Search

The incidence of oesophageal adenocarcinoma (OAC) has rapidly increased, and its prognosis remains poor. Barrett's oesophagus (BO) is considered the precursor to OAC, however, BO is not apparent adjacent to tumour in nearly half of OAC patients. We have previously demonstrated that patients with BO-adjacent tumours (BO+ve OAC phenotype) have a favourable prognosis compared to those without evidence of BO adjacent to the tumour (BO-ve OAC phenotype). It has been suggested that the BO-ve OAC tumour phenotype may arise independently of BO. Recent experimental and computational studies show that all OACs likely arise from BO, even if this precursor lesion is not histologically apparent adjacent to the tumour. However, there is a lack of consolidated clinical, epidemiological and molecular data to examine this question. In this thesis, I used orthogonal approaches to examine the overlap between the BO+ve OAC and BO-ve OAC phenotypes and explain the aetiology and the observed altered prognosis. To achieve this, I assembled a large cohort (n=4,695) comprising BO+ve OAC cases (n=1,235), BO-ve OAC cases (n= 880), OAC cases with an unascertainable BO status (BO(?) OAC; n= 985), cancer-free BO cases (n=1,091) and reflux controls (n=554). A subset of the OAC cases (n=950) with available clinical, epidemiological and whole-genome sequencing data was also examined. There was little to no association between most of the 34 clinical and epidemiological factors and the OAC phenotypes. Weak associations were observed for cigarette smoking and gender with self-reported ever-smoking and female cases being more likely in the BO-ve OAC phenotype group relative to the BO+ve OAC phenotype group. However, tumour stage, lymph node spread and metastasis (TNM) was strongly associated with increased risk of BO-ve OAC. Higher TNM stage was strongly correlated with BO-ve OAC with an adjusted odds ratio of 2.4 (95% CI:1.8-3.3), 2.9 (95% CI:2.2-3.9) and 3.2 (95% CI: 1.7-5.9) for stages II, III and IV, compared to stage I. The improved survival associated with BO+ve OAC persisted in survival analyses adjusted for the tumour stage and location as well as the effects of smoking, obesity and heartburn (adjusted hazard ratio=0.88, 95% CI: 0.77-0.95, p=0.015). Of note, the BO-ve OAC phenotype was reported for 21 OAC cases with a history of undergoing endoscopic surveillance for BO. Seven different types of genomic alterations were examined in relation to the OAC phenotypes. BO+ve OAC tumours had a slightly higher tumour mutation load relative to BO-ve OAC tumours, which was not explained by the effects of ageing and cigarette smoking. There were no differences in the distribution of driver gene alterations, frequency of whole-genome doubling events, or rates of aneuploidy between the OAC phenotypes. Similarly, the prevalence of complex events such as breakage-fusion-bridges and extrachromosomal DNA did not differ according to OAC phenotype. Importantly, signature 17, which is shown to be preserved across the BO-OAC continuum, was equally enriched among the BO+ve OAC and BO-ve OAC tumours. This thesis presents the first comprehensive evaluation of epidemiological, clinical and molecular factors between the two defined OAC phenotypes. While certain epidemiological factors differ between the phenotypes, they do not explain the observed prognostic difference. The genomic landscapes of these tumour phenotypes were remarkably similar. Taken together, it is likely that all OACs arise from BO even if this precursor lesion is no longer apparent at the time of diagnosis or resection pathology. The advanced-stage tumours suggest that the precursor lesion is overgrown in BO-ve OACs. This work contributes evidence for screening strategies focused on identifying individuals with BO to reduce the public health burden of OAC.

Communication typically requires utilising communication strategies (CSs) of either verbal or non-verbal skills to overcome the linguistic difficulties of the target language (linguistic level) and maintain communication effectiveness (pragmatic level). However, the proliferation of mobile devices has expanded these strategies by offering different features and applications. As the new form of communication strategy emerges, the topic merits further explanation, spurring the present CS research, which aims to explore the use of mobile devices as a communication strategy among ESL/EFL learners who use English as a lingua franca in communication. Existing literature indicates that the studies of CS were traditionally conducted face-to-face, but after the advent of technology, researchers began to use the CMC context on desktops as a setting to research CS. More recently, the use of mobile devices in CS has also been identified in the literature, and language learners have been found to use CS when communicating online on their devices. However, so far, no studies have investigated how ESL/EFL learners use mobile devices as CS in a hybrid context. Therefore, such a study is indeed crucial and necessary to gain a deeper understanding of the use of this device as one of the CSs among this group of people during interaction. Consequently, this present study's focus on the use of mobile devices in a hybrid CS context is noteworthy. To study this underexplored research area, I have chosen the qualitative constructivist perspective as the basis for my study. I also invented an approach which I named as a quasi-natural CS elicitation research approach that combines natural and unnatural elements to elicit CSs. Thirty non-native English language speakers (ESL/EFL) of different nationalities in a Malaysian university were recruited via purposive sampling to work in pairs. I also generated my data by observing and recording participants' interactions in the tasks in natural settings. Dyadic interviews were also carried out among them to gather further information on the CS used in interactions. The findings revealed that the participants used various CSs, namely traditional and digital CSs in communication. Their CS choice and usage were also influenced by factors such as attitude, culture, familiarity between speakers, physical context, and mobile devices properties. They were also found using a wide variety of CSs including mobile devices to communicate various language functions. One of them was using Google Search to gain accuracy of the content knowledge. The use of mobile devices as CSs also affected communication. They utilised multimodal CS, became autonomous communicators, and performed collaboration in communication. The occurrence of silence in interaction, however, was also identified as one of the effects of using mobile devices as a CS in communication. Overall, this study suggests that the use of CSs, including mobile devices, was beneficial to language speakers in communication.

Recent advances in the field of neuroscience have suggested that new generation brain computer interfaces demand a critical step in biomedical signal processing requiring online/on-chip spike sorting. Spike sorting is the process of grouping signals from an individual neuron by grouping action potentials (spikes) into a specific cluster based on the similarity of their shapes. The extraction of single-unit activity by sensors at a distance from specific neurons is necessary for a wide range of clinical applications such as disorder treatments, muscular stimulation (e.g., epidural spinal cord stimulation for treatment acceleration), cochlear implant and neural prostheses. A brain machine interface, for example, can potentially substitute the missing motor pathway/sensory information between the motor cortex and an artificial limb. With the aim of developing an energy-efficient spike sorting chip for hardware implantable systems, this thesis introduces a new feature extraction method based on extrema analysis (positive and negative peaks) of spike shapes and their discrete derivatives. The proposed method runs in real-time and does not require any offline training. Compared to other methods it offers a better tradeoff between accuracy and computational complexity using online sorting. It additionally eliminates multiplications which are computationally expensive, power hungry and require appreciable silicon area. A minimum power limit for implantable neural front-end interfaces is also derived. It involved: 1) system level optimization - the front-end specifications including the bandwidth, data converter resolution and sampling rate were defined by exploring the effect of the parameters on spike sorting via a standard spike bank; 2) block level optimization - The front-end power was minimized by using an opamp-less cyclic converter; and 3) estimating the power limit equation of the frontend. The new optimization methodology addresses the future demands of neural recording interfaces. Finally the thesis presents the design, implementation and testing of the first generation of an adaptive spike sorting processor. It enhances the accuracy-power characteristics by employing self-calibration of processing features. The chip prototype was fabricated in a 180-nm CMOS technology. It achieves an overall clustering accuracy of 84.5% using a standard spike data bank and has a power consumption of 148-μW from 1.8-V supply voltage. The fabricated spike processor has almost 10%higher clustering accuracy than the state-of-the-art. Measurements show good power-performance characteristics compared to the state-of-the-art online and offline clustering methods.

This thesis presents a detailed design, simulation, optical performance, construction and experimental validation carried out on a novel non-imaging static symmetric reflective compound parabolic concentrator (SRCPC). By considering the seasonal variation of the sun’s position, a concentrating Photovoltaic (CPV) system with precise acceptance angle and low concentrating ratio will be an ideal alternative to conventional flat plate photovoltaic (PV) modules in harvesting the power from the sun. The SRCPC is a suitable choice well designed to achieve optimum precise acceptance angles and concentration ratio for this purpose. The optical performance theory study shows that a truncated symmetric reflective CPC with acceptance half-angles of 0° and 10° (termed as SRCPC-10) is the optimum design when compared with the symmetric reflective CPC designs with acceptance half-angles of 0° and 15° and 0° and 20° in Penryn and higher latitudes. An increase in the range of acceptance angles decreases the concentration ratio but an increase in the range of acceptance angles is achieved by truncating the concentrator profile which will reduce its cost as well. Ray tracing simulations indicates that the SRCPC-10 exhibited the maximum optical efficiency and steady slope compared with others. The simulated maximum optical efficiency of the SRCPC was found to be 94%. In addition, the SRCPC-10 was found to have a more uniform intensity distribution at the receiver and a total daily-monthly energy collection compared to the other designs. Thermal modelling of the CPV system with the SRCPC-10 concentrator shows that the solar cell operating temperature can reach up to 70°C for irradiance of 1000W/m2 at an ambient temperature of 25° at a wind velocity of 2.5m/s. The integration of the thermal management system is able to control and maintain the temperature to 29°C. The modelled thermal and electrical efficiencies were 47% and 15% respectively with a heat transfer coefficient of 54.29W/m2K thereby bringing the system efficiency to 62%. The maximum power of the SRCPC-10 when characterised in an indoor controlled environment using solar simulator was 5.96W at 1000W/m2 at a cooling flow rate of 0.0079L/s with average conversion efficiency of 8.97%. The maximum power at 1200W/m2 and 0.031L/s was 7.14W with conversion efficiency of 10.57%. The maximum increase in efficiency from non-cooling to cooling is 2.54%. The efficiency increased because of cooling is relatively 40%. The outdoor characterisation (validation) of the SRCPC-10 shows that the maximum power was 7.4W at 1206W/m2 on a sunny day. The maximum electrical conversion efficiency of the SRCPC-10 in outdoor conditions was found to be 10.96%. These results revealed that this designed SRCPC-10 is capable of collecting both direct and diffuse radiation to generate power. Therefore, the SRCPC-10 could be used to provide a solution to the increasing demand on electricity to the energy mix, leaving a clean environment for future developments.

