UBIR Repository (Univ. at Buffalo)
Not a member yet
5736 research outputs found
Sort by
Auditory Function in CBA/CaJ Mice Following Blast-Induced Traumatic Brain Injury (TBI)
No full textPh.D.In this dissertation, I investigated how blast-induced traumatic brain injury (TBI) affected auditory function in mice. In experiment 1, I determined the effects of TBI on the physiological responses of the auditory brainstem, outer hair cells, and cochlea when presented with simple acoustic stimuli. In experiment 2, I used behavioral methods to evaluate differences in perception of simple and complex acoustic stimuli before and after TBI. Through these experiments, we found that the auditory system of laboratory mice is susceptible to blast exposures. In the physiological experiment, the magnitude of the response in the ear was reduced and the time it took for this response to peak increased in blasted subjects relative to uninjured controls after the blast. The high frequency sensitivity of mice was extremely elevated after blast exposures in most subjects, resulting in a reduced range of audibility. Interestingly, in the behavioral experiment, many mice were still able to detect a pure tone of the frequency that had elicited no physiological response in experiment 1. Mice could also detect ultrasonic vocalizations and tones spanning their normal hearing range after blast exposures, albeit to a lesser degree than prior to the exposures. These results have implications for the ways in which human hearing is affected following a similar exposure, because audiometry in the clinic often mirrors the behavioral and physiological measurements tested here. The variability in behavioral responses between blasted mice is significant, suggesting that clinical measurements at acute time points may not accurately reflect the hearing abilities of each individual. These experiments are unique because of the comparisons between physiology and behavior, but are also the first studies to examine changes over three months following injury acquisition for individuals exposed to a blast.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
The Many Eyes of Middle-earth: Looking at the Gaze in J.R.R. Tolkien's The Lord of the Rings
No full textM.A.This thesis examines how the gaze plays a pivotal role in J.R.R. Tolkien's The Lord of the Rings. While The Lord of the Rings is a classic tale of Elves, Men, Dwarves, and Orcs in a black and white struggle of good and evil with a simple narrative structure, the gaze, being an element of danger that the protagonists must contend with, helps give the narrative not only more weight but a sense of unpredictability as well. In the novel, Tolkien presents the gaze in a series of vignettes, episodes his characters are placed in where the gaze affects them profoundly. Through these vignettes, we are provided with a coherent account of Tolkien's thoughts about the gaze. We find that, according to Tolkien, the gaze is something that is symbolic of phenomenal power, capable of startling, unsettling, transforming, and destroying, as well as putting the self in an intense state of vulnerability. Tolkien's own theories of vision and visuality parallels those that have been advanced by Jean-Paul Sartre and Jacques Lacan. Applying both Sartre and Lacan's theories allows us to understand more completely and appreciate more fully Tolkien's own understanding and application of the gaze. Their ideas or theory can illuminate Tolkien's practice. From the very beginning of Tolkien's epic, all the way to its very end, he skillfully manipulates the gaze as an instrument with which he exercises his theories of vision and visuality. Ultimately, this thesis aims to prove that Tolkien is a serious author of the gaze and that he has something important to teach us about the gaze.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Path of Heaven
No full textPh.D.Path of Heaven for large ensemble is the second piece in the "Path" series. Similar to the other pieces in the same series, Path of Heaven explores fluctuations in timbre, pitches and structure. A portion of the fluctuation is achieved by weaving various unsteady instrumental lines together. Compositional devices such as trills, wild vibrato, pitch bends and glissandi, as well as flutter tongue, various bow positions and overpressure, provide each instrumental line with different degrees of pitch and timbre alterations. The lines with different flickering qualities are then twisted together and this resulted in a shimmering timbre. The fluctuation in the structure is a concept to get away from traditional forms. To achieve a structural fluctuation, the piece is composed measure by measure without any pre-planning. Nevertheless, the structure is not completely aleatoric. The piece has clear contrasting sections that resembles typical forms. However, the proportion of these sections has been altered so that it does not fall into the frame of any of these forms. The piece also consists of a few substantial solo passages to further twist the structure.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Survival and Biomarker Trends for Non-Small Cell Lung Cancer with the Implementation of Cuban Developed Therapies
No full textPh.D.Background: Lung cancer remains one of the most diagnosed cancers worldwide and accounts for the majority of cancer-related deaths. Mortality rates are now on the decline over the past few years due to declining smoking rates and improvements in treatment. Curative treatments are currently only an option for early-stage lung cancer but unfortunately, the majority of lung cancer is diagnosed in advanced stages. Lung cancer screening offers an opportunity for lung cancer to be detected at an earlier stage, leading to a reduction in disease-specific mortality, and potentially all-cause mortality as well. In addition to diagnosing lung cancer earlier through screening, more research is needed to identify biomarkers of lung cancer risk and treatment efficacy. An international collaboration with Havana, Cuba has afforded a new opportunity to evaluate therapies developed in Cuba for the treatment of advanced staged non-small cell