DSpace@RPI (Rensselaer Polytechnic Institute)
Not a member yet
6809 research outputs found
Sort by
A dynamic high-order variational multiscale method on unstructured meshes for transport problems
No full textAugust 2021School of EngineeringIt is well known that the Galerkin method yields a solution with numerical/spurious oscillations in advection-dominated transport problems.Specifically when the cell Peclet or Reynolds number is high with an under resolution of the solution, e.g., in case of an under-resolved boundary layer solution.
A popular technique to resolve this issue is to use a stabilization term, in particular one employing the variational multiscale (VMS) and related stabilized methods.
The focus of this work is the formulation and application of a dynamic high-order VMS approach on unstructured
meshes for stationary and transient transport problems governed by partial differential equations. The current dynamic procedure takes in the given structure/form
of the stabilization parameter with unknown coefficients and computes
them dynamically in a local fashion resulting in a dynamic VMS-based
finite element method. A variational Germano identity (VGI) based local procedure suitable for unstructured meshes and high orders is developed to perform the dynamic computation of the coefficients in the stabilization parameter in a local fashion.The overall dynamic procedure based on the local VGI (LVGI) relies on a
sequence of locally coarsened spaces, i.e., secondary coarse-scale spaces, that are constructed from the
primary coarse-scale space.
To make the current procedure practical, any
locally coarser solution is reconstructed from the primary coarse-scale
solution, which is done over local patches.
Further, averaging steps are employed to make the local dynamic procedure robust. A wide range of stationary and transient problems are considered to demonstrate the suitability of the current high-order LVGI-based dynamic
procedure together with different subscale models. Both uniform and
nonuniform meshes are employed. Different \timestep sizes are used to
evaluate the behavior of different subscale models and stabilization
parameters with small and large \timestep sizes. Orders up to are
considered. In summary, the current dynamic high-order VMS method is
shown to be effective and provide more accurate results (especially on a
coarse discretization) for both stationary and transient problems.Ph
Accessing (or not) telemedicine to treat opioid use disorder during covid-19: an ethnographic study in new york state
No full textAugust 2023School of Humanities, Arts, and Social SciencesThis dissertation presents an in-person and digital ethnography of people in New York State who use drugs and seek treatment for opioid use disorder (OUD) using phone or video connection to receive healthcare (telecare) including interviews prior to and during the COVID-19 pandemic. The interviews focus on how the shift to telecare during the pandemic has affected people seeking treatment for opioid use disorder (OUD). I leverage a feminist Science and Technology Studies (STS) approach to elucidate how the political, clinical, and juridical framing of people who use drugs shapes the interconnections of care (or inaccessible care) that are discernable. The findings described below provide a heuristic for grasping how structural stigma has intersected with the U.S. treatment system during the COVID pandemic to reproduce increased rates of use, relapse, and overdose death. Telecare offers some benefits of increased access to medication for OUD (MOUD), but people need access to the technology and the knowledge and capacity to use it, and localized resources continue to be a key barrier for accessing treatment. The narratives of people seeking treatment are analyzed through the theoretical lenses of Nelly Oudshoorn’s analysis of the technogeography of care and Nancy Fraser’s analysis of the US juridical-administrative-therapeutic in/justice system. I offer the concept of the technogeography of harm reduction (THR) to help me trace and problematize how telecare contributes to redefining the experience of familiar places, such as home, into spaces of both care and surveillance, and how the technology of telecare presents both affordances and foreclosures to accessing care and reducing harms as people struggle to conform with the requirements of telecare in order to receive care. Key findings are that the significance of hugs and tactile connection is sorely missed by people using telecare for group therapy. The proximity to in-person services even while using telecare is critically important. The resistance strategies of telecare users to surveillance mechanisms shed light on how the system of treatment continues to fall short of meeting people where they are at and accepting that their self-defined goals should govern their treatment plan. The continued stigmatization of drug use and treatment acts as a key barrier to people who are striving to produce the identity of a patient who is clinically stable for take-home medication. Mental illness has been shown to be associated with an increased prevalence of substance use disorder (SUD) (RachBeisel, Scott, and Dixon 1999), and people who experience both mental illness and SUD must contend with a twofold barrier of stigma directed at both their mental illness and their SUD, a harsh reality that was present in several of my interviews. This dual diagnosis resents a key challenge for treatment providers to reach and treat these patients (Priester et al. 2016), especially in a treatment system that prioritizes those with money and resources, a key factor for access to telemedicine especially during the economic turmoil of the pandemic (Watson et al. 2022). The importance of peers for providing compassionate care, truly meeting people where they are at, emerged as a key element of successful treatment, for both telemedicine and in-person care. Peers have gone through the treatment system themselves, and some professional peer roles in NYS, such as the Certified Recovery Peer Advocate (CRPA), are not mandated reporters, which means that people can tell their CRPA that they had a relapse and the CRPA is not legally required to tell the judge or their doctor, thereby enabling a level of openness and the establishment of a therapeutic alliance (without any need to hide a relapse) which may be foreclosed to a mandated reporter, such as a Credentialed Alcoholism and Substance Abuse Councilor (CASAC).