This thesis is a result of research aimed at analysis of the Taylor Spatial Frames (TSF) in Orthopaedics. The TSF is a ring external fixator, which are used to stabilise a fracture or provide stability during skeletal limb reconstruction procedures. A sound understanding of the mechanics of the fixator is essential, because mechanical stability is a key factor in bone healing. TSF is in fact an adaptation of the hexapod parallel manipulators for dynamisation of the classical ring fixators of Ilizarov type. Therefore, a general solution for Forward kinematics of parallel manipulators was provided and the solution is visualised in real-time. A three-dimensional visualisation tool for TSF, was developed, which offers improvements over the software provided by the manufacturer. Abaqus/CAE programming interfaces were used to develop two separate systems for automatic creation of FEMs of the TSF: one using beam elements and the other using 3D solid elements. The systems were used for a parametric study on axial compression of the TSF. Components of the TSF were also tested and analysed: o TSF rings were studied extensively, which lead to revealing important facts about their role in the TSF. o Fixation bolts in external fixators were studied by FE technique and the results used to relate bolt-load to the bolt-torque applied. o TSF struts were tested in compression and their load-deflection behaviour and the role of universal joints in them were described. TSF and Ilizarov fixators were tested and compared in axial compression. The results highlighted the important role of the pins and wires in deflection of the fixators.

In carrying out the objectives, the main body of the thesis is organized into five chapters: The second chapter will discuss the main macroeconomic trends mostly in post-revolutionary Iran. The role of the oil revenue in the national economy are examined by different graphs. The review of existing literature is discussed in chapter three. This chapter reviews the role of natural resource revenue in general and oil revenue in particular in economic performance of oil producer countries. Our hypotheses are thoroughly investigated by using ARDL Bound test, VECM time series model as well as symmetric and asymmetric methods are reviewed in chapter four. Chapter five analyses the role of institutions, oil and economic growth in Iran. It focuses on the role of institutions on economy and addresses the question: "why is oil revenue a blessing in some countries and a curse in others". Moreover, the impact of oil revenues on institutional quality in Iran has been examined. Chapter six discusses the role of oil stabilizer funds in reducing the negative impact of oil revenue fluctuation on the economy. The relation between performance of stabilizer fund and institutional quality has been assessed. Chapters four, five and six have their own literature review. Chapter seven contains a discussion of the main results emerging from the research. The chapter also presents policy implications, recommendations and directions for further research in this area. The general conclusion emerging from empirical analysis is that oil revenue have positive impact on Iranian economic growth. Moreover, we find oil revenue fluctuations have not asymmetric effects in Iranian economy. As to the effect of institutions, in our models and in all specifications we found that despite the positive effect of institutional quality on growth, resource curse is stronger the weaker of the institutional quality. This research has assessed that whether natural resource revenue in general or oil revenue in particular causes bad institutions? The result shows higher oil revenue in Iran worsen institutional quality. Managing natural resources revenue has been a serious issue in resources rich countries. We studied the impact of oil stabilization fund on inflation rate as a proxy for instability in Iranian economy and found that oil stabilisation fund reduced inflation rate in Iran. Regarding the relation between institutional quality and performance of stabilizer fund, we found that there is high correlation between fund score and institutional qualities which means higher institutional quality results in better fund performance.

In the wake of the US detention policy in the aftermath of the 9/11 attacks, the practice of detention without trial has gained a degree of attention unparalleled in the history of common law tradition. Legal analyses of all kinds have ensued, and countless policy plans and guidelines have been created. Yet, despite the pedigree of detention without trial, the historical dimension to the practice of detention without trial has not been invested with the scrutiny that it deserves. Drawing on the history of detention without trial in Britain, this research seeks to draw a roadmap for the evolving features of detention without trial. It will be argued that it is by virtue of this historical understanding that we can make sense of the modern laws governing the practice of detention without trial and its associated features.

The UK and the world are moving away from central energy resource to distributed generation (DG) in order to lower carbon emissions. Renewable energy resources comprise a big percentage of DGs and their optimal integration to the grid is the main attempt of planning/developing projects with in electricity network. Feasibility and thorough conceptual design studies are required in the planning/development process as most of the electricity networks are designed in a few decades ago, not considering the challenges imposed by DGs. As an example, the issue of voltage rise during steady state condition becomes problematic when large amount of dispersed generation is connected to a distribution network. The efficient transfer of power out or toward the network is not currently an efficient solution due to phase angle difference of each network supplied by DGs. Therefore optimisation algorithms have been developed over the last decade in order to do the planning purpose optimally to alleviate the unwanted effects of DGs. Robustness of proposed algorithms in the literature has been only partially addressed due to challenges of power system problems such multi-objective nature of them. In this work, the contribution provides a novel platform for optimum integration of distributed generations in power grid in terms of their site and size. The work provides a modified non-sorting genetic algorithm (NSGA) based on MATPOWER (for power flow calculation) in order to find a fast and reliable solution to optimum planning. The proposed multi-objective planning tool, presents a fast convergence method for the case studies, incorporating the economic and technical aspects of DG planning from the planner‟s perspective. The proposed method is novel in terms of power flow constraints handling and can be applied to other energy planning problems.

Nitric oxide (NO) is an important biological molecule within the body, which over production of this molecule in response to different stimulations can cause various inflammatory diseases. Over production of this molecule is caused by the induction of the inducible nitric oxide synthase (iNOS) enzyme. This enzyme uses L-arginine as a substrate and therefore the presence and transport of this amino acid into the cells can be a key factor in regulating NO over production. Different signalling mechanisms have been implicated in the regulation of this pathway and one of which involves the Mitogen Activated Protein Kinases (MAPK). This family of proteins respond to inflammatory conditions and may mediate effects induced by inflammatory mediators. Of the MAPKs, the role of the c-Jun-N-terminal kinase (JNK) pathway in the induction of iNOS is still controversial. JNK and its downstream target, the transcription factor Activator Protein-1 (AP-1), have shown contradictory effects on iNOS induction leading to controversies over their role in regulating iNOS expression in different cell systems or with various stimuli. The studies described in this thesis have determined the role of JNK/AP-1 on iNOS expression, NO production, L-arginine uptake and also on the transporters responsible for L-arginine transport into the cells. The studies were carried out in two different cell types: rat aortic smooth muscle cells (RASMCs) and J774 macrophages which are both critically associated with the over production of NO in vascular inflammatory disease states. The first approach was to block the expression of the inducible L-arginine-NO pathway using SP600125 and JNK Inhibitor VIII which are both pharmacological inhibitors of JNK. The results from these studies showed that the pharmacological intervention was without effect in RASMCs, but inhibited iNOS, NO and L-arginine transport in J774 macrophages. In contrast, the molecular approach employed using two dominant negative constructs of AP-1 (TAM-67 and a-Fos) revealed a different profile of effects in RASMCs, where a-Fos caused an induction in iNOS and NO while TAM-67 had an inhibitory effect on iNOS, NO, L-arginine transport and CAT-2B mRNA expression. The latter was unaffected in RASMCs but suppressed in J774 macrophages by SP600125. Examination of JNK isoforms expression showed the presence of JNK1 and 2 in both cell systems. Moreover, stimulation with LPS/IFN- or LPS alone resulted in JNK phosphorylation which did not reveal any difference between smooth muscle cells and macrophages. In contrast, expression and activation of AP-1 subunits revealed differences between the two cell systems. Activation of cells with LPS and IFN- (RASMCs) or LPS alone (J774 macrophages) resulted in changes in the activated status of the different AP-1 subunit which was different for the two cell systems. In both cell types c-Jun, JunD and Fra-1 were increased and in macrophages, FosB activity was also enhanced. Inhibition of JNK with SP600125 caused down-regulation in c-Jun in both cell types. Interestingly this down-regulation was in parallel with increases in the subunits JunB, JunD, c-Fos and Fra-1 in RASMCs or JunB and Fra-1 in J774 macrophages. Since, SP600125 was able to exert inhibitory effects in the latter cell type but not in RASMCs, it is possible that the compensatory up-regulation of certain AP-1 subunits in the smooth muscle cells may compensate for c-Jun inhibition thereby preventing suppression of iNOS expression. This notion clearly needs to be confirmed but it is potentially likely that hetero-dimers formed between JunB, JunD, c-Fos and Fra-1 could sustain gene transcription in the absence of c-Jun. The precise dimer required has not been addressed but unlikely to exclusively involve JunB and Fra-1 as these are up-regulated in macrophages but did not sustain iNOS, NO or induced L-arginine transport in the presence of SP600125. To further support the argument above, the dominant negatives caused varied effects on the activation of the different subunits. a-Fos down-regulated c-Jun, c-Fos, FosB, Fra-1 whereas TAM-67 reduced c-Jun and c-Fos but marginally induced Fra-1 activity. Associated with these changes was an up-regulation of iNOS-NO by a-Fos and inhibition by TAM-67. Taken together, the data proposes a complex mechanism(s) that regulate the expression of the inducible L-arginine-NO pathway in different cell systems and the complexity may reflect diverse intracellular changes that may be different in each cell type and not always be apparent using one experimental approach especially where this is pharmacological. Moreover, these findings strongly suggest exercising caution when interpreting pure pharmacological findings in cell-based systems particularly where these are inconsistent or contradictory.

