One of the most important challenges in dynamic magnetic resonance imaging (MRI) is to achieve high spatial and temporal resolution when it is limited by system performance. It is desirable to acquire data fast enough to capture the dynamics in the image time series without losing high spatial resolution and signal to noise ratio. Many techniques have been introduced in the recent decades to achieve this goal. Newly developed algorithms like Highly Constrained Backprojection (HYPR) and Compressed Sensing (CS) reconstruct images from highly undersampled data using constraints. Using these algorithms, it is possible to achieve high temporal resolution in the dynamic image time series with high spatial resolution and signal to noise ratio (SNR). In this thesis we have analyzed the performance of HYPR to CS algorithm. In assessing the reconstructed image quality, we considered computation time, spatial resolution, noise amplification factors, and artifact power (AP) using the same number of views in both algorithms, and that number is below the Nyquist requirement. In the simulations performed, CS always provides higher spatial resolution than HYPR, but it is limited by computation time in image reconstruction and SNR when compared to HYPR. HYPR performs better than CS in terms of SNR and computation time when the images are sparse enough. However, HYPR suffers from streaking artifacts when it comes to less sparse image data.

The objective of this work was to develop an analysis framework for the structural design of the Human-Powered Ornithopter (HPO). This framework was used in a kinematicaerostructural optimizer for apping-wing ight (Ornithia), as well as analytically to design the HPO, and focused on three goals. First was the development of an accurate and computationally inexpensive nite-element method, to be integrated with Ornithia, which would capture the geometric nonlinearity of the aerostructural interaction of the wing when subjected the large deformations in ight. Second was the assembly of a model by which the aircraft primary structure, the wing main spar especially, could be exactly characterized and designed. Third was the establishment of a process and toolbox for failure analysis which could be applied universally in the design of the HPO. The validation and tuning of these models involved extensive testing on prototype carbon ber composite components.

Background: Some studies have suggested that the risk of preeclampsia in HIV positive pregnant women has increased since the use of HAART became routine. There is also a concern that HIV positive women on HAART have a higher risk of adverse fetal outcomes compared to HIV negative women. Methods: In this matched retrospective cohort study, the risk of preeclampsia and adverse fetal outcomes was examined in 91 HIV positive pregnant women receiving HAART and 273 HIV negative pregnant women. Multivariate logistic regression models were used to adjust for confounding factors. Results: The risk of preeclampsia and preterm birth did not differ significantly between HIV positive and HIV negative women. HIV treated with HAART was an independent predictor for giving birth to a low birthweight baby. Conclusions: HIV positive women on HAART do not have a higher risk of preeclampsia. They do however have a higher risk for lower birthweight infants.

The objective of this work was to develop an analysis framework for the structural design of the Human-Powered Ornithopter (HPO). This framework was used in a kinematicaerostructural optimizer for apping-wing ight (Ornithia), as well as analytically to design the HPO, and focused on three goals. First was the development of an accurate and computationally inexpensive nite-element method, to be integrated with Ornithia, which would capture the geometric nonlinearity of the aerostructural interaction of the wing when subjected the large deformations in ight. Second was the assembly of a model by which the aircraft primary structure, the wing main spar especially, could be exactly characterized and designed. Third was the establishment of a process and toolbox for failure analysis which could be applied universally in the design of the HPO. The validation and tuning of these models involved extensive testing on prototype carbon ber composite components.

Background: Some studies have suggested that the risk of preeclampsia in HIV positive pregnant women has increased since the use of HAART became routine. There is also a concern that HIV positive women on HAART have a higher risk of adverse fetal outcomes compared to HIV negative women. Methods: In this matched retrospective cohort study, the risk of preeclampsia and adverse fetal outcomes was examined in 91 HIV positive pregnant women receiving HAART and 273 HIV negative pregnant women. Multivariate logistic regression models were used to adjust for confounding factors. Results: The risk of preeclampsia and preterm birth did not differ significantly between HIV positive and HIV negative women. HIV treated with HAART was an independent predictor for giving birth to a low birthweight baby. Conclusions: HIV positive women on HAART do not have a higher risk of preeclampsia. They do however have a higher risk for lower birthweight infants.

An algorithm for extracting event information from a Coulomb Explosion Imaging (CEI) position sensitive detector (PSD) is developed and compared with previously employed schemes. The PSD is calibrated using a newly designed grid overlay and validates the quality of the described algorithm. Precision calculations are performed to determine how best the CEI apparatus at The University of Waterloo can be improved. An algorithm for optimizing coincidence measurements of polyatomic molecules in CEI experiments is developed. Predictions of improved efficiency based on this algorithm are performed and compared with experiments using a triatomic molecule. Analysis of an OCS targeted CEI experiment using highly charged Argon ions to initiate ionization is performed. The resulting measurements are presented using a variety of visualization tools to reveal asynchronous and sequential fragmentation channels of OCS3+.