lung carcinoma (NSCLC). In the context of early phase clinical trials and new biomarkers, using minimally invasive collection strategies are being evaluated to study the safety and efficacy of CIMAvax-EGF in combination with an FDA-approved therapy. Methods: To better understand the impact of lung cancer screening guidelines, it is important to look at the change in incidence rates when comparing diagnosis data from before and after the release of such guidelines. In Chapter 2, public data available through the Surveillance, Epidemiology, and End Results (SEER) database was utilized to evaluate the change in incidence and hazard ratios of NSCLC before and after the implementation of the lung cancer screening guidelines in 2011 to understand the effects of lung cancer screening and to address issues of screening implementation. Phase 1 trials are primarily conducted to evaluate the safety and tolerability of new therapies or combinations of therapies. Chapter 3 presents the results of a phase 1 clinical trial conducted to evaluate the safety of CIMAvax-EGF in combination with FDA-approved anti-PD1 inhibitor Nivolumab using the “3+3” trial design. Biomarkers from blood and nasal brush samples were analyzed at different timepoints throughout treatment to establish a profile of which markers may be associated with treatment response. Flow cytometry and Luminex assays were used to analyze the blood samples. Multiple cox regression method was used to evaluate association of immunophenotype samples with survival outcomes. Mann Whitney test was also used to evaluate the association with response. To evaluate less invasive methods for biomarker development, bulk RNA sequencing was used to analyze the nasal epithelium samples in Chapter 4. Results: Incidence rates of NSCLC did not differ greatly after the release of the lung cancer screening guidelines when compared to pre-guideline rates. Survival rates have improved after the release of these guidelines but further research is warranted as there is limited follow-up data due to the more recent diagnosis of cases after the release of the guidelines. When treating advanced staged NSCLC cases, CIMAvax-EGF and nivolumab can be administered at each single-agent dose schedule safely together based on the phase I trial data. The objective response rate was 31% (95% CI 9% to 61%) according to irRECIST. Based on a generalized linear model, there was a significant inverse correlation between induction of anti-EGF antibody titers and depleting serum EGF levels. Within the nasal epithelium, a total of 127 genes were differentially expressed between baseline (pre-treatment) and end of loading dose (post-treatment) samples, p<0.005. The EGF analyte signature seen in the nasal brushings suggested a trend that was associated with the response status of the patients. Initial analysis using pathway enrichment, down-regulated genes were suggestive of EGFR signaling within cancer, the Pi3K pathway, and the TP53 pathway. Conclusions: Broad-based implementation of the lung cancer screening guidelines remains an issue in clinical practice even after definitive research has shown that screening significantly decreases the mortality rates of lung cancer. Further policies need to address how clinical practices can increase the implementation of these guidelines so lung cancer can be caught earlier to further decrease the mortality rate from this cancer. While the Cuban developed therapy of CIMAvax-EGF for lung cancer has shown that it can be administered safety with checkpoint inhibitors, further research is needed to understand how these therapies affect the tumor microenvironment and which biomarkers identify the subgroup of patients most likely to benefit from this drug combination. A phase II clinical trial is currently ongoing that will evaluate overall survival, progression free survival and objective response rates of patients receiving CIMAvax-EGF in combination with immune checkpoint inhibitors. Biomarkers from the nasal epithelium may provide additional information on the effect of CIMAvax-EGF on the lung injury field. Further research is needed to tease out the gene expression changes associated with the individual therapies when administered alone.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Controlling the Global Turn in Art History: Doris Salcedo as Understood in the Global North
No full textPh.D.This dissertation investigates the oeuvre of contemporary sculptor Doris Salcedo, notable as one of the few Colombian artists with a broad international profile. The mechanics of translation as the work moves from a local to a global context is the primary concern. Salcedo's practice is highly responsive to political violence, especially in Colombia. She seeks to make survivors, most often underprivileged women, worthy of empathy by working to make them visible, lending them a specificity that is usually precluded by their status as marginal. Yet, a close analysis of the discourse surrounding her work demonstrates that once she garnered international prestige, her work was pervasively universalized and its historical specificity effaced. Art historians and critics from the US and Europe not only neglected to address the situated histories that inform her work but, in contradistinction to the perspective of Colombia's own art historians and critics, framed her according to their own socio-political agendas. The result is that the very raison d'être of the work, making visible the specificities of Colombian violence is itself disappeared in an account of the "disappeared." In fact, no extant art-historical account offers a comprehensive and detailed explanation of Colombia's history nor the dramatic influence of US foreign policy on the violence there. And rarely is a Colombian art critic or historian cited in the wealth of material written on Salcedo. As a result, art history overlooks the intricate and multifaceted play between local and global registers found in Salcedo's work. I demonstrate the way in which Salcedo infuses "universal" forms with local signifiers, while equally insisting that the local, specifically the plight of victims of political violence, are worthy of "universal" attention. Her international prestige thus offers a critical case study in the politics of a global art market that promoted artists under the guise of multiculturalism, while neglecting to provide the necessary cultural background and analysis critical to the very definition of multiculturalism. In my investigation of Colombia's history as refracted through Salcedo's work, I show that despite her intent to make the survivors of violence visible to a global audience, US and European discourse doubly negates the already negated.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Selective Auditory Attention in Auditory Scene Analysis