This work addresses a better understanding of water dynamics in cement paste and mortar materials. Permeability to water and water transport diffusion are measured using magnetic resonance profiling. Water self diffusion in one and two dimensions is measured using nuclear magnetic resonance pulsed field gradient diffusometry. With regard to water permeability of cement paste and mortars, three topics are addressed. The permeability is measured as a function of relative humidity for cement pastes of different sizes and as a function of sand size for mortars. The relative liquid water and water vapour permeability are calculated using a modified version of a model suggested by Baroghel-Bouny et al. The intrinsic permeability values for cement pastes are similar to those measured by beam bending methods by Vichit-Vadakan and Scherer. The values found for mortars are similar to those measured by Baroghel-Bouny et al. The time dependence of the equilibration of the water concentration profile is also measured. With regard to water transport diffusion, mortars with different sand sizes are studied during water ingress/egress cycles. It is shown that ingress of water into dried mortars slows dramatically after some time most likely due to pore blocking by particulate fines carried with the water. Two indicative transport diffusion coefficients are obtained corresponding to the water diffusion before and after the blockage is encountered. With regard to diffusometry, the water self-diffusion coefficient was measured in a cement paste and found to decrease rapidly in the early stages of hydration. The capillary pore size distribution is measured from an analysis of restricted diffusion data in a mature sample. No evidence of capillary pore anisotropy is found.

Exercise based interventions are common in the treatment of musculoskeletal disorders. These interventions have been investigated in other manual therapy disciplines, but little empirical data exists about. osteopathic approaches to exercise. There is a need to examine the role of exercise in osteopathic undergraduate education and the osteopathic treatment and management of patients in order to identify, clarify and develop upon educational and professional needs of the practising osteopath. The aims of this thesis were to explore the integration of exercise therapy in the undergraduate osteopathic curriculum, gain an understanding of the interpretation and uses of exercise therapy in osteopathic practice and explore potential concordance between education and practice. To provide context for the studies in the thesis, preparatory work included . examining the historical and philosophical developments within osteopathy and the utilisation of exercise therapy and physical activity in wider health care provision. Curriculum evaluation of the intended (content analysis of course documents) delivered (faculty member perspective) and received (student perspective) undergraduate osteopathic curricula revealed the idiosyncratic and sporadic nature of documented and delivered exercise content. The intended curriculum was seen to pertain to academic education whilst the delivered and received curriculum was reflective of clinical education. There was evidence of shared desires from osteopathic students and faculty members who suggested that they would welcome a standardised, experiential, patient centred approach to exercise therapy. The interview study with practitioners revealed a patient centred approach to the use of exercise therapy. However there remains some confusion over the use and understanding of exercise terminology. Paradoxically with the stated patient centred approaches practitioners expressed exercise delivery using paternalistic language. Favoured modes of exercise showed common trends with other manual therapies such as the use of 'core stability' programmes, but bore little resemblance to those delivered during undergraduate education. Exercise therapy and its potential for use is a substantial issue for osteopaths and for education providers in the UK. There is some dissonance between clinicians reported patient centred care approaches and actual delivery of exercise advice. Exercise education in osteopathy and clinical practice are not concordant and clinical aspects of undergraduate education are not documented clearly and are largely opportunistic. There is a need for further consensus about the role of exercise therapy in osteopathic practice and this should be a driver for a more coherent approach across education and practice.

In this project various neurochemical and immunological approaches were used to identify and analyse the role of glycoproteins in neural plasticity. Where appropriate, the one-trial passive avoidance paradigm was used (Gibbs & Ng 1977). Initially, the role of a glycosylation inhibitor, 2-deoxy-D-galactose (2DGal), was examined both in vivo and in vitro. The results of these studies agreed with previous reports that 2DGal inhibits incorporation of fucose and galactose into glycoproteins. However, 2DGal inhibited fucosylation and galactosylation of only a sub-set of these macromolecules. The localization of these glycoproteins in synaptic plasma membranes (SPMs) and postsynaptic densities (PSDs) suggested differential glycoprotein synthesis and distribution across the synaptic cleft. Subsequent studies attempted to identify the fucosylglycoproteins involved in learning and memory formation and to characterize their structure, sub-cellular distribution and the timecourse of their expression. An in vitro increase in the rate of fucosylation was found in the right forebrain base after training. Further, the rate of fucosylation of a protein component of Mr 110-120kD increased significantly within the first four posttraining hours. A number of glycoproteins showing significant increase in fucosylation either 6hr or 24hr after training, or at both posttraining time-points were identified in vivo. Most were concentrated in SPMs and PSDs of LPO as summarised in figure A.1. Polyclonal antibody R15 was raised and it recognized a 60-62kD antigen in SPM and PSD fractions. It rendered chicks amnesic when injected bilaterally 30min before training. Immunocytochemical studies showed medium density labelling of axonal membranes and SPMs and PSDs. Finally, the sugar contents of SPM and PSD membrane-bound glycoproteins were analysed using specific lectins. Chick brain glycoproteins were separated by two-dimensional gel electrophoresis and identified using the lectin Ulex Europaeus type I (UEA.I). Some 55 fucosylated polypeptides were found in SPMs and PSDs. Specific antibodies whose antigens have been implicated in learning and memory formation were used to examine whether their immunoreactivity changed as a result of passive avoidance training. Both anti-N-CAM antibodies and anti-ependymin antibodies cross-reacted with chick antigens from SPM and PSD fractions, but immunoblotting did not show training-induced changes. The results reported in this thesis are discussed in the light of current understanding of neural mechanisms of learning and memory. Several hypotheses are considered which collectively recognize glycoproteins as major contributors to neural network formation/re-formation during learning and formation and consolidation of memory.

This dissertation, written in April 2000, focuses on African American, Hispanic, and White community college students' attitudes toward affirmative action in college admissions. Specifically, the study examines the association between attitudes toward affirmative action in college admissions, student background characteristics, educational aspirations, self-interest, racial affect, and institutional characteristics. The author focuses on finding out whether attitudes toward affirmative action in college admissions varied by race/ethnicity, gender, and future aspirations among students attending community colleges. Some of the key findings of the document are as follows: (1) 74% of African American and 57.5% of Hispanics agreed strongly with racial discrimination still being a problem while 46.4% of White students reported similarly; (2) half of Whites and over two thirds of students of color favored using affirmative action in college admissions; (3) the lower the GPA of White students, the more they favored abolishing affirmative action in college admissions whereas the higher the GPA of African American students they more they favored using affirmative action; and (4) whereas race/ethnicity, gender, annual family income and political views significantly predicted the odds of support or disfavor for affirmative action, separate logistic regression analyses produced different affects by racial/ethnic group. Contains 3 appendices and numerous graphs and charts. (MZ)

Dimension reduction techniques are valuable for both data-rich and data-poor problems. For applications involving massive high-dimensional data, dimension reduction can be utilized for data compression and data-driven discovery. In the data-poor regime, low-rank subspaces enable field reconstruction with only a few sparse measurements. Moreover, reduced-order modeling effectively propagates parametric uncertainty in high-dimensional partial differential equations. This work aims to develop dimension reduction techniques based on spatiotemporal subspaces for applications across the data-availability spectrum as well as performing uncertainty quantification in high-dimensional dynamical systems. First, we develop a low-rank approximation that compresses the size of transient simulation data in real-time, which helps with storage and input/output limitations. These limitations also restrict data analysis and visualization in large-scale simulations. To address this issue, we present an in-situ dimension reduction technique that decomposes the streaming data into a set of time-dependent bases and a core tensor in real-time. This method is adaptive and controls the compression error through the addition or removal of modes. We then develop dimension-reduction methodologies for prediction in data-poor regimes. While it is possible to predict blood flow using machine learning models, clinical measurements, such as Transcranial Doppler ultrasound, may be insufficient or too low-resolution for the training process. Therefore, developing a computational model that provides predictions based on sparse data is crucial. To this end, we present a physics-informed regression framework based on Gaussian process regression to predict blood flow properties using very few sparse measurements. Lastly, we extend the application of low-rank approximation to uncertainty quantification in blood flow simulations. In clinical settings, measurements are often intrusive and inherently uncertain. Numerical simulations could aid in developing non-invasive assessments. However, physiological variability introduces uncertainties in simulation parameters, necessitating a large number of computationally expensive simulations. To address these challenges, we explore implementing a low-rank approximation approach that reduces computational costs while maintaining accuracy.

