University of Maryland, Baltimore County
Not a member yet
17643 research outputs found
Sort by
The (Conference) Interpreter: An Intercultural Mediator at the Crossroads of Competing Narratives. An Intercultural Communication Approach
No full textThrough an ethnography of the Community of Practice ‘Spanish Conference Interpreters in Brussels’, I gained insight into the translation and communication problems interpreters face derived from intercultural differences. This thesis demonstrates the need for an explicit inclusion of Intercultural Communication into the field of Translation and Interpreting Studies and more specifically into training programs for so&ndashcalled ‘conference interpreters’. To illustrate my claim I provide two case studies of two very common interpreted communicative events in the Brussels’ market that are mediated by the same community of interpreters. The events are: Free Trade Agreements (FTA) negotiations between European and Latin American government delegations and Latin American Indigenous Peoples public hearings at the European Parliament at which those very signings of the FTA’s are denounced. This demonstrates that interpreters are oftentimes at the crossroads of competing discourses informed by different ideologies, cultures, assymetrical power relationships, and the particular social and situational context. All of these factors exert pressure on the development of an interpreted communicative event. In its response to the “cultural and narrative turn” in the field of Interpreting Studies, this thesis advocates the inclusion of ethical, pragmatic, ethnographic and political–economic aspects of critical intercultural communication in the training of future (conference) interpreters in today’s globalized world in which discourses compete for power and legitimacy.Includes 2 .wmv video files
Independent Vector Analysis: Theory, Algorithms, and Applications
No full textThe field of blind source separation (BSS) is a well studied discipline within the signal processing community due to its applicability to a variety of problems when the data observation model is poorly known or difficult to model. For example, in the study of the human brain with functional magnetic resonance imaging (fMRI), a neuroimaging sensor, BSS algorithms are able to provide medical researchers and practitioners with a decomposition of a three-dimensional `movie' of the brain that is amenable to analysis. BSS algorithms achieve this decomposition with only a few justifiable assumptions; this is contrary to methods based on the general linear model, which require prespecified models of the expected or desired response to achieve analysis of fMRI data. Most BSS algorithms consider just a single dataset, but it also desirable to have methods that can analyze multiple subjects or data collections in fMRI jointly, so as to provide insights beyond that achieved with individual analysis of single datasets. Several frameworks for using BSS on multiple datasets jointly have been proposed. The subject of this dissertation is the study of one of these frameworks, which has been termed independent vector analysis (IVA). IVA is a recent extension of the classical independent component analysis (ICA) model to BSS of multiple datasets and it has been the subject of significant research interest. In this dissertation, we provide a formulation of IVA that accounts for sources which possess properties such as a) following Gaussian or non-Gaussian distributions; b) samples are independently and identically distributed (iid) or are dependent; and c) having either linear or nonlinear dependence of sources between datasets. The proposed IVA formulation utilizes the likelihood to define the objective function. This formulation admits to theoretical analysis. In particular, we provide the identification conditions, i.e., we determine when the sources can be `blindly' recovered by IVA, and give a lower bound on the source separation performance. Several algorithms exist for achieving IVA. We provide several new approaches to developing IVA algorithms and apply these approaches using a Gaussian distribution source model and a more general Kotz distribution model. The former, in addition to leading to efficient IVA algorithms, serves as the distribution model that directly connects canonical correlation analysis (CCA) and ICA
ELUCIDATING A COMMON MECHANISM IN WORKING MEMORY PATHOPHYSIOLOGY IN SCHIZOPHRENIA: A COMBINED FUNCTIONAL MRL-COMPUTATIONAL MODELING APPROACH
No full textCognitive dysfunction in schizophrenia is the most frequent and most stable symptom/sign over time and the degree of this dysfunction is the best predictor of long-term functional outcome. Despite impairing socio-occupational aspects of the individual and generating a burden for family and society in general, there is no effective pharmacological treatment for this cognitive dysfunction. Advances in the field of neuroimaging have increased our understanding about cognitive processes to the extent that functional magnetic resonance imaging (MRI) has been instrumental in identifying working memory as an important factor in the cognitive dysfunction in schizophrenia. Despite the improved understanding of brain mechanisms underlying aberrant cognitive processes in schizophrenia and the numerous neurochemicals involved in its pathophysiology, there is little information as to how to link the functional neuroimaging results of working memory (WM) impairment, and the known pathophysiology in schizophrenia. An important obstacle to understand this link is the different behaviors of the signal in the prefrontal cortex when comparing healthy and schizophrenia groups in neuroimaging studies of WM. Can different prefrontal signal behaviors(hypofrontality, hyperfrontality, both, or none) be explained by the differential WM interference caused by the interaction of the different Gamma-Aminobutryic Acid (GABA) pathology with different inter-stimulus interval (ISI) timings? This Ph.D. work successfully answered this research question by combining experimental functional MRI and theoretical computational approaches. Comparing healthy and schizophrenia groups in a functional MRI study using a parametric manipulation of phonological and orthographic interference during a one-back task at two different ISI (1000 and 1600 milliseconds), I provided evidence that interaction of group x interference x ISI generates different prefrontal outcomes. Simulating parametric perturbations to different prefrontal GABA interneurons in a working memory model at different ISI, I generated different outcomes of the modeled neuroimaging signal in the prefrontal cortex model. By generating different prefrontal cortex activity outcomes when comparing healthy versus schizophrenia in both approaches (experimental and theoretical), I was able to suggest a link between functional neuroimaging results and underlying pathophysiology in schizophrenia
