a superior-anterior or inferior-anterior position was adverse. The introduction of multi-scale modelling in order to investigate cut-out with a discrete bone model was not achieved. Discrete cancellous bone models were used to examine some of the cancellous screw characteristics, including pitch, inner diameter and proximal half angle, while a cancellous screw was also studied using a model of cancellous bone with a range of bone densities. The calculated reaction force for a pull-out of 0.2mm shows the influence of some parameters. Change in the proximal half angle increased the stiffness and strength by about 15% in line with the experimental findings of others, while apparent density changes of 2.5% increased the forces threefold. A significant reduction in reaction force was observed when a particular screw geometry in lower apparent density bone was modelled and rotated through 180° on a plane. Examination of the geometry of the bone/screw interface shows that in certain positions there is very little cancellous bone to support the implant. This will lead to low strength and is very difficult to predict. The same models were used to examine the effect of increasing bone stiffness adjacent to the implant and the use of a cement layer to augment the screw model. The increasing stiffness concluded that an increase in pull-out stiffness can be achieved, even in low quality bone, while the cement augmentation showed a significant increase in pull-out strength, though it was idealised as bonded to the bone and screw.