Since Iran's Islamic revolution in 1978, there has been a notable shift in how the country engages with the global community. This paper explores this transformation within the context of Iran's nuclear program, focusing on the Joint Comprehensive Plan of Action (JCPOA) signed in Vienna in July 2015. Involving major world powers and Iran, the JCPOA aimed to ensure Iran's nuclear program remains peaceful by setting specific limits, removing sanctions, and establishing monitoring mechanisms. This research examines the historical context, negotiation timeline and complexities, operationalization impact, challenges during implementation, and the JCPOA's geopolitical implications. Additionally, it analyzes subsequent developments, including the US withdrawal, ensuing tensions, and diplomatic efforts to salvage the agreement. Employing qualitative research methodology, this study contributes to understanding multilateral approaches to nuclear proliferation challenges and the balance between diplomacy and geopolitical realities.

Polarization domains are the fundamental elements in ferroelectric materials that contribute to their properties. A natural polarization domain, which possesses a specific polarization direction or state, typically ranges in size from one unit cell to a few micrometers. Therefore, studying polarization domains requires the fabrication of materials and the measurement of ferroelectric properties at the nanoscale. In this study, molecular beam epitaxy (MBE) is employed to fabricate barium titanate (BTO) crystals, and piezoresponse force microscopy (PFM) is used to measure the properties of polarization domains at the nanoscale. In the beginning, this study identifies the reflection of high-energy electron diffraction (RHEED) oscillations for the growth of stoichiometric BTO crystals. Then, it demonstrates the MBE growth of (1) single crystal BTO thin films with thicknesses of 2, 10, and 40 nm using the shuttered RHEED method and (2) self-assembled BTO nanodots through a novel two-step method using the Stranski-Krastanov (SK) and droplet epitaxy (DE) techniques. These crystal structures are characterized using X-ray photoemission spectroscopy (XPS) to identify their compositions, atomic force microscopy (AFM) to assess the quality of the film surfaces, and X-ray diffraction (XRD) to determine the lattice parameters and strain and quality of the crystal structures. PFM uses the electric field generated by an AFM tip to manipulate and measure the direction of polarization domains on the surface of ferroelectric crystals. To conduct quantitative studies with PFM, I have established an analytical expression for the electric field beneath the AFM tip. This analytical expression for the electric field is derived and then compared with simulation and experimental results. The AFM tip is modeled as a sphere and a cone, with the sphere being substituted by two image point charges and the cone by a linear charge density that characterizes the electric field. The expressions for the image point charges depend on the thickness and dielectric constant of the film, radius of curvature, and cone half angle of the tip, while the linear charge density is a function of the cone half angle of the tip. The derived electric field expression will be employed to quantitatively investigate the expansion of polarization domains using the electric field under the AFM tip. The dynamic of the expansion of the polarization domains in BTO thin film follows Mrez’s laws. The material constants for BTO thin films in Merz’s law are found and show dependence on strain but not on the thickness and the applied voltage. The effect of temperature on the polarization domain as a function of thickness is also studied. It is found that wall motion and stability of polarization domains are thickness-dependent. This can be attributed to the electric field induced by screening charges in thin films, which is dependent on the thickness. The self-assembled BTO nanodots with a width of 20-40 nm and height of 2-5 nm exhibit a natural downward polarization at their edges, indicating the presence of a skyrmion state for their polarization. This polarization state can be switched by applying an electric field using an AFM tip. This nanodot structure shows great potential as an effective means to enhance the capacity of ferroelectric memory devices.

Abstract: The world's energy consumption and greenhouse gas emissions are on the rise due to increasing energy usage in buildings. To address this problem, it is important to design energy-efficient buildings with advanced control systems. One key aspect of these systems is having reliable thermal dynamic models can provide necessary information for the control systems. Thermal resistor-capacitor (RC) models are one of the commonly used methods for thermal dynamic modeling in control systems. RC models represent the thermal behavior of building components using parameters related to fundamental physical principles. The R's and C's in the model are equivalent thermal resistors and capacitors that represent the effective resistance to heat flow and thermal energy storage capacity of the nodes, respectively. These model parameters are used to relate system inputs (e.g., heating and cooling supply) and temperature states. RC model parameters and inputs (e.g., actual solar heat gain) are difficult to be directly measured but can be estimated with historical data. Accordingly, this thesis aims to develop a practical and dependable method for estimating RC model parameters and input simultaneously with and without partially missing states. By estimating model parameters and creating a trustable thermal dynamic model the developed method will be used to obtain required heating, ventilation, and air conditioning (HVAC) outputs for temperature control purposes.To estimate unknown parameters and inputs, the method uses unscented Kalman filter (UKF) in combination with nonlinear least square (NLS) estimation method. To evaluate the effectiveness of this method, two case studies are conducted. The first case study involves made-up data, while the second case study uses real-world data from a single detached house. The performance of the method is assessed by comparing estimated parameter and input values to true values in the made-up case, as well as the accuracy of the updated thermal dynamic model (created based on the last estimated model parameters) in predicting temperature responses in the real-word case. Both estimation and prediction studies indicate that the developed method can accurately estimate the unknown model inputs and parameters.The method can also be used to estimate the required heating and cooling supply for controlling the temperatures of multiple zones. To evaluate the effectiveness of this approach, two case studies are conducted: one with made-up data and one with real-world data from a single detached house. The performance of the method is assessed by simulating the thermal model with applying the estimated heating and cooling supply to the model and generate the system response for the zone temperature that requires control and verify whether the controlled zone's temperature meets the expected temperature or not. The results of the two case studies indicate that the method can accurately estimate the heating and cooling supply.Overall, the primary objective of this thesis is to develop a practical and dependable method for estimating RC model parameters and inputs simultaneously, even when some states are partially missing. Furthermore, the developed method will be used to obtain the required HVAC outputs for temperature control purposes. Achieving these goals will not only advance the relevant field but also provide feasible solutions for real-world applications.

Transdermal drug delivery is a rapidly expanding industry with significant advantages over conventional oral and intravenous methods. However, in order to gain a better understanding of drug permeation mechanisms and obtain accurate estimates of drug permeability to the skin, there is an urgent need for modeling techniques. In this thesis, we present a multi-scale modeling approach to investigate solute transport through the human stratum corneum. Firstly, continuum finite element brick-and-mortar models were constructed to study water permeation. Subsequently, molecular dynamics simulations were employed to investigate the permeation of various solutes through lipid bilayers within the stratum corneum. [...]

Abstract: Industrial flaring is a notable global contributor to carbon dioxide emissions and other key pollutants, such as black carbon, oxides of nitrogen, and unburnt hydrocarbons. Introducing a separate assisting fluid near the base of these flames affects their hydrodynamics, thermodynamics, and chemistry, which in turn affects their overall efficiency and emissions. The extent of these three effects can differ depending on the injection geometry, the composition of assisting fluid (inert gases, air, steam, or atomized liquid water), and the quantity of the assisting fluid added. This study comprises three connected experimental investigations to address the effects on efficiency, emissions, and stability of lab-scale co-flow jet diffusion flames in either a co-annular burner or a slot burner. In Investigation I, a burner was constructed of two concentric tubes allowed for various generic burner geometries, where different fuels (methane or propane) and co-flow assisting fluids (air, steam, or inert gases) flowed through the annular space and the center tube, respectively. The effects of the composition and flow rate of the fuel and assisting fluid, as well as the burner head geometry, were investigated in terms of carbon conversion efficiency (CCE) and emission indices of black carbon and oxides of nitrogen. Adding low flow rates of the assisting fluid co-flow significantly reduced the black carbon emission, which was not related to the CCE collapse with the main flame blow-off occurring at higher flow rates of assisting fluid. Moreover, any changes that resulted in a higher/lower characteristic flame temperature increased/decreased the emission of oxides of nitrogen. These results showed that, irrespective of the fuel or assisting fluid composition (other than the oxygen-enriched case), there was a range of assisting fluid flow rates, where the CCE was approximately 100 %, while the emissions of black carbon and nitrogen oxides were highly suppressed. To focus on the effects of low flow rates of water as the assisting fluid on emissions, in Investigation II, a burner was constructed of a contoured nozzle with an inner tube at the center. Different fuels (methane, propane, or methane-propane mixture) and co-flow assisting fluids with the same chemical composition (steam or atomized deionized water) flowed through the annular space and the center tube, respectively. The results revealed that water addition suppressed the emission of black carbon and nitrogen oxides more significantly compared to the same mass of steam addition, which is due to the stronger thermal effect of water as the assisting fluid. In order to explore the hydrodynamic effects of assisting fluid on the stability of the diffusion flames, in Investigation III, a multi-slot burner was designed and tested. This burner allowed for stabilizing normal and inverse diffusion flames, as well as the coexistence of both in various configurations with 2D optical access. To finalize the design dimensions and specifications, numerical simulations and non-reacting flow particle image velocimetry (PIV) tests were performed on single slots to get the exit velocity profile along their short and long side. After constructing the multi-slot burner, the coexistence of the normal and inverse diffusion flames was investigated by performing reacting flow PIV tests and overall emission measurements. The notable finding is that the inner air flow rate must be high enough to lift off the inner flame and enable fuel and air premixing to lower the total BC emissions.