Simultaneous Feature Acquisition and Cost Estimation
No full textThis thesis will address classification problems with two sources of cost: the cost of acquiring feature values and the cost of incorrect classifications. In particu- lar, I address problems with feature costs and instance-dependent misclassification costs. Many real-world applications, such as medical diagnosis, contain both feature acquisition costs and instance-dependent misclassification costs. The goal of my re- search is to minimize the total cost of classifying an unknown instance. This goal is accomplished with a new approach: Simultaneous Feature Acquisition and Cost Estimation (SFACE), which combines feature acquisition methods with a regression algorithm that estimates misclassification costs. The estimated cost values are used to estimate the expected cost reduction for the acquisition of each feature. SFACE is evaluated by comparing the total cost of operation to the cost incurred by existing cost-insensitive, cost-sensitive, and feature acquisition algorithms. The results show that SFACE results in lower total cost for the tested datasets
Working on Many Levels: A History of Second-Wave Feminism in Baltimore
No full textThis thesis is a place-based social and intellectual history of the second-wave feminist movement that lived and worked in Waverly and Charles Village, two adjacent neighborhoods in northeastern Baltimore. 'Working on Many Levels:' Second-Wave Feminism in Baltimore investigates the events and institutions organized and created by feminist and other social movement actors, and analyzes the political content of their activism and discourse. Sources include oral history interviews, underground and mainstream newspaper articles, and a plethora of archival materials including neighborhood studies, feminist organization meeting minutes, position papers, flyers, and more. The thesis argues that the largely white self-defined feminism that emerged and grew up in Baltimore, in this neighborhood in the northeastern part of the city, through its interactions with others and with place, became a staunchly antiracist socialist feminism that did not insist on gender as the primary oppression and instead incorporated gender, race, and sexuality into its anti-capitalist, anti-imperialist analysis. Through their involvement with antiwar and black liberation organizing in Baltimore and elsewhere, the feminists cultivated an increasingly militant political analysis that came to view gender, racial, sexual, and economic struggles as inextricably linked. The feminists rooted themselves and enacted their politics in this mixed-income and mixed-race neighborhood in Baltimore during a period of economic decline, instead of fleeing to the suburbs like much of the city's white population. Further, they recognized the tendency of white leftist organizations to create insular communities of privilege sheltered from an oppressive society, and so created accessible institutions and ideologically sound organizing bodies that worked to embody their politics but also to serve the community. In this neighborhood, these institutions and projects included a the first nationally distributed second-wave feminist journal, Women: A Journal of Liberation, the People's Free Medical Clinic, a couple of food co-ops, a Lesbian Community Center, Bread an Roses Coffee House, the women's 31st Street Bookstore, a daycare called Red Wagon Child Center, and more. These prefigurative institutions were seen as only a part of a greater societal transformation, but served to help people survive in the meantime. `Working on Many Levels' recounts how these feminists formulated (and re-formulated) a politics that refused universality and equality, and instead embraced difference and proponed justice, responding to issues that came up throughout the course of their organizational life
ON OPTIMIZING CONTRAST QUALITY AND ACQUISITION TIME OF SSFP-SQUENCE-BASED TECHNIQUES FOR STRUCTURAL AND FUNCTIONAL MR IMAGING VIA EXTENDED PHASE GRAPH (EPG) ANALYSIS
No full text"Structural and functional magnetic resonance imaging (MRI/fMRI) has been an excellent neuroimaging tool, relying on which neuroscientists have been able to define different brain functional regions and measure neural activities within those regions. As it stands now, the most commonly used functional MRI technique, Echo Planar Imaging (EPI), performs reasonably well, but this imaging technique faces a number of long-standing issues. Specifically, the EPI images suffer from two types of image artifacts, signal dropout and geometric distortion, especially when images are acquired from brain regions around the air/tissue interface. In addition to the known image artifacts, it is understood that the technique does not provide functional images with the optimal resolution to capture spatially distributed fine neural activation. For structural imaging, many different techniques are available to allow delineation of detailed brain regions. A particular technique, Magnetization-Preparation RApid Gradient Echo (MP-RAGE), has received much attention for its potential of providing structural images with great tissue contrast and obtaining them with rapid imaging time. However, this technique has not been fully optimized in terms of image quality and acquisition time. Our main goal is to improve/optimize both structural and functional MRI in order to contribute to the study of neurophysiology. In light of our motivation, we have designed an optimization framework for MP-RAGE to enable acquisition that satisfies specified image quality criteria of signal intensity and contrast ratio of/between different tissues in the shortest amount of time. It allows us to perform highly quantitative structural imaging and customized imaging for any individuals. We have also designed a novel functional MRI technique called Frequency-Modulated TRUe Fast Imaging with Steady-state Free Precession (FM TruFISP) and different post image processing techniques to take advantage of the FM TruFISP functional image acquisition. The combined use of the new acquisition and processing techniques allow us to address the issue of the EPI image artifacts and perform highly quantitative functional imaging. The MP-RAGE optimization framework and the FM TruFISP design are discussed in detail in the body of this dissertation, and the Appendix B includes the developed Matlab programs that perform magnetization (signal) calculation/simulation necessary for applying the framework and the different post image processing techniques.