No full textPh.D.In the present study, I attempted to understand the effects of auditory attention on auditory scene analysis process in budgerigars. Synthetic pure tone sequences and complex natural vocalizations were used in two separate experiments. These experiments were the first comprehensive investigation of the effects of auditory attention on the auditory streaming process in behaving animals, which adds to studies investigating the mechanisms of auditory processing in animal models. In experiment 1, I used modified ABAB... patterned two-stream mixtures and replicated the buildup effect in an objective auditory streaming paradigm in budgerigars. By adding different distractors in an unattended stream, I investigated the effect of attention capture on the auditory streaming process. Sequentially unpredictable distractors in the unattended stream deteriorated the buildup process in birds, and perceptually more salient distractors tended to be more influential. However, making the distractors sequentially predictable mitigated the deterioration effect elicited by attention capture in birds. Different from humans, distractors from the intensity or frequency feature dimension tended to affect birds' attention to the frequency dimension similarly. In experiment 2, birds' sensitivity to spectral or temporal modulations in conspecific vocalizations were measured in conditions requiring different attentional demands. Compared to spectral modulations, birds' sensitivity to temporal modulations were more susceptible to birds' attentional sets. The presence of competing backgrounds extended the birds' attention spans during the task. Distractors appearing right before the target narrowed the birds' attentional spotlight, which facilitated their discrimination of the target. In sum, birds' attentional sets can have a critical influence on their perception of complex auditory scenes, which should be considered in comparative neurophysiological studies.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Morality in the Making of Urban Space
No full textPh.D.This dissertation is a comparative sociology of morality in the making of urban space. In it, I inquire into how various sets of social actors and urban stakeholders draw on, make use of, and deploy moral concepts, ideas, and discourses in evaluating, organizing, and arranging the urban built and social environment. To study how morality figures in animating the architecture, design, and plans for (re)production of urban space, I look at two neighborhoods: Regent Park in Toronto, Canada, and Kampong Glam, Singapore. In Chapter 2, I examine how urban reformers in Toronto during the early 20th-century drew on a moral perspective to assess, judge, evaluate, to identify problems and diagnosing Cabbagetown (and other slum areas of Toronto) as a pathological space inflicted with poor physical and moral health that would spread and infect the rest of the city of Toronto, framing as urban social problem in the public sphere, facilitating and strengthening the case for slum clearance and urban renewal. Chapter 3 explores how, in the second half of the 20th century, public officials, journalists, and community leaders drew on moral concepts to critique Regent Park on the basis of its transformation of its ethnic composition, the architecture and design of its built environment, and representations of the neighborhood as a place teeming with crime, vice, and violence. In doing so, they cultivated fertile ground for arguments for redevelopment, particularly along the lines of dissolving the concentration of pathology via social mix-based revitalization. In Chapter 4, I turn to Kampong Glam and explain how the efforts to re-fashion the neighborhood hinge on two moral logics—one directed toward economic growth and another towards the ideals of nation-building—which at times come into competition and conflict with each other. I explain how stakeholders work to manage and resolve such moral tensions. Chapter 5 discusses what can be learnt from this dissertation, its theoretical and practical significance, and proposes some avenues for potential future inquiry.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
VCO ADC in Electrical Impedance Tomography Measurements and Physical Unclonable Function for Hardware Security Application
No full textPh.D.In this thesis, novel techniques and methods of designing and implementing ring voltage-controlled oscillator (VCO) analog-to-digital converters (ADC)s for the application of electrical impedance tomography (EIT) measurements are presented. All the EIT systems, reported so far and have been implemented by successive approximation register (SAR) ADC, have two main issues of 1) consuming tens of mW power due to such power hungry elements as instrumentation amplifier, ADC driver and SAR logic and 2) kick-back noise. In this thesis, we showed that traditionally use of SAR ADCs in EIT systems can be abolished and replaced by VCO ADCs since our taped-out VCO ADC has reached to stunning and excellent results. The measurement results show that the designated readout circuit, in which ADC playes the most significant role, consumes only 90µW power dissipation bringing about the figure-of-merit (FoM)of 0.13pW/Hz which is 9x better than other state-of-the-art works using SAR ADC. Also, using the VCO-based ADC in EIT system, the measured amplitude and phase error of the phantom impedance have been obtained by only 2.8% and 0.86 degrees which are very competitive