X-ray spectroscopy provides a great deal of information for analyzing reactivity and characteristic properties of matter by probing the interactions of molecules with their local chemical environment. The nature of core orbitals and elemental specificity of core-level transitions provide an advantage for resolving spectroscopic signatures with continually advancing experimental X-ray techniques. Computational modeling of X-ray processes can be achieved with ab initio methods for describing the electronic structure of core-excited states. This dissertation explores quantum chemical models for simulating X-ray photoelectron spectroscopy (XPS), X-ray absorption spectroscopy (XAS), and X-ray emission spectroscopy (XES). Specifically, the following research examines the applicability o fΔ-based self-consistent-field (ΔSCF) and Møller-Plesset (MP) perturbation theory (ΔMP) methods and introduces different composite models coupling these techniques with electron propagator theory (EPT) for studying element-specific K-shell transitions. Firstly, as a practical approach, vertical core excitation energies are obtained using a combination of the ΔSCF method and the diagonal second-order self-energy approximation with the inclusion of relativistic effects. For core excitations involving delocalized symmetry orbitals, the applied method improves upon the overestimation of ΔSCF by providing approximate values close to experimental K-shell transition energies. Furthermore, spin projected ΔUHF and ΔUMPn (n=2,2.5,3) methods are also used to calculate vertical core excitation energies. These methods are applied to a set of symmetrical molecules with equivalent atoms. The role of core localization, SCF orbital relaxation, pair correlation, and different relativistic corrections on the accuracy and reliability of the results is examined. Additionally, the limitations of using core-hole reference determinants and complications that may arise in perturbative calculations are addressed. Lastly, a practical ab initio composite method for modeling X-ray absorption and non-resonant X-ray emission is presented. Vertical K-edge excitation and emission energies are obtained from core-electron binding energies calculated with spin-projected ΔHF/ΔMP and outer-core ionization potentials/electron affinities calculated with electron propagator theory. An assessment of the combined methodologies against experiment is performed for a set of small molecules containing second-row elements. Methods for obtaining transition intensities are applied for reconstructing non-resonant X-ray emission spectra. Results from the various ab initio models examined indicate that sub-electron-volt accuracy can be obtained for core-level energetics while maintaining a satisfactory balance between accuracy and computational cost.

Abstract: Approximately 1 in 3 Canadians experiences addiction or mental health challenges at some point in their lifetime. Unfortunately, there are multiple barriers to accessing mental healthcare, including system fragmentation, episodic care, long wait times, and insufficient support for health system navigation. In addition, stigma may further reduce an individual’s likelihood of seeking support. Digital technologies present new and exciting opportunities to bridge significant gaps in mental healthcare service provision, reduce barriers pertaining to stigma, and improve health outcomes for patients and mental health system integration and efficiency. Chatbots (ie, computer programs designed to simulate conversation with human users, especially over the Internet) may be explored to support those in need of information or access to services and present the opportunity to address gaps in traditional, fragmented, or episodic mental health system structures on demand with personalized attention. The recent COVID-19 pandemic has exacerbated even further the need for mental health support among Canadians and called attention to the inefficiencies of the system. As healthcare workers and their families are at an even greater risk of mental illness and psychological distress during the COVID-19 pandemic, this technology is first piloted to support this vulnerable group. In this project, we developed mental health software Mira Chatbot to support healthcare workers and their families in the Canadian provinces of Alberta and Nova Scotia. The software features four major elements: the Mira Chatbot, Mira Resource Portal, Mira Dataset, and Mira Interface. Mira Chatbot’s primary purpose is to provide strategic resources for users responding to the custom needs that they share. Users provide their unique information through two main tasks defined within the Mira Chatbot: intent detection and entity extraction. Intent detection is a process where Mira Chatbot identifies the needs of the user based on a chat experience with an accuracy rate of 99.1%. Through the task of entity extraction, Mira Chatbot recognizes important keywords from a sentence with an accuracy rate of 95.4%.

The current trend of ever-increasing performance in high performance computing (HPC) applications comes with tremendous growth in energy consumption. Because existing libraries are mainly concerned with performance, they do not make efficient use of heterogeneous computing systems, resulting in energy inefficiency. Hence, improving the energy efficiency of critical applications running on HPC systems is necessary to deliver better performance at a given power budget. The aim of this dissertation is to develop techniques and frameworks to improve the energy efficiency of the high-performance applications on heterogeneous system with GPUs while maintaining the reliability and performance requirements. In our first approach, we present GreenMM framework for matrix multiplication, which reduces energy consumption in GPUs through undervolting without sacrificing the performance. The idea is to undervolt the GPU beyond the minimum operating voltage (Vmin) to save maximum energy while keeping the frequency constant. Since such undervoltingmay give rise to faults, we design an Algorithm Based Fault Tolerance (ABFT) algorithm to detect and correct those errors. We target Matrix Multiplication (MM), as a key kernel used in many scientific applications. Empirically, we explore different errors and derive a fault model as a function of undervolting levels and matrix sizes. Then, using the model, we configure the proposed fault tolerant MM algorithm. We show that energy consumption is reduced up to 19.8%. GreenMM also improves the GFLOPS/Watt by 9% with negligible performance overhead.In our second study, we present a framework for GPU applications, which reduces energy consumption in GPUs through Safe Overclocking and Undervolting (SAOU) without sacrificing performance. The idea is to increase the frequency beyond the safe frequency fsafeMax and undervolt below VsafeMin to get maximum energy saving. Since such overclocking and undervolting may give rise to faults, we employ an enhanced checkpoint-recovery technique to cover the possible errors. Empirically, we explore different errors and derive a fault model that is used to determine the appropriate undervolting and overclocking level for maximum energy saving. Similarly, we target MM kernel for error correction using the checkpoint and recovery (CR) technique as an example of scientific applications. In case of MM, SAOU achieves up to 22% energy reduction through undervolting and overclocking without sacrificing the performance. In our third study, we introduce GreenMD, an energy-efficient framework for heterogeneous systems for LU factorization utilizing multi-GPUs. LU factorization is a crucial kernel from the MAGMA library, which is highly optimized. The aim is to apply DVFS by leveraging slacks intelligently on both CPU and multiple GPUs. To predict the slack times, accurate performance models are developed separately for CPUs, GPUs, and PCIe bus based on the algorithmic knowledge and manufacturer’s specifications. We also determine the appropriate level of undervolting for both CPUs and GPUs through offline profiling. Reducing voltage below threshold values may give rise to errors; hence we extract the minimum safe voltages (VsafeMin) for the CPUs and GPUs utilizing a low overhead profiling phase and apply them before execution. It is shown that GreenMD improves the CPU, GPUs, and total energy about 59%, 21%, and 31%, respectively, while deliveringsimilar performance to the state-of-the-art linear algebra library MAGMA. In our fourth study, we introduce a fault tolerant algorithm for LU factorization on heterogeneous systems with GPUs. We have developed a fault tolerant algorithm byconstructing a local and global checksum. The local checksums are used to detect the errors and the global checksums are used to correct the errors. Using the local checksums, we can detect the errors in the middle of the computation which enables us to tolerate more number of faults during the whole execution. LU factorization has three main phases. These phases have different sensitivity to the error. For each phase, we introduce an appropriate level of fault tolerance to prevent the error propagation to other phases. Since, we check the correctness of the computation in each iteration, if there is any error in the system, only the small fraction of the computation is affected and can be covered easily in compared to the previous works such as ABFT.

This study investigated equity-based instruction for Culturally and Linguistically Diverse Students (CLD) by examining public school teachers’ perceptions and preparation on diversity and inclusion. Preparing pre-service teachers to improve opportunities for bilinguals and CLD education is still a critical need (Williams et al., 2019). As public schools implement instructional programs for ESLs, teachers should be well prepared to teach culturally and linguistically diverse students. Teacher preparation is essential for both teaching effectiveness and student academic achievement (Boyd et al., 2009). A quantitative research design was used for this study. A CLD Research-Based Practices Survey on public school teachers’ attitudes and perceptions on the inclusion of CLD students in mainstream classrooms was administered to 200 elementary and secondary teachers in ten public schools who had CLD students in their mainstream classrooms. The study was framed based on the five dimensions of diversity, linguistics needs, cultural needs, awareness, and preparation. The study examined literature on teachers’ perceptions, attitudes, self-awareness, preparation, and CLD resources. The analysis of the survey data uncovered information related to the themes of diversity, instructional awareness and practices, resources, support, and access to linguistically and culturally responsive teaching. The major findings of the study included the role of gender, ethnicity, and ESL background on the effectiveness of teacher preparation and their perceptions towards diversity of the students. The study concluded by providing implications for teachers, schools, educators, and policymakers on providing an equity-based educational environment for all the students including the CLD ones.

Nanobiotechnology often requires a significant amount of nanoparticles with controlled morphology and properties that cannot be achieved using the current state of the arts. In this dissertation, we introduce a scalable approach for the mass production of elongated nanoparticles by harnessing the current distribution during template-assisted electrodeposition technique. This approach not only substantially reduces the synthesis cost and time (by a factor of 4X) but also significantly enhances monodispersity and fabrication yield (by a factor of 80X to 100X). Interestingly, this scalable approach can unlimitedly be scaled up for several magnitudes of orders higher fabrication yields. Practically, high-yielding fabrication methods usually do not allow perfectly identical nanoparticles, leading to variation in properties and functionalities. In this context, we advised a novel, fast, and universal characterization technique, named the projection method, that can potentially be used for characterizing hysteretic behaviors (dependency of a property of a system on its history) of magnetic nanoparticles. The projection method not only speeds up extracting magnetic signatures by a factor of 20X to 100X but also removes the trade-off between accuracy and characterization speed, thus beneficial for both research development and industrial quality control levels. Lastly, extensive characterizations and surface chemistry modifications of numerous magnetic nanowires have been conducted to advance Nanobiotechnology by tackling challenges in quantitative biolabeling and nanobarcoding, cell manipulation, selective detection and stimulation of cancer cells in multimodal therapeutic platforms, cancer hyperthermia, and cryopreservation applications. Specifically, this dissertation addresses long-standing obstacles, including magnetic nanowires agglomeration, surface biofunctionalization, colloidal stability in biological media, and selective detection of biological entities, to boost the progression of Nanobiotechnology. The proven success of magnetic nanowires in the aforementioned applications opens new directions to transform the future of Nanobiotechnology, where my scalable synthesis and universal characterization techniques are the cornerstones.

