Environmental and Occupational Health Sciences Institute
Rutgers University Community RepositoryNot a member yet
58345 research outputs found
Sort by
The spectral urban: visualizing the Indian metropolis
No full textThis dissertation considers literary fiction, graphic narratives, and films from Delhi, Kolkata, and Mumbai, tracing the ways in which their contested histories have found expression in equally conflicted media forms. This multilingual interdisciplinary project attends to the local and global processes that have impacted and transformed the three cities since the neoliberalization of the Indian economy in 1991, which deepened the nation’s desire to create world class cities that fit into a fantasy futurism, and analyzes the ways in which the economic transformation of the national landscape transformed its cultural landscape. Using Indian fiction, film, and graphic narratives as its primary archive, and drawing upon the idea of spectrality to read urban absent-presences that do not make it into the traditional cartographic representations of cities, this dissertation recognizes and preserves images of urban spaces now on the brink of extinction or radical transformation. Through a multimodal reading of the legacies of the past that shape contemporary representation of Indian cities, I propose that the city, continually shape-shifting and mercurial, is marked by its lost or forgotten avatars, which are memorialized through the textual/visual dimension of representations. Since recognizing the lost or forgotten specters are critical to understanding and constituting the contemporary urban, reading the spectralized layer of absent-presences of the marginalized past and unacknowledged present allows for a visualizing of the urban beyond the known and mapped cities.Ph.D.Includes bibliographical reference
Essays on methods for forecasting using high dimension data
No full textWith the advent of big data and machine learning methods (ML), economists can utilize more information from high dimensional and high frequency dataset for financial analysis and economic forecasting. This dissertation consists of two essays that explore issues in empirical macroeconomic forecasting using machine learning methods to construct a variety of latent factors and economic uncertainty indexes from high dimensional dataset. In the first essay, we construct several new business conditions predictors and economic uncertainty indexes for use in improving the forecast accuracy of macroeconomic variables. In the second essay, we introduce a novel variable selection procedure that correctly distinguishes between relevant and irrelevant variables to be used when estimating latent factors in a factor augmented vector autoregressive framework.
In Chapter 2, we evaluate the marginal predictive content of a variety of new business conditions (BC) predictors as well as nine economic uncertainty indexes (EUIs) constructed using these predictors. Our predictors are defined as selected observable variables and latent factors extracted from a high dimensional macroeconomic dataset and our EUIs are functions of predictive errors from models that incorporate these predictors. Estimation of the predictors is based on a number of extant and novel machine learning methods that combine dimension reduction and shrinkage. When predicting 14 monthly U.S. economic series selected from 8 different groups of economic variables, the new indexes and predictors are shown to result in significant improvements in forecast accuracy, relative to predictions made using benchmark models. Moreover, while inclusion of either BC predictors or EUIs often yields forecast accuracy improvements, greater predictive gains accrue when using BC predictors with real economic activity type variables. Also, adding both BC predictors and EUIs together is particularly useful when forecasting housing market variables at short horizons.
In Chapter 3, we propose a new variable selection method that allows researchers to distinguish between variables that are relevant in the sense that they provide useful information for estimating underlying latent factors and variables that are irrelevant in the sense that they do not load on underlying factors, in an FAVAR model. In our context, variable selection methods are needed because using too many irrelevant variables may lead to inconsistency in factor estimation. Our procedure is designed to facilitate consistent factor estimation and can be viewed as the factor model analog of the type of multiple hypothesis testing or variable selection procedures that people use to select regressors when specifying linear regression. One key difference between our method and the typical multiple hypothesis testing procedure used in the literature is that rather than controlling the overall Type I error at some fixed non-zero level, our procedure is completely consistent in the sense that the probability of both Type I and Type II errors goes to zero asymptotically as sample sizes approach infinity. Monte Carlo evidence indicates that our method has very good finite sample properties. Additionally, we analyze a real-time macroeconomic dataset, where it is shown that our method delivers factors that result in improved marginal predictive content, relative to cases where standard principal components as well as hard-thresholding methods are used in factor estimation.Ph.D.Includes bibliographical reference