Los virus de influenza aviar de alta patogenicidad (IAAP) causan una enfermedad muy severa en pollos, los cuales frecuentemente inducen lesiones en el sistema nervioso central (SNC). Esta tesis recoge los resultados de tres estudios que se llevaron a cabo para determinar el mecanismo de patogénesis y neurotropismo, así como establecer la ruta de entrada al SNC para un virus H7N1 IAAP. En el primer estudio se estableció un modelo animal de infección en pollos libres de patógenos específicos, que consistía en la inoculación intranasal con el virus H7N1 IAAP. Para establecer este modelo, se utilizaron tres diferentes dosis del virus, obteniendo que las dosis más altas producen una enfermedad similar a la reportada para otros virus de IAAP. Además, se observó que las dosis más bajas causan infección demostrada porque con las dosis más bajas, el virus es hallado en muestras de tejido, muestras de heces y secreciones respiratorias. Adicionalmente, se pudo comprobar el alto neurotropismo del virus, ya que aún en pollos inoculados con bajas dosis el RNA viral es hallado en el CNS. La viremia fue detectada a un día post infección (dpi), sugiriendo que está podría ser la vía de diseminación al SNC. En el segundo estudio, se determinó la distribución topográfica del antígeno viral en el SNC durante las primeras horas post infección, lo cual permitió determinar que el virus se disemina de forma simétrica y bilateral en núcleos neurales del diencéfalo, mesencéfalo y rombencéfalo. La distribución del antígeno viral indica que el bulbo olfatorio y los nervios periféricos están involucrados en el proceso de invasión del SNC. El hallazgo de receptores aviares y humanos en las células endoteliales explica porque estas células son tan sensibles a la infección. El RNA viral fue hallado en el líquido cerebro espinal el primer dpi, lo que indica que el virus atraviesa la barrera hemato-encefálica (BHE). En el tercer estudio, la alteración de la BHE inducido por el virus H7N1 IAAP fue demostrado usando tres diferentes métodos que incluye la perfusión intracardial de Azul de Evans, la detección de la extravasación de la proteína del suero IgY, y evaluación del patrón de tinción con el marcador de las uniones fuertes de la BHE, ZO-1 y claudin-1. El antígeno viral fue observado a las 24 hpi en las células endoteliales, mientras que el daño de la BHE fue observado a las 36 hpi y 48 hpi. En resumen, se puede afirmar que el virus H7N1 IAAP se disemina por la vía hematógena durante las primeras horas pi, posiblemente favorecido por la presencia de receptores en las células endoteliales del sistema nervioso central, y poco después daña la BHE durante las primeras horas de infección como se demuestra por la presencia de extravasación del azul de Evans and IgY del suero. / Highly pathogenic avian influenza viruses (HPAIV) cause a very severe systemic disease in chickens, in which is also frequent to find central nervous system (CNS) lesions. In this thesis, three studies were undertaken in order to determine the mechanism of pathogenesis, the neurotropism and establish the route of entry into the CNS use for a H7N1 HPAI virus. In the first study, an animal model was set up that consisted of SPF chickens inoculated intranasally with the H7N1 HPAI virus. To do that, three different doses were used, obtaining that the highest dose induced a disease similar to the produce by other HPAI viruses, moreover, it was also observed that very low doses also cause infection demonstrated because viral RNA was found in tissues samples, faeces and respiratory secretions. Besides, the high neurotropism of this virus was demonstrated because still in chickens inoculated with low doses, viral RNA is found in the brain. Viremia was detected at one dpi, which indicated that the bloodstream is the pathway of viral spreading to the brain. In the second study, the topographical distribution study of the viral antigen during the first dpi was determined, which allow to determine that the virus disseminates showing a symmetrical and bilateral pattern in the diencephalon, mesencephalon and rhombencephalon, whereas in the telencephalon and cerebellum it was multifocal and random. Viral antigen distribution indicates that the olfactory bulb (OB) and peripheral nerves are not involved in the process of virus invasion into the brain. Avian and human influenza receptors were found in endothelial cells which explain why these cells are so sensitive to the infection. Viral RNA was found in cerebrospinal fluid (CSF) at one dpi, indicating that the virus was able to cross blood brain barrier (BBB). In the third study, the disruption of the BBB induce by the H7N1 HPAI was demonstrated using three different methods that include the intracardial perfusion of the tracer Evans blue (EB), detection of the extravasation serum IgY, and evaluation of the pattern of staining of the tight junction proteins ZO-1 and claudin-1. Viral antigen can be observed as early as 24 hpi in the endothelial cells, whereas disruption was detected at 36 and 48 hpi. In summary, it can be asserted that this H7N1 HPAIV disseminates via the haematogenous route early during the infection, favored by the presence of abundant receptors on the CNS endothelial cells, and soon after it disrupts the BBB during the first hours of infection as demonstrated by the presence of EB and serum IgY extravasation.