Developing Improved Mouse Models of Prostate Cancer
No full textGenetically modified rodent models provide a platform to dissect the complex and multifactorial mechanisms of prostate cancer initiation and progression. Existing mouse models based on MYC overexpression, including Lo-MYC and Hi-MYC strains, show distinct features of prostatic intraepithelial neoplasia (PIN) and adenocarcinoma that are quite similar to those observed in humans. However, the cancers that emerge in these models rarely progress to metastasis. Evidence is mounting that multiple molecular changes, including loss of tumor suppressor PTEN and MYC overexpression, cooperate to promote human prostate carcinogenesis. To better understand the pathobiology of this disease, we have modeled multiple genetic changes associated with prostate carcinogenesis by combining conditional loss of Pten and concomitant activation of MYC within prostate luminal epithelial cells. A triple transgenic mouse model in which the androgen independent Hoxb13 promoter drives MYC expression concurrent with conditional loss of Pten (Hoxb13-Myc/Hoxb13-Cre/Pten-l-) was generated. To date, tissues from thirty triply transgenic (Myc+, Cre+, Pten-/-) mice (referred to as BMTs) have been examined for disease progression and by immunohistochemical analyses. By two months of age, BMTs have developed high-grade PIN lesions and have lost expression of the key tumor suppressor Nkx3.1. By four months of age, these mice have developed adenocarcinoma that has metastasized lymph nodes (20/20), liver (13/20), and lung (11/20), mimicking the progression of human prostate cancer. Expression of several key markers including Cytokeratin 18 (CK18), Cytokeratin 8 (CK8), FoxA1, focal AR expression, and loss of p63 in the primary and metastatic tumors provides strong evidence that these tumors are adenocarcinomas of prostate origin. Comparative Genomic Hybridization (CGH) analyses using DNA from primary and metastatic lesions of four different mice were performed. Over 250 DNA copy number (DCN) changes were identified, indicating that genomic instability accompanied disease progression, in parallel with human prostate cancer. This innovative mouse model provides a powerful platform in which the molecular events underlying the progression from PIN through highly penetrant lethal metastatic disease can be systematically analyzed
Stigma and Social Relations in a Dementia Care Unit
No full textDementia is a highly stigmatized and stigmatizing condition affecting more than one in eight older adults in the U.S. People with dementia have been characterized by American society as being not quite whole individuals; they are categorically different, the other. In long-term care (LTC) settings for older adults a high level of surveillance combines with existing prejudices against aging and decline. These settings create a microcosm of culture with the potential for heightened stigma against people perceived to be different, such as those with dementia. Some LTC settings include a dementia care unit (DCU), a separate level or unit designed specifically for people with dementia. DCUs have the potential to both exacerbate and shield residents with dementia from the stigmatizing attitudes of others. Utilizing qualitative research methods, this dissertation sought to address three specific aims examining: 1) the stigma of admission to and residence in a DCU in a multi-level LTC setting; 2) whether and how entry into and residence in a DCU affects residents' social relations; and 3) how staff, visitors, family, and other residents react to residents within a DCU. Ethnographic interviews were conducted with eighteen individuals including staff members, family members, and residents of the DCU over the course of ten months of participant observation. Analysis also included data from a previous study at the research site in order to retrospectively examine stigma. Evidence was found of stigmatizing attitudes toward DCU residents which manifested in lying to residents, the infantilization of residents, and the physical, psychological, and social separation of DCU residents from others at the setting. Many other aspects of stigma were identified and are discussed throughout this dissertation. In addition, residents of the DCU were found to recognize differences among each other and to react to one another on the basis of those differences. Although the DCU is envisioned as a safe, supportive environment for older adults with dementia, it falls short of this ideal. Inasmuch as the stigma against dementia remains part of the greater cultural and social environment, the DCU is likely to remain a stigmatized and stigmatizing setting for residents with dementia