Affordable, point-of-care diagnostics are needed for disease monitoring in low- resource settings (LRS). Gold-standard laboratory tests require expensive benchtop equipment, which limits their widespread use. In this work, I developed low-cost technologies for the transduction of two independent types of biosensors. In the first part of my work, I developed inexpensive gold leaf electrodes that can be used to build electrochemical biosensors; I use these electrodes to detect clinically relevant viral loads human papillomavirus (HPV) from clinical samples. In the next part of my work, I built an optical setup using off-the-shelf parts for the inexpensive detection of a fluorescent progesterone biosensor; next, I translated this solution-based progesterone biosensor into a paperfluidic format for inexpensive point-of-care testing. The first technology I developed were gold leaf electrodes as a cheaper alternative to conventional gold electrodes. Gold electrodes are commonly used as substrates for biosensors due to the ease of surface modification through the self- assembly of thiolated molecules. However, conventional gold electrodes are often costly and laborious to fabricate, rendering them inappropriate for many applications in low- resource settings. Here, I developed the first integrated, three-electrode biosensor fabricated using pure gold leaf, an inexpensive and widely available material; each device uses only $0.16 worth of gold. Importantly, my fabrication scheme does not require any specialized equipment or clean room space, enabling my fabrication process to be carried out anywhere it is needed. As a proof of concept of this platform, electrodes were modified with DNA for CRISPR-based detection of clinically relevant viral loads (1.2 x 104 total copies) of HPV 18 DNA that had been extracted from clinical samples. Next, I developed a benchtop optical device using inexpensive-off-the shelf parts for fluorescent detection of a transcription-factor based biosensor. This transcription- factor based biosensor detected progesterone as a model target. This optical device worked as well as a multi-thousand dollar plate reader. Next, I adapted the solution- based progesterone biosensor to a paper-based biosensor for the affordable and portable progesterone monitoring. This device had a lower limit of detection of 27 nM, which is a clinically relevant level of progesterone in both cows and premenopausal women.

This project develops different compression techniques, both in algorithm and hardware, dedicated to distributed brain neurochemistry recording devices. In particular, first, two signal processing algorithms called compressive fast-scan cyclic voltammetry (C-FSCV) and compressed principal component regression (C-PCR) are proposed for data rate reduction and hardware-efficient processing of neurochemical signals, respectively. C-FSCV uses compressive sensing (CS) to compress the neurochemical signals recorded with fast-scan cyclic voltammetry (FSCV). It employs block-sparsity concept to efficiently compress the signals up to ~5 times while achieving high fidelity (correlation coefficient, r > 0.97) in signal reconstruction. On the other hand, C-PCR employs the compressed learning (CL) concept to train and test a chemometrics algorithm, based on principal component regression (PCR), with compressed signals. This approach results in ~10 times reduction in the required resources for hardware implementation of the chemometrics algorithm.Next, C-FSCV and C-PCR are implemented in a digital signal processing (DSP) unit, called Compressive processor, fabricated with 0.18μm TSMC CMOS technology, for real-time compression and processing of neurochemical signals. Compressive processor is validated with different pre-recorded datasets from dopamine recording experiments in vivo. Moreover, a neurochemical recording microsystem employing analog background-current subtraction as well as Compressive processor is developed. The analog building blocks of the microsystem are fabricated using 0.18μm TSMC CMOS technology, while the required memories, controlling logics, and DSP unit are implemented in a field-programmable gate array (FPGA) device. The microsystem applies 400V/s FSCV at a carbon-fiber microelectrode (CFM) to sense neurochemicals with a 1st-order delta-sigma modulator, followed by real-time compression and processing of the recorded signals to differentiate neurochemicals from interferents. Measured results from benchtop electrical testing of the microsystem and flow injection analysis (FIA) are presented.

In power/energy systems, free-piston linear machines are referred to as a mechanism where the constrained crank motion is eliminated and replaced with free reciprocating piston motion. Depending on the application, the piston motion can be converted into other types of energy and includes compressed air/fluid, electricity, and high temperature/pressure gas. A research group at West Virginia University developed a free-piston linear engine alternator (LEA) in 1998 and have achieved significant accomplishment in the performance enhancement of the LEAs to date. The present LEA design incorporates flexure springs as energy restoration components and as bearing supports. The advantages of using flexure springs are threefold and include: (1) it increases the LEA’s stiffness and resonant frequency, and hence the power density; (2) it eliminates the need for rotary or linear bearings and lubrication system; and (3) it reduces the overall frictional contact area in the translator assembly which improves the durability. The current research focuses on the design and optimization of the flexure springs as the system’s resonant dominating component for a 1 kW free-piston LEA. First, the flexure springs were characterized according to the LEA’s target outputs and dimensional limitations. The finite element method (FEM) was used to analyze the stress/strain, different modes of deformation, and fatigue life of a range of flexure spring designs under dynamic loadings. Primary geometric design variables included the number of arms, inside and outside diameter, thickness, and arm’s length. To find the near-optimum designs, a machine learning algorithm incorporating the FEM results was used in order to find the sensitivity of the target outputs to the geometrical parameters. From the results, design charts were extracted as a guideline to flexure spring selection for a range of operations. Then, methods were introduced, investigated, and analyzed to improve the overall energy conversion performance and service life of the flexure springs and the overall LEA system. These included: a transient FE tool used for fatigue analysis to quantify the life and factors of safety of the flexure springs as well as the spring’s hysteresis; a fluid/structure interaction model used to quantify the energy loss due to drag force applied on the flexures’ side surfaces; packaging of multiple flexures to increase the overall stiffness and to reduce the vibration-induced stresses on flexure arms due to higher harmonics; a model to investigate the two-way interactions of the flexures’ dynamics with the alternator and engine components to find an optimum selection of the LEA’s assembly; a non-linear friction analysis to identify/quantify the energy losses due to the friction of the sliding surfaces of the flexures and spacers; and a series of static and transient experiment to determine the non-linearity of flexures’ stiffness and comparison to FEM results and for validation of the energy audit results from numerical and analytical calculations. With over 6000 flexure designs evaluated using artificial intelligent methods, the maximum achievable resonant frequency of a single flexure spring for a 1 kW LEA was found to be around 150 Hz. From the FEM results, it was found that under dynamic conditions the stress levels to be as high as twice the maximum stress under static (or very low speed) conditions. Modifications of the arm’s end shape and implementation of a shape factor were found as effective methods to reduce the maximum stress by 20%. The modal analysis showed that the most damaging modes of deformations of a flexure spring were the second to fourth modes, depending on the number of arms and symmetry of the design. Experiment and FEM results showed that using bolted packaging of the springs can damp a portion of the vibration and improve the performance. The drag force loss was found to account for 10-15% of the mechanical losses in a 100 Wnet LEA prototype. From the manufacturing perspective, use of water jet was found the most economical method for manufacturing the flexures which could make the commercial production of the LEAs feasible; however, for high-efficiency, high-durability machines, additional material treatments, and alternative manufacturing methods are essential.