Coronary artery disease and chronic pain are serious health issues that cause severe discomfort and suffering in society today. Antithrombotic agents and highly potent analgesics play a critical role in improving the recovery process for patients being treated for these diseases. This thesis focuses on the design and study of pellet-based drug dosage forms which allow diffusion-controlled delivery of drugs with the aim of achieving optimal therapeutic outcomes. A wet slurry process was used to mix the drug and the polymer and/or clay precursor excipients into a paste. The pellets were then shaped via ionotropic gelation (alginate hydrogel beads/pellets), extrusion/spheronization (halloysite clay pellets) or geopolymerization. The decrease in the drug diffusion rate in the alginate beads was affected by the drug's molecular size and charge and the characteristics (such as concentration and chemical structure) of the surrounding alginate gel. The halloysite clay pellets provided sustained release of the highly potent drug fentanyl at both gastric pH 1 and intestinal pH 6.8. As expected, crushing the pellets reduced the diffusion barrier, resulting in more rapid release (dose dumping). The use of mechanically strong geopolymer gels was investigated as a potential means of preventing dose dumping as a result of crushing of the dosage form. Variations in the synthesis composition resulted in drastic changes in the microstructure morphology, the porosity, the mechanical stability and the drug release rate. Pore network modeling and finite element simulations were employed to theoretically evaluate the effects of porosity and drug solubility in the geopolymer structure on the drug release process. Fitting the model parameters to experimental data showed that increased average pore connectivity, a greater pore size distribution, and increased drug solubility in the pellet resulted in an increased drug release rate. Furthermore, incorporation of pH-sensitive organic polymers in the geopolymer structure reduced the high drug release rate from the pellets at gastric pH. These results indicate that geopolymers have potential for use in pellet form; both the release rate of the drug and the mechanical stability of the pellets can be optimized to prevent dose dumping.

This thesis contains results of electron-ion recombination processes in atomic ions relevant for plasma applications. The measurements were performed at the Stockholm Refrigerated Electron Beam Ion Trap (R-EBIT) and at the CRYRING heavy-ion storage ring. Dielectronic recombination (DR) cross sections, resonant strengths, rate coefficients and energy peak positions in H-like and He-like S are obtained for the first time from the EBIT measurements. Furthermore, the experimentally obtained DR resonant strengths are used to check the behaviour of a scaling formula for low Z, H-and He-like iso-electronic sequences and to update the fitting parameters. KLL DR peak positions for initially He- to B-like Ar ions are obtained experimentally from the EBIT measurements. Both the results from highly charged sulfur and argon are compared with the calculations performed with a distorted wave approximation. Absolute recombination rate coefficients of B-like C, B-like Ne and Be-like F ions are obtained for the first time with high energy resolution from storage ring measurements. The experimental results are compared with the intermediate coupling AUTOSTRUCTURE calculations. Plasma rate coefficients of each of these ions are obtained by convoluting the energy dependent recombination spectra with a Maxwell-Boltzmann energy distribution in the temperature range of 103-106 K. The resulting plasma rate coefficients are presented and compared with the calculated data available in literature. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 4: Accepted. Paper 5: Accepted. Paper 6: Manuscript. Paper 7: Manuscript.</p>