The narratives described below provide a heuristic for understanding how addiction is (re)produced by the clinical and juridical systems that intersect with telecare for OUD, and one implication of my research is that in order to reverse the drug overdose crisis, the United States needs a complete reimagination and overhaul of the juridical and therapeutic systems that (re)produce the problem of addiction. A second implication centers on the prospect that people who use drugs (PWUD) may avoid using telecare phone apps because of concerns that they could be subject to data and privacy breaches, an issue that grows more urgent and potentially harmful as increasing numbers of people develop a reliance upon apps for telecare. Therefore, a second implication of this research is to affirm the importance of legal and professional protections of digital individual privacy and algorithmic fairness, a topic that is increasingly important as AI machine learning enters the space of healthcare apps and telecare.Ph
Inflammation and immunomodulation: considerations towards developing treatments for symptoms of autism spectrum disorder
No full textMay 2023School of EngineeringAutism Spectrum Disorder (ASD) is an increasingly prevalent developmental disorder characterized by moderate to debilitating social, behavioral, and communicative deficits. It is frequently accompanied with one or more co-occurring conditions that both diminish the quality of life of individuals with ASD and can significantly increase the financial burden on families for medical and care costs. Despite the growing urgency to address ASD, there currently are no broadly accepted treatments or biological diagnostic methods for the disorder. Research into these have been hampered by the high degree of heterogeneity in ASD and the lack of a clear etiology for the disorder. However, studies have identified consistent pathological features – aberrant neuroanatomical structure, chronically altered immunological state, and metabolic dysfunctions – which has driven research towards exploring these avenues for pharmacological intervention. While several therapeutic strategies have demonstrated some clinical promise, there remains large populations of patients whose symptoms either see no improvement or worsen within most trials. This necessitates clarification of both the targeted pathological features and the cellular mechanistic underpinnings of the treatments in order to develop broadly effective pharmacological strategies.The work herein focuses on three pharmaceuticals that have shown some clinical promise for treatment of core symptoms of ASD: 1) cannabidiol (CBD), 2) Tocilizumab (Toc), and 3) suramin (Sura). These three treatments focus on dysregulated components of the immune system that are potentially complicit in the neuroanatomical abnormalities observed in ASD, namely through endocannabinoid, interleukin 6 (IL-6), and purinergic signaling.
xv
CBD, an anti-inflammatory and anxiolytic phytocannabinoid, targets the endocannabinoid system (eCBS), a short-range signaling pathway with roles in immunomodulation and synaptic depression. As such CBD has shown some efficacy in small clinical trials, but much remains to be elucidated. Towards this, we expand upon previous studies characterizing the anti-inflammatory effects of CBD on peripheral immune cells in terms of secreted cytokines, chemokines, and growth factors as well as provide novel insights into their mechanistic effects on autophagy and oxidative stress. We also explore the effects of CBD on microglia and neurons under ASD-mimetic inflammatory stimuli – IL-6 and purinergic adenosine triphosphate (ATP) – and demonstrate its capacity to ameliorate ASD-mimetic deficits to inflammatory markers and synaptic protein expression.
