Submitted by Irene Nascimento (irene.kessia@ufpe.br) on 2015-03-06T19:38:17Z No. of bitstreams: 2 DISSERTACAO_DIGITAL_ALMIR_BUARQUE_FEV_2012.pdf: 7439176 bytes, checksum: 3b4db72fb5cc5f5cf69bc769fc7dc50e (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-03-06T19:38:17Z (GMT). No. of bitstreams: 2 DISSERTACAO_DIGITAL_ALMIR_BUARQUE_FEV_2012.pdf: 7439176 bytes, checksum: 3b4db72fb5cc5f5cf69bc769fc7dc50e (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Previous issue date: 2012-02-27 / Nos últimos anos, o paradigma de desenvolvimento de software dirigido a modelos tem se consolidado e ganho popularidade. Desta forma, um produto de software pode ser obtido através de transformações dos modelos de mais alto nível de abstração para os de mais baixo nível, sendo essa transformação guiada por metamodelos. Por exemplo, requisitos iniciais de um sistema de software podem ser transformados em modelos conceituais deste sistema. Observamos que, tanto a abordagem de modelagem de requisitos i* (iStar) como a linguagem de modelagem conceitual OO-Method (OO-M), têm ganho notoriedade no âmbito acadêmico e industrial. Nesta dissertação, estamos interessados em transformar modelos de requisitos iniciais, descritos em i*, em modelos conceituais especificados em OO-Method. Em particular, investigaremos como os requisitos não-funcionais (NFR) de produto, relacionados a softgoals internos a um ator sistema do i*, poderão ser analisados nas abordagens que transformam modelos de requisitos i* em modelos conceituais descritos em OO-Method. Erros e omissões dos requisitos não-funcionais têm impacto bastante negativo na qualidade do produto final: custos elevados de correção pós-implantação e inviabilização do projeto de software. Esses requisitos estão entre os mais caros e difíceis de corrigir. Nesta dissertação, é apresentada a proposta OOM-NFR, que analisa os softgoals internos a um ator sistema em modelos i* para verificar conflitos, prioridades e satisfação dos mesmos. Além disso, com base nos metamodelos i* de origem e OO-Method de destino, a proposta especifica um processo de transformação contendo regras que convertem uma instância de um modelo i* em outra instância do modelo OO-Method. Com o processo OOM-NFR, o modelo OO-M gerado reflete as prioridades dos softgoals especificados em i*. Para ilustrar e validar a proposta, foi concebido um exemplo de uma aplicação baseada num sistema real, utilizando-se a linguagem de transformação de modelos Query/View/Transformation (QVT) para automatizar e implementar essas transformações.

Submitted by Pedro Henrique Rodrigues (pedro.henriquer@ufpe.br) on 2015-03-04T17:08:03Z No. of bitstreams: 2 DISSERTACAO_DIGITAL_ALMIR_BUARQUE_FEV_2012.pdf: 7439176 bytes, checksum: 3b4db72fb5cc5f5cf69bc769fc7dc50e (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-03-04T17:08:03Z (GMT). No. of bitstreams: 2 DISSERTACAO_DIGITAL_ALMIR_BUARQUE_FEV_2012.pdf: 7439176 bytes, checksum: 3b4db72fb5cc5f5cf69bc769fc7dc50e (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Previous issue date: 2012-02-27 / SERPRO / Nos últimos anos, o paradigma de desenvolvimento de software dirigido a modelos tem se consolidado e ganho popularidade. Desta forma, um produto de software pode ser obtido através de transformações dos modelos de mais alto nível de abstração para os de mais baixo nível, sendo essa transformação guiada por metamodelos. Por exemplo, requisitos iniciais de um sistema de software podem ser transformados em modelos conceituais deste sistema. Observamos que, tanto a abordagem de modelagem de requisitos i* (iStar) como a linguagem de modelagem conceitual OO-Method (OO-M), têm ganho notoriedade no âmbito acadêmico e industrial. Nesta dissertação, estamos interessados em transformar modelos de requisitos iniciais, descritos em i*, em modelos conceituais especificados em OO-Method. Em particular, investigaremos como os requisitos não-funcionais (NFR) de produto, relacionados a softgoals internos a um ator sistema do i*, poderão ser analisados nas abordagens que transformam modelos de requisitos i* em modelos conceituais descritos em OO-Method. Erros e omissões dos requisitos não-funcionais têm impacto bastante negativo na qualidade do produto final: custos elevados de correção pós-implantação e inviabilização do projeto de software. Esses requisitos estão entre os mais caros e difíceis de corrigir. Nesta dissertação, é apresentada a proposta OOM-NFR, que analisa os softgoals internos a um ator sistema em modelos i* para verificar conflitos, prioridades e satisfação dos mesmos. Além disso, com base nos metamodelos i* de origem e OO-Method de destino, a proposta especifica um processo de transformação contendo regras que convertem uma instância de um modelo i* em outra instância do modelo OO-Method. Com o processo OOM-NFR, o modelo OO-M gerado reflete as prioridades dos softgoals especificados em i*. Para ilustrar e validar a proposta, foi concebido um exemplo de uma aplicação baseada num sistema real, utilizando-se a linguagem de transformação de modelos Query/View/Transformation (QVT) para automatizar e implementar essas transformações.