Design, 3D printing, and characterization of porous orthopedic fixation devices
No full textThis thesis begins with a review on the use of Laser-based Powder Bed Fusion (LPBF) to print metallic lattices for use in orthopedic applications. The benefits of such implants, including promoting osteointegration and limiting stress shielding, have been reviewed. A discussion of the specific implant geometries and materials utilized to achieve these benefits follows. Firstly, three lattice designs are optimized using the software package nTop for use in porous blanks. Theoretical porosity, mean pore size, min pore size, max pore size, and overall pore distribution data are calculated and analyzed. Secondly, porous blanks are printed using an LPBF machine, specifically the Aconity3D Mini. The blanks are then machined to create M7 threads, a thread analogous to those used in headless compression screws. Thirdly, a 2D characterization was performed using a Keyence VR-3100. A 3D characterization was performed using a Nikon 225KV industrial CT. The data gathered are compared to the theoretical nTop data. Surface and line roughness measurements are obtained from the as-printed and as-machined surfaces using a Keyence VR-3100. Finally, theoretical stiffness tensors are calculated for each lattice design and is compared to that of a theoretical solid sample.M.S.Includes bibliographical reference
Increasing readability to online inherited retinal disease information
No full textInherited retinal diseases (IRDs) affect more than five million people in the world and many of them have overlapping features. An increasing number of individuals are using the internet for medical information; however, many sources are written at a higher literacy level than the average individual in the United States. Despite this, almost seventy-five percent of Americans use online health information to influence their medical decisions. Many online webpages for IRDs are written at a higher reading level and lack information about the value of genetic testing, information on low vision resources, and possible therapies. The purpose of this project was to create a patient-friendly webpage for both the general public and healthcare professionals. A draft of the webpage was distributed and evaluated via a survey and readability scoring. Our findings show that although readability scales may indicate information is at a higher than recommended reading level, readers are able to understand the information provided. We also found that readers are interested in having more information about genetic counseling. Our hope is that this free, public access webpage will help to make online health information of IRDs more accessible.M.S.Includes bibliographical reference
Natural biomaterials aid pressure wound healing following spinal cord injury
No full textPressure wounds (PWs) are among the secondary complications after spinal cord injury (SCI) that hinder an individual’s social, psychological, and physical well-being. Studies have shown that PWs below the SCI site are especially prone to heal slowly or not heal at all, further spotlighting the severity of these wounds. While many preventive measures and therapeutic options are available for those facing PWs, they have high recurrence rates and can prove to be costly to treat for many patients. A low-cost, easy-to-use treatment modality that promotes wound healing progression under such suboptimal circumstances is necessary. We proposed the utilization of natural biomaterials applied to PWs to accelerate the progression of wound healing stages. The claims were investigated in experimentally induced excisional wounds on male and female mice with complete transection of the spinal cord. Skin wounds on male SCI were 42% smaller than similar untreated wounds (used as controls) on post-wounding day 20. Histology analysis revealed significant differences in wound width (p<0.05), epidermis and dermis thickness (p<0.05), macrophage presence (p<0.01), blood vessel formation (p<0.01), cell proliferation (p<0.05), and extracellular matrix (ECM) deposition (p<0.01) when the biomaterials were used as a treatment. When the biomaterials were applied topically on female SCI mouse wounds, the same wound closure trends were observed, and in addition, overall healing occurred about six days earlier compared to male mouse wounds. In contrast, the same biomaterials had no effect on healing of similar skin wounds in non-SCI mice. Ultimately, our findings demonstrate that these natural biomaterials significantly improve skin wound healing mechanisms and present a promising treatment option for promoting PW healing after complete SCI.M.S.Includes bibliographical reference
Improvement of road and runway safety in wet weather with friction modeling and speed limit analysis
No full textThe skid resistance on the road and runway surface plays a critical role in the overall functionality and safety by influencing the interaction between tires and pavement surface. The challenge of predicting skid resistance requires addressing the complex interplay of factors affecting tire-pavement interaction. Moreover, the skid resistance is the fundamental brick to evaluate the braking performance for determination of speed limit. The main objective of this research is to develop the validated numerical models and integrated analysis procedures to improve the road and runway safety based on the predicted skid resistance on the wet pavement surface.