Following, we further explore IL-6 and purinergic targeted treatments with Toc, an anti-IL-6R monoclonal antibody treatment, and general antipurinergic P2 receptor antagonist Sura. This was done in terms of inherent dysregulations exhibited by neurons differentiated from induced pluripotent stem cells derived from individuals with a monogenetic syndromic form of ASD, Fragile X Syndrome (FXS). We demonstrate that Toc and Sura both have similar beneficial effects on the inherent dysregulations to components of the transmethylation and transsulfuration pathways and secreted proteins in FXS neurons.Ph
Iterative image reconstruction for electrical impedance tomography using adaptive techniques
No full textDecember 2014School of EngineeringThis thesis will focus on the development of EIT reconstruction methods that strike a balance between reconstruction accuracy and cost efficiency in solving forward and inverse problems. Previous work has focused more on the accuracy of the algorithms rather than the efficiency of the algorithms. In the forward problem, an overly refined forward model is always used to obtained the predicted voltages to ensure the accuracy of the forward solution. However, the accuracy should only achieve the measurement precision of the hardware equipment and any refinement effort to improve beyond this limited precision will be a waste of computation power. In the inverse problem, direct reconstruction methods, e.g. NOSER and D-bar, are considered to be efficient but less accurate. Iterative methods, e.g. Gauss-Newton, are cost expensive and limited their use in only small-scale problems. Another issue with the inverse problem is that fixed uniform meshes are always used for reconstruction, which may generate a large number of elements that are not necessarily needed and lead to waste in computational power. To improve efficiency without reducing the accuracy of the solutions in both the forward and inverse problems, the thesis will first study the effects of the FEM mesh on the accuracy of forward solution and suggest a relatively accurate and efficient model size for the three-dimensional cylinder geometry. Secondly, an efficient reconstruction algorithm will be proposed to adaptively improve the accuracy of the reconstructed images using optimal current patterns. It will be shown that accurate and stable solutions will be obtained with much lower memory cost and storage. The algorithm is further improved by combining it with adaptive meshing to adaptively determine the reconstruction mesh at each iteration. The combination could optimize the reconstruction process with efficient meshing and optimal current patterns. Thirdly, a method in estimating the relationship between the ventilation and the perfusion during the breathing cycle is proposed and evaluated using human subject data.Ph
Non-convex optimizations for machine learning with theoretical guarantee: robust matrix completion and neural network learning
No full textDecember 2021School of EngineeringDespite the recent development in machine learning, most learning systems are still under the concept of “black box”, where the performance cannot be understood and derived. With the rise of safety and privacy concerns in public, designing an explainable learning system has become a new trend in machine learning. In general, many machine learning problems are formulated as minimizing (or maximizing) some loss function. Since real data are most likely generated from non-linear models, the loss function is non-convex in general. Unlike the convex optimization problem, gradient descent algorithms will be trapped in spurious local minima in solving non-convex optimization. Therefore, it is challenging to provide explainable algorithms when studying non-convex optimization problems. In this thesis, two popular non-convex problems are studied: (1) low-rank matrix completion and (2) neural network learning. In low-rank matrix completion (MC), the objective is to recover a low-rank matrix from partial observations that may contain significant errors. MC problem is non-convex due to the natural constraint of low-rankness. However, the low-rank structure does not capture the temporal correlations in some time series, i.e., power system monitoring, magnetic resonance imaging, and array signal processing. As a result, low-rank MC cannot handle the whole column/row being fully lost. In this thesis, a new model, termed multichannel Hankel matrices, is proposed to characterize the intrinsic low-dimensional structures in some multichannel time series. By exploiting the new model in this thesis, several projected gradient-based algorithms are developed to solve the non-convex MC problems with fully lost/corrupted columns. In neural network learning, a reliable learned model requires a small generalization error, which simultaneously achieves a small training error and generalization gap. According to the classic generalization theories, bounded generalization requires a larger number of training samples than the model complexity. However, solving the optimization problems with such a number of samples is not guaranteed to find a local minimum with a small training error due to the high non-convexity of the objective functions. Therefore, studying the convergence to the global optimum when training neural networks is vital and challenging. This thesis provides the convergence analysis to the global optimum when the number of samples is larger than the model complexity for one-hidden-layer neural networks with Gaussian inputs. Also, the minimal required training samples to guarantee zero generalization error are presented for various neural network architectures. Nevertheless, there are cases where the training process is not accessible to adequate training samples due to the difficulty of generating reliable data. Therefore, this thesis further explores the methods with a limited number of training samples, focusing on network pruning and self-training algorithms. The motivation for studying self-training comes naturally from a semi-supervised framework, which leverages many unlabeled data to improve learning when the labeled data are limited. In contrast, the network pruning is mainly inspired by the recent Lottery Ticket Hypothesis (LTH), which claims that a good pruned network achieves a faster convergence rate and higher test accuracy than the original dense network.Ph
Material politics in sound
No full textDecember 2022School of Humanities, Arts, and Social SciencesIn this dissertation, “Material Politics in Sound,” I utilize an experimental methodology that prioritizes a material-based practice in art and electricity, and in my artistic and scholarly analysis I explore my artistic practice that I refer to as “Material Politics in Sound “(MPS). Fostered by the disciplines of photography, sculpture, experimental electronic music, perception, and circuitry, “Material Politics in Sound” is an examination of sensory thresholds and influences mediated by the material world through lenses of both deception and care. In this dissertation, I outline influences from composers and artists who utilize technology within their conceptual framework.