No description available.

Evolution and Adaptability of Complex Applications Abstract In these days the applications become more complex that causes maintenance problems while evolving these applications. A change in one part of the applica- tion can significantly affect other parts of the application. The next aspect can be related systems which communicate with this application. They must be up- dated to satisfy their correct functionality. These problems can concern multiple domains, e.g., UML diagrams, XML schema diagrams, relational schemas, APIs, etc. We focus on this problem from the perspective of the MDA, which uses the platform independent model (PIM) for a general view of the problem and the platform specific model (PSM) for particular domains. Moreover, these models can be interconnected and related to each other. We propose novel PSM models from various widely used domains, operations over these models and algorithms for model transformations. Thanks to the MDA principle, it is possible to combine presented models and model a complex application. All models and related algorithms we present were experimentally implemented and tested in the DaemonX framework on real-world data for their verification. 1

Molecular profiling of blood for several purposes, 1) To identify prostate cancer biomarkers, 2) to identify commonalities between asthma and mood disorders, and 3) to identify mRNAs that may be involved in psychobiotic changes to behaviour. / Every cell of the body has the opportunity to secrete molecules into the blood. These molecules: proteins, RNAs, and DNAs, can be secreted freely, or within extracellular vesicles (EV). The complement of specific molecules secreted by cells can vary in accordance with changes to their immediate environment, such as disease in a particular organ. Cells of the immune system which circulate in the blood may also change the rates at which they produce these molecules in response to a disease or unusual event occurring somewhere within the body. The full complement of proteins, RNAs, or DNAs from all sources within the blood can therefore be measured to garner information about disease states and communication between every tissue of the body. In this body of work, we leveraged this to address three separate challenges within medical science. First, we utilized blood as a source of biomarkers for disease and disease severity; isolating EVs from the blood of prostate cancer patients and healthy subjects and characterized their proteins with mass spectrometry to identify potential biomarkers for prostate cancer and its stages. Next, we explored the ability of blood to identify commonalities between distinct but often comorbid diseases; here we utilized publicly available datasets to identify transcripts or gene sets potentially facilitating the relationship between PTSD, MDD, and asthma. Finally, we utilized differential gene and gene sets expression to gain mechanistic insight into microbiota-gut-brain axis; investigating the hippocampus and blood of mice fed one of two psychobiotic bacteria: Lactobacillus rhamnosus JB1, Lactobacillus reuteri 6475. The analysis identified several mRNA expression differences potentially responsible for the mood-altering characteristics of these psychobiotic bacteria. This body of work illustrates the utility of blood omics data for addressing many problems within medical science, and highlights the large scale of information stored within the blood. / Thesis / Doctor of Philosophy (Medical Science) / Every cell of the body has the opportunity to secrete molecules into the blood. These molecules: proteins, ribonucleic acids (RNAs), and deoxyribonucleic acids (DNAs), can be secreted freely, or within small membrane compartments called extracellular vesicles (EV). Specific molecules are secreted more or less by cells depending on changes to their immediate environment, such as disease in a particular organ. We leveraged this to the benefit of medical science in three separate scenarios: 1) using the molecular contents of EVs to determine when someone has prostate cancer, and at what stage; 2) examining RNAs of the blood to determine why so many with asthma also have depression or PTSD; 3) measuring RNAs in the blood and hippocampus of mice to better understand how certain bacteria in the gut can alleviate depression. This work illustrates the utility of blood in tackling many challenging problems within medical science.

No description available.

No description available.