Firstly, this study developed a finite element model using Coupled Eulerian-Lagrangian algorithm for both passenger car and aircraft tire to evaluate the skid resistance on the wet pavement surface considering various water film depths and tire operational scenarios. Secondly, the predicted wet skid resistance considering the adverse wet weather effect was substituted into the computation of the braking distance for both passenger car and aircraft tires. Third, the stopping sight distance of passenger car on straight sections were further evaluated with driver visibility for analysis of speed limit and car following distance. Finally, the of braking performance were evaluated using an integrated tire-vehicle model to analyze road safety on curved sections.
This study concluded that the current roadway design criteria based on the empirical estimation is not sufficient to guide the roadway design and guarantee the road safety during wet weather. The determination of the speed limits considering visibility and safe braking speed at both straight and curved sections can help the transportation agencies for better roadway design and safety management. The evaluation of the aircraft braking distance during landing operation considering both new and worn tires, and the recommendations for proper approaching speeds can help airport agencies and pilots to mitigate the overrun risk of aircraft landing from different aspects.Ph.D.Includes bibliographical reference
Sensory fatigue: militarization, redevelopment and state exceptionalism in Kingston Jamaica
No full textThis dissertation looks at power, domination, and resistance amongst the urban underclass in Kingston, Jamaica, within the context of postcolonial struggles of state legitimacy and economic competitiveness through increased practices of militarization and exceptional forms of policing. Using data retrieved from archival sources and ethnographic methods, this project investigated how, at the macro level, punitive security policies, specifically reasserts state power as a path for economic development for the most marginalized while simultaneously drawing existing contradictions of Jamaican urban political life into the formal space of urban governance. Given that the military is being used as security infrastructure to regulate urban life, at the micro-level, this project examined the effects of this militarization on residents. I lived in Trench Town for twelve consecutive months, observing residents’ everyday movements in highly constrained conditions, focusing on how and where state control manifested, and how households and families adapted and reshaped their urban environments. Using various participatory methods as well as knowledge from my own lived experience in Trench Town, I was able to understand and appreciate the complexities of their everyday urban experiences, increasingly shaped by conditions of extreme violence, militarization, and state surveillance. My research adopts a multidisciplinary approach that draws from theoretical and methodological perspectives from various fields, including environmental justice, critical development studies, postcolonial and urban geography, Black geographical scholarship, and Caribbean studies. Drawing critical inspiration from the histories of social and environmental struggle in multiple compromised environments, attending to practices of Black space making, and experimenting with decolonial qualitative methodologies, this research project challenges dominant ways of knowing by centering local knowledge and experiences. The central novel idea posited in this dissertation is the idea of sensory fatigue, which is the situational cognitive-interpretative as well as affective response to the layered excesses of militarization, poverty, and gang violence.Ph.D.Includes bibliographical reference
Protein on display: a comprehensive analysis of front-of-package labels and consumer perceptions
No full textConsumers are searching for food products that will improve their health and are convenient and tasty. To meet this demand, the popularity and sales of protein foods are growing, along with consumer perceptions that they will lead to greater satiety, energy, and weight loss. Marketers have responded by increasingly placing declarations of protein on the front-of-package of food products. These declarations communicate salient product attributes to consumers yet may also lead to misinterpretations of the overall nutritional composition of these products, which may not be substantially better than similar products without protein claims. Many contain added sugars, sodium, and fats to make them more palatable. Since many American consumers choose foods based on their protein content, it is important to determine which types of products most frequently contain protein claims, as well as the types of protein claims that are being presented. Additionally, examining how consumers perceive these protein-labeled products in the context of health will shed light on their implications for individual health and societal rates of obesity. Few academic studies that examine protein labels beyond the traditional snack bar format exist in the literature. Thus, our research aims to fill the knowledge gap surrounding an important and timely topic. The objectives of this research were to characterize the types of products that make protein claims, identify how these protein claims are displayed, and determine how these protein-labeled products compare nutritionally to other similar products without protein claims. Drawing on the results of this analysis, two online surveys addressed the objective of evaluating how consumer perceive these protein-labeled products in the context of healthfulness and purchase intentions, as well as examining how protein claims affect attention to other nutrition information on the package.
Products with front-of-package claims of protein were analyzed using Label Insight, an online product database of label information and product images for over 330,000 foods. Product categories with the greatest proportion of products with protein claims were determined to be Dairy, Frozen Food, and Cereal & Breakfast Foods. These categories are comprised of 15.33%, 14.32%, and 13.83% of products with protein claims, respectively. Overwhelmingly, protein amount declarations (e.g., “10g of protein per serving”) were the most frequently made front-of-package protein claims across all products. Most products made only one front-of-package protein claim, but also contained at least one other claim, such as “Organic” or “No preservatives” in addition to the protein claim.