to other readout circuits. However, this is not the end of story. Since we have verified the superiority of VCO-based ADCs in EIT systems, the main focus has gone for new design techniques and improving VCO ADC performance. In this regard, we took the advantage of multi-stage noise shaping (MASH) architecture to eliminate the excessive loop delay (ELD) issues of closed-loop architecture which was used in the EIT measurement. We have presented two 1-1 and 1-2 MASH VCO ADC in which the first stages has applied an open-loop VCO ADC to get ride of ELD. In the 1-1 MASH ADC, the second stage designed to be similar to the first stage so that brings about a second-order noise shaping VCO ADC while the 1-2 MASH ADC has a second-order VCO ADC in the second stage to provide ultimately a third-order noise shaping ADC. In VCO-based modulators, mismatch of VCOs are unavoidable and affects the modulators performance. In this regard, we took the advantage of using the VCOs' mismatch to create a physical unclonable function (PUF) response as a hardware secret key. We taped-out this PUF based on the mismatch of VCOs and current steering digital-to-analog converters (DAC)s. Basically, use of VCO as a phase-domain integrator allows amplification of small mismatches with time and and is used to creat an N-bit PUF key from a single PUF cell by quantizing the VCO phase for N cycles. This PUF is highly digital and its energy consumption is expected to improve with technology scaling. The PUF circuit has been taped-out in 65nm technology process with the supply voltage of 1.2V. The PUF chip achieved power consumption of 36µW and the energy/bit of 85 for the 2048-bit PUF response that is more than 12x better than other state-of-the-art works. Intra and Inter-hamming distance (HD) have been measured for the supply and temperature range of 0.9V-1.2V and 0-50C for 4 PUFs and 16 PUFs, respectively. The measured Inter-HD and Intra-HD of 0.4859 and 0.0906 shows the robust reliability of PUF response.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Towards Effective and Controllable Neural Text Generation
No full textPh.D.Deep neural networks have recently achieved remarkable empirical success in text generation tasks. Users demand both effective and controllable generation, which mean respectively to generate high-quality human-level sequences of words, and to control text output in certain aspects to accommodate different practical needs. Researchers have therefore adopted various neural architectures and techniques. Deep latent variable models, especially variational auto-encoders, have played important roles in representation learning of languages, owing to which promising performances have been shown in several tasks. In recent years, self-attention architectures become the main-stream workhorses and have promoted state-of-the-art generation performances by large margins. Given such context, this thesis presents several studies towards effective and controllable neural text generation. The first part studies the main challenge in using traditional RNN or LSTM based deep latent variable models for text generation, which is the notorious "posterior collapse" issue that leads to ineffective usage of input representations. Sample-based representation for natural language is proposed to mitigate the issue and learn better latent features for downstream generation applications. More effective deep latent variable models are trained with mutual information taking into account. The second part deals with controllable generation of open-domain long text in the era of self-attention architectures. When large scale pre-trained language models have shown surprising generation capabilities and the dominant paradigm of "pre-training + fine-tuning" emerges to shine in natural language understanding, the thesis presents long text generation based on pre-trained model with fine-grained control. A new task named "outline to story" is introduced as a test bed and a simple yet powerful baseline model named "fine-tune with special tokens" is proposed. The third part explores deep latent variable models, specifically variational auto-encoders with self-attention architectures. By integrating latent representation vectors with self-attention based pre-trained components, the conditional variational auto-encoder shows effective representation learning performance as expected, and provides a principled way for controlled and conditional generation. This part takes conditional story generation as main test bed and presents satisfying empirical results. Overall, this manuscript advocates to revive the power of representation learning in the era of large-scale pre-trained Transformer-based modeling, in order to achieve both effective and controllable text generation in a principled way.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*
Negotiating the Boundaries of Our Right to Privacy: The Landscape of Privacy Behaviors, Surveillance Capitalism, and Public Policy in the United States
No full textM.A.Scholars from Communication and Economic Behavior have scattered theories about how individuals in the United States negotiate and navigate their privacy in online contexts. This thesis lays out the three leading theories of privacy from the last 30 years: the privacy paradox, communication privacy management, and the privacy calculus. By drawing connections across the theories, we can address the gaps within each and build upon what previous scholars have written to establish a model of privacy behavior. Then after outlining the political theory of surveillance capitalism, the author will draw connections to establish an overarching picture of the surveillance landscape. The final section will be a review of the current federal policy agenda. The depth of the thesis is to draw connections across fields to bring attention to the lack of cohesive literature that is impeding lawmakers' understanding of privacy behaviors. This ultimately leads to a lack of protection for citizens.**To request an accessible version of the file(s) associated with this item, contact [email protected]. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.*