In particular, I am interested in the practices and psychologies behind the making of technologies in relationship to concepts of agency and representation in the world as related to our sensory thresholds. I suggest that by considering sensory experiences and identities outside of the standards perpetuated through our technologies, and by examining the materials that mediate them, this might engender a more diverse and rich opportunity for a wider spectrum of representation in the arts.
Through my most recent work, Sonic Spells (EMPAC, 2022) and an analysis of the many collaborative and experimental pieces prior, I demonstrate the practice of a material-based approach to art and technology-making that considers the importance of sensory curiosity, care, and potential deceptions embedded within them. These practices, experiments and conceptual frameworks contribute to the many artistic practices of artists, engineers, and makers working to bring awareness to represent a myriad of sensory capabilities within the electronic arts. Most importantly, this work joins others in the fields of care work and disability justice, with the intension of bringing awareness to the material tangible influence our practices and making have to a greater political and socio-economic system.Ph
Design analysis of 3d printed internal cavity lens for lighting applications
No full textDecember 2022School of ArchitectureLED lighting systems consist of an LED light source and several subsystems, including optical, electrical, and thermomechanical components. The optical subsystem uses reflective, refractive, or a combination of reflective and refractive components to transfer the luminous flux from the LED light source to the target area to satisfy the application’s light level and distribution requirements. A refractive optic or lens is a transparent material with a given index of refraction and a shaped external surface that redirects the incoming beam. Refractive optics exposed to the application environment can cause lumen depreciation when dust and dirt accumulate within the crevices of its contoured external surfaces. Such light loss can cause the secondary optics to become less effective over time. Furthermore, lenses with non-flat external surfaces pose challenges when assembling LED lighting systems. An optic with internal refractive cavities and flat, smooth external surfaces can reduce such problems. Therefore, this dissertation study aims to investigate the design strategy and manufacturing method of internal-cavity flat optics for use in LED illumination systems. Although many studies have examined lens design methods for external refractive surfaces, no study has proposed design strategies for internal cavity lenses with planar external surfaces. In this dissertation study, a novel design method was investigated and analyzed for creating flat lenses with internal cavities. The internal refractive cavity is created between two internal freeform surfaces that are formed based on the light-energy mapping method and that consider the edge ray principle and Snell’s law. The study objectives of this research project were established by understanding the knowledge gaps in the areas of freeform lens design algorithms for designing internal cavity lenses and manufacturing methods for such lenses. First, an algorithm was developed for designing the surfaces of the internal refractive cavity by extending the freeform design method for producing a circular symmetric light distribution on a target plane. Later, this algorithm was advanced to achieve non-circular symmetric distributions as well. In both cases, a mathematical relationship was formed based on geometric optics to design the internal refractive surface geometries. Defined surface geometries were used to generate the 3D model of the internal cavity lens. The design method was validated using the results from a Monte Carlo ray-tracing simulation study and a laboratory experiment that analyzed the output beam of the 3D-printed internal cavity lenses. Tolerance analyses were performed to assess the effects of different design parameters on the beam quality and lens efficiency. The initial ray-tracing results demonstrated that the proposed lens design method can construct internal cavity lenses with optical efficiencies of 83% and 80%. The calculated uniformities were 1:1.9 and 1:2.3 during the ray-tracing simulations. The experiment results showed that for the two 3D-printed flat internal cavity lenses that formed the different beam patterns, the optical efficiencies were 72% and 70% and the beam uniformities were 1:2.2 and 1:2.2 with the material and 3D printer used in this study. The lower optical efficiencies in the experimental results could be mainly due to Fresnel and scattering losses incorporated with the 3D-printed lenses. To the best of the author’s knowledge, the proposed internal cavity lens design strategy-that is, by simultaneously designing a pair of internal refractive surfaces by extending the freeform method has never been studied before. Therefore, the new knowledge contribution to the field of optics through this dissertation study includes a design method for a flat internal cavity lens and the manufacturing of such lenses. The author believes this study will open up a new field of exploration for the concept of internal refractive cavity lenses. Moreover, it will provide information on how to use additive manufacturing optics to develop new optical designs that add value to a wide range of applications.Ph