Mann-Whitney U tests were used to determine whether there were differences in the nutrient content per 100g between three product types (yogurt, cereal, and frozen entrees) with and without protein claims. Yogurt, cereals, and frozen entrees were chosen for this analysis due to their consumer popularity and proportion of protein claim products relative to non-protein claim products within each category of Dairy, Frozen Food, and Cereal & Breakfast Foods. Yogurts with protein claims had statistically significantly more protein, and less calories, sodium, and sugar compared to yogurts without protein claims. Cereals had significantly more protein and fat, and less sodium and sugar compared to cereals without protein claims. Nutrients in frozen entrees with protein claims did not significantly differ from frozen entrees without protein claims.
To determine if consumers perceived these differences in protein-labeled products, two online surveys were conducted. The first, consisting of 1,022 participants, presented images of the front packaging of two familiar American cereals (Special K Original and Special K Protein) and asked comparative questions about product healthfulness, health benefits, and nutrients. Participants perceived Special K Protein as healthier, more nutritious, though less tasty than Special K Original. Special K Protein was perceived as providing greater health benefits, such as being more likely to help build muscle, stay healthy, and live longer. Many participants perceived no differences in the amounts of certain nutrients between the cereals, such as sugar (54.5%), sodium (59.2%), and calories (49.1%). Yet, when serving sizes are equalized to one cup, Special K protein has more sugar, sodium, and calories than Special K Original. Though most participants reported viewing the Facts Up Front label that was also present on the package displaying the serving size amounts, only 21.3% correctly chose Special K Original as having the larger serving size.
The second consumer survey randomly assigned 1,195 participants to experimental conditions to test perceptions of differing types of front-of-package protein labeling schemes on cereal and yogurt products: “Good Source of Protein,” “Excellent Source of Protein,” “protein” in the product name, amount declaration (e.g., “12g protein”), or no protein claim with less healthy and healthy Nutrition Facts Labels (NFL). Though the overall perceived healthfulness of products with protein labels did not appear to be greater compared to control cereals, there were some more nuanced differences between label types for other perceptions that were studied, indicating the presence of a health halo. Protein-named cereals were perceived as more likely to have the benefit of promoting healthier digestive systems, and as having more protein and more Vitamin D than control cereals. Participants who viewed cereals with the word protein in their names also viewed the corresponding NFL the longest compared to all other conditions. Previous research on products using “protein” in the name was only conducted on snack bars and did not examine attention to NFL for these products. The present study, as well as the prior Special K study, furthers this research by establishing that the phenomenon of health halos surrounding protein-titled foods extends beyond protein bars, but may not be universal. It is present in cereals, but not in yogurts.Ph.D.Includes bibliographical reference
Guild level response of human gut microbiome in mitigating oxygen-induced dysbiosis via dietary fibers and elemental iron
No full textThe human gut is predominantly inhabited by obligate anaerobic bacteria, maintaining a hypoxic environment. An increased ratio of facultative to obligate anaerobes indicates gut microbiome dysbiosis, associated with a range of chronic conditions including metabolic syndromes and inflammatory bowel diseases. Dietary fibers mitigate dysbiosis by producing short-chain fatty acids (SCFAs), regulating glucose homeostasis, and suppressing facultative anaerobic bacteria. Different processing methods alter the structure and fermentability of dietary fibers, shaping the microbiome accordingly. Antioxidants such as polyphenol and vitamin E also alleviate dysbiosis induced by oxidative stress. However, most studies targeting the relationships between food components and the gut microbiome use taxon-based analysis, which is limited by what is available in the database as well as resolution for differentiating signal from noise. In this study, guild-based analysis was applied to understand the food-microbiome relationship.