Major Depressive Disorder (MDD) is a severe psychiatric disorder affecting approximately 12% of men and 25% of women nationally in the nonclinical population. The aim of this research was to determine if lay individuals could differentiate between MDD and normal sadness. To evaluate the lay understanding of MDD and normal sadness, students at a large Southeastern university read four vignettes describing varying severity levels of MDD and normal sadness and then answered a variety of questions relating to the vignettes. Additionally, the lay conceptualizations of MDD were compared and contrasted to the professional conceptualizations of MDD. The principal hypothesis was that lay individuals could not differentiate between clinical depression and normal sadness because the two concepts have become synonymous in today’s society. In fact, results showed that lay individuals could not differentiate between threshold MDD and subthreshold MDD.

This thesis presents research investigating objectively and subjectively examined cognitive impairment in Bipolar Disorder (BD) in comparison to disorders with similar cognitive symptomatologies. First, a systematic review and meta-analyses compared the cognitive performance between BD and Mild Cognitive Impairment (MCI) or dementia. Studies included in this review and meta-analyses assessed cognitive performances using multiple objective cognitive assessments. Results from these meta-analyses found greater impairment in BD relative to MCI on motor initiative abilities. Additionally, there were similarities in cognitive deficits on delayed memory recall and visuoconstructional abilities between BD and MCI. For the comparison between BD and dementia, we analyzed the findings of studies comparing BD across different mood states with different types of dementia, where BD in acute mood episode demonstrated greater deficits in attention, working memory, verbal memory, and executive function than behavioral variant frontotemporal dementia (bvFTD). In contrast, overall cognitive functioning and verbal fluency was more impaired in Alzheimer’s disease (AD) in comparison to BD during euthymia. Next, we shifted the focus on examining subjective cognitive complaints in BD relative to Major Depressive Disorder (MDD). Our study is unique from previous literature with the same aim considering that it only involved patients recently diagnosed with BD, and subjective complaints were assessed with the Cognitive Complaints in Bipolar Disorder Rating Assessment (COBRA), an instrument specific to cognitive complaints detected in BD. The findings demonstrate higher subjective cognitive complaints in euthymic BD in comparison to euthymic MDD, suggesting greater self-perceived difficulties in BD, even in the beginning of the illness. Taken together, findings from the studies presented in this thesis highlight the importance of early detection and intervention of cognitive impairments in BD, with the aim of enhancing cognitive abilities, and prevention of further cognitive degradation with the progression of the disorder / Thesis / Master of Science (MSc)

<p> Major Depressive Disorder (MDD) is a common affective disorder associated with persistent states of negative mood and selective cognitive impairments. Fronto-temporal dysregulation in MDD patients is thought to contribute to the symptoms seen in these patients.</p> <p> Based on prior evidence of structural and functional alterations in the hippocampus (Hc) and prefrontal cortex (PFC) in MDD patients, we were interested in examining the changes in cerebral function that underlie the cognitive dysfunction seen in two different MDD populations. We studied psychotropically naïve depressed patients experiencing their first treated episode (FTE) of depression, MDD patients who had experienced multiple past treated episodes (MTE) of MDD and healthy controls.</p> <p> Two functional magnetic resonance imaging studies (fMRI) were conducted. The first study used an Hc dependent process dissociation task to examine Hc activation during recollection memory. The second fMRI study examined the activation in the PFC during reward and punishment conditions of a reversal-learning paradigm. Finally, we conducted magnetic resonance spectroscopy scans to measure levels of metabolites indicative of neuronal and glial cell integrity in the Hc of depressed patients and controls. </p> <p> We observed differing results across all three studies in our FTE and MTE depression groups. Our studies examining the Hc suggest that MTE patients have decreased activation in this region as well as corresponding memory errors during recollection memory. Additionally, these patients have smaller Hc volume and signs of increased neuronal membrane turnover. Conversely, our FTE patients displayed heightened Hc activation without memory deficits. Moreover, FTE patients had signs of increased glial cell density in the Hc without volumetric differences in this region. Our examination of reward processing revealed several health-to-illness gradients of activation in areas as the nucleus accumbens, anterior cingulate and ventral prefrontal cortices during the processing of rewards and punishers. </p> <p> These findings suggest that several regions in the brain may be sensitive to the impact of disease burden and repeated episodes of MDD. In the Hc, first treatment patients may engage in compensatory processes during the early stages of illness that are attenuated with repeated episodes of illness. Moreover, reward processing may be affected in the early course of the disorder, however with a protracted course of illness these regional alterations in activation become more pronounced.</p> / Thesis / Doctor of Philosophy (PhD)