This dissertation is divided into two parts. The first part is concerned with developing algorithms for the compression of emerging 3D audio format, while the second part investigates optimization techniques for error-resilient predictive compression systems design.In the first part, advances in development of compression algorithms for higher order ambisonics (HOA) data is presented. HOA has proven to be the method of choice in virtual reality applications, given its capability in reproducing spatial audio and its rendering flexibility. Recent standardization for HOA compression adopted a framework wherein HOA data are decomposed into principal components that are then encoded by standard audio coding, i.e., frequency domain quantization and entropy coding to exploit psychoacoustic redundancy. A noted shortcoming of this approach is the occasional mismatch in principal components across blocks, and the resulting suboptimal transitions in the data fed to the audio coder. In this dissertation, we propose a framework where singular value decomposition (SVD) is performed after transformation to the frequency domain via the modified discrete cosine transform (MDCT). This framework not only ensures smooth transition across blocks, but also enables frequency dependent SVD for better energy compaction. Moreover, we introduce a novel noise substitution technique to compensate for suppressed ambient energy in discarded higher order ambisonics channels, which significantly enhances the perceptual quality of the reconstructed HOA signal. In the next step, to reduce the burden of side information, a new encoding architecture is presented, where transform matrices are estimated backward-adaptively. This framework allows a more frequent usage of optimal SVD, thereby approaching the full potential of frequencydomain SVD. Also the division of HOA data into predominant and ambient components in current schemes, is difficult to perceptually optimize and ignores spatial inter channel masking effects. To address this issues, a new encoding framework for compression of HOA data is presented, where a null-space basis vector extension technique enables all compression to be performed in the SVD domain, and a jointly computed common masking threshold accounts for effects of spatial masking across components.The second part is concerned with developing optimization techniques for error-resilient predictive compression systems design. Prediction is used in virtually all compression systems and when such a compressed signal is transmitted over unreliable networks, packet losses can lead to significant error propagation through the prediction loop. Despite this, the conventional design technique completely ignores the effect of packet losses, and estimates the prediction parameters to minimize the mean squared prediction error, and optimizes the quantizer to minimize the reconstruction error at the encoder. While some design techniques have been proposed toaccurately estimate and minimize the end-to-end distortion (EED) at the decoderthat accounts for packet losses, they operate in a closed-loop, which introduces a mismatch between statistics used for design and statistics used in operation, causing a negative impact on convergenceand stability of the design procedure. The first contribution of the dissertation is this part is proposing an effective technique for designing a compression system with a first order linear predictor, that accounts for the instability caused by error propagation due to packet losses, and enjoys stable statistics during design by employing open-loop iterations that on convergence mimic closed loop operation.End-to-end distortion (EED) estimation, accounting for error propagationand concealment at the decoder, has been originally developed for video coding, and enables optimal rate-distortion (RD) decisions at the encoder. However, this approach was limited to the video coder’ssimple setting of a single tap constant coefficient temporal predictor. This thesis considerably generalized the framework to account for: i) high order prediction filters, and ii) filter adaptation to localsignal statistics. We demonstrate how this EED estimatecan be leveraged, by an encoder with short and long term linearprediction, to improve RD decisions and achieve major performance gains. The approach is further extended to estimate EED in speech coders. The error propagation problem is exacerbated in this case, as standard coders not only predict the signal from past frames, but also the parameters (in the line spectral frequency domain) employed for such prediction. Hence, the prediction loop propagates errors in the reconstructed signal as well as errors in the prediction parameters. A recursive algorithm is proposed to estimate, at the encoder, the overall EED, by the subterfuge of parallel tracking of decoder statistics for prediction parameters and signal reconstructions, in their respective domains, which are then combined to obtain the ultimate EED estimate.

Understanding fatigue is important to the success of dental materials as evident from the eventual failure of restorations and cyclic loads during mastication. However, detection of such failures often poses clinical challenges as most cracks are invisible or subsurface when initiated and only identified when catastrophic failures occur. Current detection methods are limited in clinical application. Therefore, the aim of this research study was to design a test method and parameters that simulate the oral environment to assess the detection of crack initiation and growth using quantitative percussion diagnostics (QPD) technology. Nine 3D tooth models were subjected to cyclic fatigue loading amplitudes of 92 N accompanied by periodic QPD testing. Normal fit error values determined by QPD before loading and after fatigue cycling testing were recorded for the 3D simulated teeth. Results showed that location and depth of notch were important determinants of the number of cycles to crack initiation. The QPD device was able to monitor crack initiation and growth in at least three of the simulated teeth with accuracy. Although each tooth was found to reach failure at a different cycle number, the paired comparison was non-significant (p= 0.099) indicating results are comparable. Future analysis requires more simulated teeth to confirm these findings. Also, a comparison of the data obtained from 3D tooth models to natural extracted teeth as well as comparison to existing diagnostic tools used for crack identification would be warranted. These preliminary findings provide further evidence for the use of QPD as a diagnostic aid for early crack identification and monitoring structural integrity of teeth.

Abstract: There is no shortage of community mining algorithms for discovering structure in complex information networks; most with unique advantages, however, all with drawbacks, including efficiency, correctness, resolution limit, and field of view limit. We introduce a novel efficient approach for community detection based on the notion of edge strength and a formal definition of the notion of community. We consider that there are two types of links in a graph: links that run between communities whose removal may divide the graph into disconnected subgraphs that we refer to as weak links, and links that are inside communities whose removal would not change graph connectivity, that we refer to as strong links. We put forward a new objective function, called SIWO (for Strong In, Weak Out), which encourages adding strong links to the communities while avoiding weak links. Optimizing this function allows us to discover dense subgraphs from which qualified communities that comply with the definition are extracted. This process allows us to effectively discover communities in social networks without the resolution and field of view limit problems some popular approaches, considered the state of the art, suffer from. Moreover, time complexity of our method, which piggybacks on the optimization of Louvain, known to be very efficient, is linear in the number of edges. We experimentally demonstrate the effectiveness of our approach on various real and artificial datasets with large and small communities to show the resilience to the resolution and field of view limits, including large size networks having million edges.

Inter-organizational knowledge exchange is an important process for stimulating innovation and improving collaboration among multiple organizations. With the use of information and communication technologies (ICT), this process could be greatly affected for effectiveness and efficiency. However, existing studies provide an insufficient understanding of how knowledge exchange processes are shaped by the nature of ICT in the context of inter-organisational settings. Therefore, this research attempts to address this research gap through conceptualising the role of ICT in the shaping of inter-organisational knowledge exchange (IOKE) practice by investigating the context of European Living Labs, as an exemplary inter-organisational collaboration setting. In methodological terms, this study adopts an inductive, qualitative research approach, and follows a combination of Grounded Theory, Multiple Case Study and Documentary Research, making use of in-depth semi-structured interviews with European Living Labs stakeholders and analysing Living Labs project documents. The data has informed the use of Practice Theory, Sociomateriality Theory, and Technology Affordance Theory at different levels of analysis to understand the research phenomena. This research has found that 1) ICT-based IOKE practice can be characterised by ICT affordances and constraints, human knowledgeabilities, and triple-layer contexts (inter-organisational context, intra-organisational context, and wider environmental context). 2) ICT mediates such enactment of knowledgeability by affording or constraining distinct human abilities that allow actors to accomplish their knowledge work or hold them back as they engage in knowledge exchange practice. Diverse ICT affordances (e.g., cross-distance networking, instant workaround) and constraints (e.g., accessing, communication continuity) can mediate the enactment of four different types of knowledgeabilities: inter-connecting, interactive learning, co-creating, and co-ordinating. 3) Different knowledgeabilities, their associated ICT affordances/constraints, and their embedded contexts are inter-related, and they co-evolve over time across the project lifecycle. This has been conceptualised into an ICT-based IOKE practice framework (integrated) and its three variations (three variated frameworks for different project lifecycle stages). The findings have strong theoretical and practical implications that signal the salient areas for future research to study, particularly for ICT affordance studies.

Study and prediction of classical and non-classical mechanical properties of GaN is crucial due to the potential application of GaN nanowires (NWs) in piezoelectric, probe-based nanometrology, and nanolithography areas. GaN is mainly grown on sapphire substrates whose lattice constant and thermal expansion coefficient are significantly different from GaN. These discrepancies cause mechanical defects and high residual stresses and strains in GaN, which reduce its quantum efficiency. Specifically, for nanoscale applications, the mechanical properties of materials differ significantly compared to the bulk properties due to size-effects. Therefore, it is essential to investigate the mechanical properties of GaN NWs using the non-classical solid mechanic theories, modified couple stress theory (MCST) and modified strain gradient theory (MSGT). Experimentally the GaN NWs were prepared by using top-down approach to etch c-plane GaN layer grown in Metal Organic Chemical Vapor Deposition (MOCVD) chamber to achieve high aspect ratio NWs with high uniformity. An Atomic Force Microscope (AFM) was used to apply an infinitesimal deflection on the top of clamped-free NWs while monitoring the lateral and normal forces. According to the MCST, the Young’s modulus, shear modulus and length scale were measured to be 323 GPa, 133 GPa and 13 nm, respectively, and according to the MSGT, they were measured to be 319 GPa, 132 GPa and 8 nm, respectively. Furthermore, a quantum mechanics based approached was conducted to estimate the classical and non-classical mechanical properties of the GaN NWs as well.

Pore Network Modelling (PNM) is a well-established micro-scale numerical method for the prediction of permeability in reservoir rocks. We study the capabilities of this method for coal as a heterogeneous unconventional rock. The fracture structure of coal known as cleats is captured using high resolution micro-Computed Tomography (micro-CT) images at the core-scale. A network model is extracted to idealise the fractured geometry and its interconnectivity. This three-dimensional arrangement of pore nodes and throats replaces the entire complex geometry of coal. Flow simulation is then performed on the extracted network model. The absolute permeability is compared with direct computations on micro-CT images by the Lattice Boltzmann Method (LBM). Obtained relative permeability is also compared against Capillary Drainage Transform (CDT) and microfluidic framework as direct simulation and experimental methods, respectively. Comparison of PNM with LBM, CDT and microfluidic results shows applicability of PNM as a predictive method for permeability in anisotropic geometries such as coal. However, the accuracy of calculations is tied to the connectivity of the fractures in the flooding direction. The effects of flow direction, wettability and pore network geometrical statistics such as coordination number, pore size distribution and shape factor on absolute and relative permeabilities are investigated. Then, a numerical simulation model is developed using a commercial reservoir simulator to estimate the average relative permeability at the core-scale. The effect of different wettability conditions on gas production from coal is investigated by moving the cross point in the obtained gas-water relative permeability. Our results quantified the effects of aperture size, wettability and porosity of coal on its permeability. However, to improve the accuracy of the results, it is recommended to combine PNM with stochastic methods to simulate flow on more representative samples in terms of size. In addition, effect of micropores is ignored in this work. Dual networks are recommended to be developed to take them into account.