La tesi esta basada en l'adaptació de l'ús de codons a genomes procariotes, especialment l'adaptació de l'ús de codons a una alta expressió. Hi ha un grup de genomes procariotes, els quals estan sota una selecció traduccional, que tenen un grup de gens amb un ús de codons esbiaixat de la resta de gens del genoma i adaptats a l'abundància dels tRNA. Hem desenvolupat un nou algoritme per a avaluar si un genoma esta sota selecció traduccional i predir els gens altament expressat de tots els genomes sota selecció traduccional. Aquestes prediccions són públiques a la base de dades HEG-DB (http://genomes.urv.cat/HEG-DB), la qual s'ha publicat a la revista Nucleic Acids Research. Les prediccions de gens altament expressats s'han fet servir com a filtre en les prediccions de gens adquirits per transferència horitzontal, ja que els gens altament expressats molts cops son predits com a falsos positius en la predicció de gens adquirits. Amb les dades de la predicció de gens altament expressats, també hem desenvolupat una nova eina Bioinformàtica, anomenada OPTIMIZER (http://genomes.urv.cat/OPTIMIZER) i publicada al Nucleic Acids Research, per tal d'optimitzar l'ús de codons d'un gen per a incrementar la seva expressió en experiments d'expressió heteròloga de proteïnes. També hem estudiat un cas particular d'adaptació de l'ús de codons. El cas de l' 'amelioration', que és l'adaptació de l'ús de codons que pateix un gen inserit en un genoma hoste. Aquest cas l'hem estudiat amb els gens mitocondrials que varen saltar al genoma nuclear i varen haver d'adaptar el seu us de codons mitocondrial a l'ús de codons del genoma nuclear. Per tal d'estudiar l''amelioration', hem desenvolupat un nou índex anomenat CAI esperat (eCAI) i una nova eina Bioinformàtica anomenada CAIcal (http://genomes.urv.cat/CAIcal), que està en procés de revisió a la revista BMC Bioinformatics. Analitzant l'anàlisi de l'ús de codons dels genomes completament sequenciats vàrem realitzar una troballa que s'aparta una mica del tema central de la tesi. Vàrem veure que els genomes que estan adaptats a la (hiper)termofília tenen un patró de l'ús de codons i d'aminoàcids diferent a la resta de genomes (mesòfils). Aquest fet ens ha permès descobrir casos de guany i pèrdua (recents i antics) de la capacitat d'adaptació termofílica en genomes procariotes. Aquests resultats han donat lloc a una publicació a la revista Trends in Genetics. Durant la tesi he realitzat una estada de 4 mesos (Febrer - Juny, 2006) en el laboratori de bioinformàtica del departament de biologia de la universitat nacional d'Irlanda a Maynooth sota la supervisió del Dr James McInerney on vaig desenvolupar un nou programa per a la comparació d'arbres filogenètics anomenat TOPD/FMTS (http://genomes.urv.cat/topd) el qual està publicat a la revista Bioinformatics. / This thesis is based in codon usage adaptation in prokaryotic genomes, especially the codon usage adaptation to a high expression. In genomes under translational selection, the group of highly expressed genes has a codon usage adapted to the most abundant tRNA species. We have developed a new iterative algorithm which predicts a group of highly expressed genes in genomes under translational selection by using the Codon Adaptation Index and the group of ribosomal protein genes as a seed. We have developed a new genomic database, called HEG-DB, to store genes that are predicted as highly expressed in prokaryotic complete genomes under strong translational selection. The database is freely available at http://genomes.urv.cat/HEG-DB and it has been published in Nucleic Acids Research. The predicted highly expressed genes are used as an initial filter to reduce the number of false positives of the Horizontal Gene Transfer Database, due to highly expressed genes are usually false positive in predictions of acquired genes. We have developed a new web sever, called OPTIMIZER (http://genomes.urv.cat/OPTIMIZER), which has been published in Nucleic Acids Research, to optimize the codon usage of DNA or RNA sequences. This new web server can be used to predict and optimize the level expression of a gene in heterologous gene expression or to express new genes that confer new metabolic capabilities in a given species. We have also analyzed an especial case of codon usage adaptation, which is called 'amelioration'. The 'amelioration' is the adaptation of foreign genes to a new genome. This is the case of mitochondrial genes encoded in the human nuclear genome and originally encoded in the proto-mitochondria. To test the 'amelioration' process we have developed an expected value of CAI (eCAI) to find out whether the differences in the CAI are statistically significant or whether they are the product of biased nucleotide and/or amino acid composition and a new bioinformatics tool called CAIcal (http://genomes.urv.cat/CAIcal). We have also analyzed the evolution of thermophilic adaptation in prokaryotes and we suggest that the amino acid composition signature in thermophilic organisms is a consequence of or an adaptation to living at high temperatures, not its cause. Our findings suggest that there have been several cases where the capacity for thermophilic adaptation has been gained or lost throughout the evolution of prokaryotes. These results have been published in Trends in Genetics. During my thesis I have worked for four months in the Bioinformatics Laboratory of the Biology Department at the National University of Ireland under the supervision of Dr James O. McInerney where I developed a new software program to compare phylogenetic trees called TOPD/FMTS (http://genomes.urv.cat/topd), that has been published in Bioinformatics.