Resource-rational cognitive modelling : an information-theoretic approach
No full textDecember 2022School of Humanities, Arts, and Social SciencesHow do humans coordinate perception and memory when learning and making decisions? Additionally, how do cognitive limitations and behavioural goals influence the optimal functioning of these faculties? Many accounts have sought to explain one or more of these faculties and how they impact behaviour. Relatively little attention has been given to how these cognitive faculties and goals are interrelated. This thesis will provide an account for how the human mind might optimally coordinate perception and memory with learning and decision making, relative to individual cognitive constraints. To achieve this goal, this thesis presents a cognitive model inspired by two areas of research. Firstly, computational modelling of biological visual perception and memory, which seeks to understand and predict these cognitive functions. Secondly, resource-rational analysis which seeks to understand how cognitive agents behave optimally under cognitive constraints, specifically information-theoretic constraints. The result of these connections is a cognitive model that makes predictions of perception, memory, learning, and decision making, while explaining how individuals coordinate these faculties relative to their goals and limitations. This model is first applied onto predicting human behaviour in a visual learning task collected in a previous experiment. Next, two novel experiments are introduced that incorporate utility judgements, change detection, and learning. Results from these experiments demonstrate that the proposed model is better able to account for detailed aspects of human behaviour compared to related methods. This improvement is due to the successful integration of multiple areas of research in biological perception and memory with learning and decision making, all under the resource-rational approach to cognitive modelling. This thesis concludes with a broad discussion of the importance of the proposed model and how it relates to remaining open questions in computational models of biological perception, memory, learning, and decision making.Ph
Made more radiant : "psychic practice" and other somatic technologies for re-enchantment and recovery
No full textDecember 2022School of Humanities, Arts, and Social SciencesThis practice-based dissertation, “Made More Radiant: ‘psychic practice’ and other somatic technologies for re-enchantment and recovery,” is grounded in “experimental dance” and queer and feminist performance art. The heart of the project is what I call “psychic practice,” a way to activate embodied re-enchantment and intimacy with self, other, and place. This project emerged collaboratively in March 2020 and is now ongoing. It consists of remote, one-on-one sessions that include a phone conversation, improvised dancing, “psychically” witnessing practice partners, and writing. Based on years of contemplative movement practices and dance-based performances, as well as teaching movement and performance workshops, I have noticed ways of orienting that are the foundation of this creative activity. I have formulated these into four unique somatic technologies/philosophies, or guides to this embodied practice: 1. The Eroticism of Nothing(ness), 2. WitHnessing, 3. Flesh Circus, and 4. Moshing with Multitudes. Augmenting “psychic practice” is a video piece that acts as a creative document of Made More Radiant (in prismatic undertaking), an in-process performance and installation work that I began developing at Rensselaer’s Experimental Media and Performing Arts Center in May 2022. This project expands upon the practice-as-research around re-enchantment and intimacy with self, other, and world conducted through the last two years of “psychic practice,” and encompasses dance, sound, animation, text, and sculpture. A corresponding publication, which creatively documents “psychic practice” through image, poetic text, instructions for practice, and excerpts from this dissertation text will be released in December 2022 through Publication Studio, Hudson. This dissertation text contextualizes “psychic practice,” the video/installation and performance, and the publication through an articulation of my motivations, intentions, and my creative constellations of influence and creative and scholarly communities.Ph