We live in an era when political and commercial entities are increasingly engaging in sophisticated cyber attacks to damage, disrupt, or censor information content and to conduct mass surveillance. By compiling various patterns from user data over time, untrusted parties could create an intimate picture of sensitive personal information such as political and religious beliefs, health status, and so forth. In this dissertation, we study scalable and robust distributed algorithms that guarantee user privacy when communicating with other parties to either solely exchange information or participate in multi-party computations. We consider scalability and robustness requirements in three privacy-preserving areas: secure multi-party computation (MPC), anonymous broadcast, and blocking-resistant Tor bridge distribution. We propose decentralized algorithms for MPC that, unlike most previous work, scale well with the number of parties and tolerate malicious faults from a large fraction of the parties. Our algorithms do not require any trusted party and are fully load-balanced. Anonymity is an essential tool for achieving privacy; it enables individuals to communicate with each other without being identified as the sender or the receiver of the information being exchanged. We show that our MPC algorithms can be effectively used to design a scalable anonymous broadcast protocol. We do this by developing a multi-party shuffling protocol that can efficiently anonymize a sequence of messages in the presence of many faulty nodes. Our final approach for preserving user privacy in cyberspace is to improve Tor; the most popular anonymity network in the Internet. A current challenge with Tor is that colluding corrupt users inside a censorship territory can completely block user's access to Tor by obtaining information about a large fraction of Tor bridges; a type of relay nodes used as the Tor's primary mechanism for blocking-resistance. We describe a randomized bridge distribution algorithm, where all honest users are guaranteed to connect to Tor in the presence of an adversary corrupting an unknown number of users. Our simulations suggest that, with minimal resource costs, our algorithm can guarantee Tor access for all honest users after a small (logarithmic) number of rounds.

These days, sensor chips in very small form factors are ubiquitous. In this dissertation, I explain my efforts and achievements in proposing the first center-clamped 3C-SiC RF microdisk resonators with out-of-plane vibration as a candidate for future resonant sensing applications. First, nanomachining of 3C-SiC material using Focused Ion Beam (FIB) to find sputtering yield and lateral deformations for many different FIB conditions as well different patterns (and scales), is investigated accompanied by SRIM simulations.Then, I report theoretical analysis, fabrication and measurement of (center-clamped) RF multi-mode micromechanical resonators based upon vibrating circular disks made of a ~500nm thin SiC epilayer grown on single crystal Si. The fabrication method - composed of FIB sputtering and HNA etching - has been used for the first time and the vibration resonance peaks (in frequency spectrum) are detected through laser-interferometry measurements. A ~40µm-diameter SiC disk with a slender (~800nm) anchor exhibits more than a dozen flexural modes between ~2MHz¿20MHz with quality factors (Q’s) of ~1000¿4000. Two other disks with diameters of ~40µm and ~30µm, and wide anchors (~20µm and ~10.3µm, respectively) have their set of major flexural peaks (associated with modes of zero-circular nodes) between 15¿19MHz with Q’s of ~500¿2500. This dissertation then describes the efforts to achieve an improved self-sustained oscillator based on such resonators. First, a theory-inspired method is developed for design of electrostatically-transduced MEMS/NEMS referenced Pierce oscillators to improve phase noise (by increasing motional inductance and thus motional resistance) while also addressing oscillation start-up and other specifications e.g. frequency detuning and power consumption. Finally, the laser-based optical transduction scheme is used to make self-sustained oscillators from microdisk resonators. Laser actuation is very compatible with 3C-SiC material and the targeted application of sensing in harsh-environments and very immune to the undesired circuit loading effects and power handling inefficiencies of other transductions. The oscillator is analyzed by behavioral simulations and also measured with a home-built setup. The transduction with the microdisk resonator is via interferometry-based sensing and laser-based optical actuation. The measured phase noise of such oscillator shows a noticeable improvement of the effective Q (~2.3) over that of the open-loop microdisk resonators.

The primary goal of orthodontic treatment is a functional dental occlusion as well as an aesthetic smile in harmony with the lower 1/3 facial soft tissue. A 3-dimensional analysis of the patient’s dentofacial relations is crucial for proper treatment planning and it is ultimately the soft tissue that dictates the direction of the orthodontic treatment plan. The purpose of this pilot study is to study the difference of three different medias (3D photography, 2D photography, clinical patient evaluation) for facial form analysis. Specifically, to determine which photographic method most closely mimics clinical analysis of facial form for the purpose of orthodontic facial form diagnosis. Two orthodontists analyzed twelve facial measurements on 54 patients (21 male) via 3D photographs, 2D photographs and clinical evaluations. In addition, twelve of these patients were randomly chosen for two extra evaluation sessions for intra-rater reliability testing. The twelve categories are the most commonly used measurements by orthodontists to evaluate lower 1/3 facial form for diagnosis and treatment planning purposes. Krippendorff’s Alpha reliability testing was used with values >0.6 deemed substantially reliable. Results showed poor intra-rater reliability for both evaluators across all three media (only 22 out of 72 total categories substantially reliable) demonstrating the examiners were not reliably consistent with their evaluations. Furthermore, results showed poor reliability testing between the three different medias making comparisons of 3D vs. 2D vs. clinical evaluation difficult. Lastly, results showed poor inter-rater reliability for the two evaluators across all three media (zero substantially reliable categories) demonstrating that the two examiners did not agree on any of the facial evaluations. In conclusion, the two orthodontist did not reliably diagnose most of the 12 facial categories using the three given media of clinical, 2D photography and 3D photography and they also did not agree with each other’s diagnosis; therefore, it is difficult to determine if 2D or 3D photography provides more diagnosis similarities to clinical evaluation of facial form diagnosis.

Nickel barrier layers are commonly-used in metallisation of silicon-based microelectronic and photovoltaic devices to prevent copper diffusion from metal contacts to the underlying silicon. This thesis compares the effectiveness of nickel barrier layers, formed using: (i) electroless plating; and (ii) light-induced plating (LIP), in preventing copper diffusion into mono-crystalline silicon laser-doped selective emitter (LDSE) solar cells. Energy dispersive X-ray spectroscopy was used in conjunction with transmission electron microscopy for chemical identification of copper-rich regions in silicon under the nickel/copper metal contacts of heat-treated cells. It was observed that copper can penetrate through both electroless-plated and LIP Ni barrier layers in LDSE cells that have been heated for 5 hours at 200 °C or 400 °C, followed by quenching in ethylene glycol (fast cooling). Current-voltage measurements were used to characterise the electrical performance of LDSE cells with different nickel barrier layers. Significant degradations in cell efficiency, fill factor and open-circuit voltage were observed for LDSE cells with electroless-plated nickel barrier layers after heating at 200 °C and subsequent quenching in ethylene glycol. In the worst case, the fill factor was reduced from 66% to 31%, causing severe efficiency degradation. However, the fill factor of cells with LIP nickel barrier layers were only reduced from 66% to 58% under the same conditions. When cells were cooled slowly the electrical performance of the cells was only slightly affected and was independent of the type of nickel barrier layer. Furthermore, LDSE cells with electroless-plated nickel barrier layers that were originally heat-treated followed by fast cooling, were exposed to a second heat treatment at 250 °C for 2 hours, followed by slow cooling. This resulted in a significant improvement in the lifetime. These results suggest that copper may diffuse through both electroless plated and LIP nickel barrier layers at temperatures as low as 200 °C. However, on slow cooling (which is likely in the field), the copper can out-diffuse to regions of the cell where the cell’s electrical performance is not degraded. These observations suggest that some commonly-accepted views on the ability of nickel barrier layers to prevent copper diffusion in silicon solar cells may need to be challenged and further investigated.

Cyber-Physical Systems (CPSs) are complex systems resulting from intricate interaction of digital computational devices with the physical systems. With the recent dazzling advances in computational devices, CPSs have become ubiquitous in modern technology. The increasing presence of CPSs on one hand and the incapability of current methods to analyze them on the other hand, impel the development of novel approaches for analysis and design. In CPSs, embedded computers have the responsibility of monitoring and controlling the physical plants using feedback loops using which physical plants affect computations and vice versa. In these closed-loop fashions, controllers implemented in software are termed embedded control software. Increasing use of embedded control software in life critical applications, such as aircraft flight control systems and automotive engine control systems, demands lots of efforts on software verifications and validations which are very costly. On the other hand, by changing the center of gravity from verification to design, it is possible to synthesize correct-by-design embedded control software while providing formal guarantees of correctness. The foundation of this proposed approach relies on some technical results showing how to construct equivalent finite state models for differential equation models describing physical plant. These finite state models are simpler descriptions of physical plant in whicheach state of the finite model represents a collection or aggregate of states in the physical plant. Similar finite state models are used in software and hardware modeling, which enable the composition of such models with the finite models of the physical systems. The results of this composition are finite models capturing the behavior of the physical systems interacting with the digital computation devices. Once such models areavailable, the methodologies and tools developed in computer science for verification and control synthesis purposes can be easily employed to physical systems, via these models. In the first part of this thesis I take an important step in my quest to synthesize correct-by-design embedded control software for CPSs by constructing finite state models for control systems. I propose a novel technique to compute bisimilar finite state models of incrementally stable nonlinear control systems. I show on practical examples that the finite state models computed by my procedure can be several orders of magnitude smaller than existing approaches. Moreover, I propose another technique to compute (not necessarily bisimilar) finite state models of any nonlinear control system as long as I am interested in its behavior in a compact set. In the second part of this thesis I will show some incremental properties under which nonlinear control systems admit finite state models. I propose some analysis tools to check those properties. Moreover, I provide some design techniques providing controllers enforcing those incremental properties for some special classes of nonlinear control systems.