In chapter 1, we investigated the fermentability of dietary fibers with different processing methods. The raw oats and pinto beans were canned, boiled, extruded with high or low temperature, and introduced in the in vitro gut microbiome fermentation system after in vitro digestion. We hypothesized that different processing methods may alter the fiber structures. Bacteria harboring certain CAZymes may utilize fibers with various structures and proliferate, changing the whole microbiome. Results showed that materials and processing methods both influence the gas production and SCFA generation of gut microbiome. Guild-based analysis found that co-abundance group (CAG) containing facultative anaerobic bacteria such as Escherichia-Shigella ASV0010 were promoted in extruded oats, while CAG containing SCFA producers were significantly promoted in canned and high-temperature extruded pinto beans. This study illustrated that the fermentability of dietary fibers can be affected by processing methods. Therefore, when developing gut microbiota-targeted food, the effects of processing methods on fermentability need to be considered.
In chapter 2, we investigated if dietary fibers protect gut microbiome from oxygen-induced dysbiosis. In vitro fermentations utilizing human fecal samples were set up: Control (no oxygen, no fibers), Oxygen (oxygen, no fibers), Fiber (no oxygen, fibers), and Oxygen_Fiber (oxygen, fibers). The hypothesis was that a fiber formula with various dietary fibers can reverse the oxygen-induced disruption of gut microbiota by promoting the growth of anaerobic beneficial bacteria as a preferred energy source, generating SCFAs and lowering pH to repress detrimental bacteria. Results showed that microbiome structure shifted significantly with the introduction of oxygen, which was mitigated by dietary fibers. The Fiber and Oxygen_Fiber groups had lower pH levels and significantly higher SCFA production compared to the Oxygen group. Guilds with facultative anaerobes such as Escherichia-Shigella ASV0010 and Enterobacteriaceae ASV01P9 were significantly proliferated by oxygen but reduced following fiber treatment. In contrast, guilds with potential SCFA producers like Eubacterium ruminantium_ASV00LH were restored by fiber treatment under microaerobic conditions. The study suggested that dietary fibers can safeguard the gut microbiome from oxygen disruption, indicating a potential host-independent beneficial modulatory mechanism of dietary fibers on the gut microbiome.
In chapter 3, elemental iron was introduced to investigate its protective effect against oxygen-induced dysbiosis. The hypothesis was that elemental iron might alleviate dysbiosis in gut microbiota by scavenging oxygen and its derivatives. It was found that elemental iron partially reversed the oxygen-induced disruption of the gut microbiota. Inter-individual differences, corresponding to iron oxidation patterns, were observed for the tested donors. Guild with Escherichia-Shigella ASV0010 was promoted by oxygen and depressed by elemental iron, while guilds including bacteria such as Bifidobacterium ASV001D, Blautia ASV054R, were reversed by iron_oxygen treatment. SCFA production was inhibited by oxygen, which was partially reversed by elemental iron. These results suggest that elemental iron can partially regulate the oxygen/ROS state and protect the gut microbiota from oxidative stress.
Overall, this thesis illustrates that oxygen-induced dysbiosis can be alleviated by dietary fibers and elemental iron independent of host. Integrating the results from all three studies, key bacteria responding to oxygen and diet factors were identified by guild-based analysis. Some bacteria such as Escherichia-Shigella ASV0010 and Bifidobacterium ASV001D are common responders across different test donors. These bacteria are potential key players in developing microbiome-targeting food and require isolation and characterization in the future.Ph.D.Includes bibliographical reference
On generalized surgery operations for Lagrangian submanifolds
No full textA Lagrangian submanifold of a symplectic manifold is a half-dimensional submanifoldon which the restriction of the symplectic form vanishes. This dissertation explores the problem of constructing new Lagrangians from old by performing surgery type operations.
In the second chapter we extend the construction of higher mutation, as initially introduced by Pascaleff-Tonkonog, to local higher mutation. We prove that up to a change of local systems, the Lagrangian intersection Floer cohomology of a pair of Lagrangians remains invariant under local mutation. This result extends the wall-crossing formula presented in Pascaleff-Tonkonog.
The third chapter contains an account of a collaborative research with Amanda Hirschi and Luya Wang. We extend Vianna’s result about the existence of infinitely many monotone Lagrangian torus in projective space to any dimension. We define a surgery type operation called solid mutation, which is a circle fibered version of usual mutation, and show that lifts of Vianna tori, are related by solid mutations. We utilize Pascaleff-Tonkonog’s wall-crossing result to understand the effect of solid mutation on disk potentials.Ph.D.Includes bibliographical reference