University at Albany, State University of New York
University at Albany, State University of New York (SUNY): Scholars ArchiveNot a member yet
6197 research outputs found
Sort by
Big House, from Conjoined States
No full text“Big House” is the third section from Conjoined States, a novel that depicts the United States’ dependence on forced labor, incarceration, and surveillance. In 1971, two years after Ohio’s governor deploys the National Guard to violently suppress a protest in Columbus, Ella Schmidt and her parents relocate to Cleveland, where they experience renewed displacement due to racism. The child of an Austrian father and Barbadian mother, Ella endures bullying in middle school but finds solace in high school in the writings of Jane Addams. Inspired by Addams’s career in social work, Ella dreams of repairing Cleveland’s abandoned houses so their former residents can return. Two people believe Ella’s plan could change the city: her outspoken Puerto Rican classmate, Rita Lugo, who insists Cleveland will benefit from Ella’s presence no matter what; and Mrs. Leonardi, a teacher who offers to help Ella enact her plan—but only in exchange for intel that could lead to Rita’s arrest. Informed by Critical Race Theory and loosely based on Italo Calvino’s Invisible Cities, “Big House” combines real accounts of police brutality with imagined details that empower the surveillance state, offering a glimpse of what happens when schools and churches operate as intelligence arms for the police
The Two Faces of Workplace Digital Surveillance: A Mixed-Method Study of Monitoring Technologies\u27 Affordances and Employee Experiences
No full textThe rapid proliferation of employee electronic monitoring technologies presents organizations with complex challenges at the intersection of productivity, privacy, and workplace trust. While these technologies have become increasingly prevalent, with adoption rates surging 58% between 2020 and 2021, existing research has failed to systematically analyze both the technological capabilities and human experiences of workplace surveillance. This dissertation addresses this gap through two complementary studies examining workplace monitoring from technological and human perspectives.
The first study uses qualitative analysis of vendor materials and product documentation for 10 monitoring tools. The study identifies three hierarchical levels of monitoring affordances: observational affordances focusing on surveillance and visibility enhancement, operational affordances enabling automated decision-making and enforcement, and strategic affordances supporting resource and compliance management. A framework for understanding workplace monitoring technologies through the lens of affordance theory is developed to advance our understanding of how monitoring technologies enable different forms of organizational control while highlighting potential areas for technological improvement and policy intervention.
The second study employs advanced topic modeling techniques to analyze 4,188 Reddit posts and 120,381 comments discussing workplace monitoring. Using Latent Dirichlet Allocation and BERTopic algorithms, the analysis reveals 16 distinct topics of employee concern, with topic coherence scores exceeding 0.7. Key themes include privacy concerns with personal device monitoring, debates about remote work surveillance, and sophisticated employee response strategies. The second study\u27s findings demonstrate the complex interplay between technological surveillance and workplace trust, particularly highlighting employees\u27 concerns about the blurred boundaries between personal and professional digital spaces.
Together, these studies reveal fundamental tensions between the expanding technical capabilities of monitoring systems and employee acceptance of monitoring practices. The dissertation makes three primary contributions: (1) developing a structured framework for analyzing monitoring technologies through affordance theory, (2) providing empirical evidence of employee perspectives and response strategies, and (3) offering practical guidance for balancing organizational control needs with employee privacy concerns. These insights are particularly relevant as organizations navigate the challenges of remote work and the increasing digitalization of work
Using Convolutional Neural Networks to Identify Rare Weather Events: Application to Kona Low Classifications with Large-Scale Wind Pattern
Full text linkExtreme events pose challenges in prediction on account of their low frequency and erratic nature. Among these events, Kona lows are rare subtropical cyclones that significantly impact the weather pattern of Hawaii, sometimes causing heavy rainfall, strong winds, and coastal flooding. In the past, studies have used traditional statistical methods such as synoptic climatology, regression analysis, and manual classification that rely on expert judgment to identify Kona lows; however, they often struggle to capture the complex spatial patterns of the Kona storms. Convolutional Neural Networks (CNNs) exhibit a property called translation invariance, which allows them to learn and recognize Kona lows regardless of any changes in the spatial patterns. The goal of this study was to assess the CNN’s ability in correctly identifying extreme events. It uses the zonal wind data of Hawaii from the ERA5 reanalysis between the years 1990 and 2010 as an input. Several CNNs are trained using techniques like undersampling, oversampling, and manual weight distribution to overcome data imbalance, which occurs due to the high frequency of non-extreme events compared to the rare Kona lows. The best model achieved an Extreme Dependency Score (EDS) of 0.84, where 1 represents the best possible score reflecting perfect prediction with no false positives and a frequency bias of 0.90, which suggested slight underprediction. Combining EDS with the bias provides a more comprehensive picture where the scores indicate that the model is well calibrated and non-biased. The CNN model outperforms the reference random prediction and a simple artificial neural network model that uses the numerical PC1 ( First Principal Component ) and PC2 (Second Principal Component ) values derived from the Kona lows as an input. This is followed by a physical interpretation of the forecast using one of the Explainable AI (XAI) approaches. Randomized Input Sampling for Explanation (RISE) explains how the CNN model makes decisions by randomly masking a part of the input image to see how it affects the model output. The composite of the resulting heat map of the true positive events indicates the features of the zonal wind that are important for prediction .The heat map is consistent with the observed trend that Kona storms typically form to the southwest of Hawaii, as this area shows the highest importance, followed by the region surrounding the Hawaiian islands. These results demonstrate the potential of using deep learning techniques in forecasting rare events, promising avenues of future research in climate risk assessment
The Invisible Costs of Privatizing Welfare: Administrative Burdens in Contracted Medicaid
Full text linkWhy do governments favor contracting out social services? Do private actors guarantee a better service experience for citizens? How should we understand the trade-off between efficiency and high-quality service delivery? These are the central questions that motivate this dissertation. Focusing on Medicaid, one of the largest social safety net programs in the United States, this dissertation investigates the impact of outsourcing Medicaid delivery to private health insurance companies. It explores how this privatized structure generates administrative burdens and shapes policy feedback effects. Theoretically, this project seeks to bridge the administrative burden and policy feedback literature by examining how burdensome policy experiences influence individuals’ political engagement. Employing a mixed-methods approach, this dissertation combines causal inference techniques, in-depth interviews with health Navigators and a nationally representative survey experiment.
In sum, this dissertation finds that outsourcing social services does not necessarily improve access or service quality. Although contracting out is often introduced to enhance efficiency in public programs, private actors may achieve cost reductions by restricting access to services. Additionally, the research shows that private service providers can impose further administrative burdens when Medicaid beneficiaries seek care. By exercising bureaucratic discretion as a form of “hidden politics,” private entities shape how individuals experience and interact with government programs. Moreover, respondents perceived higher administrative burdens when the service provider was private, and greater burdens were associated with increased political engagement. These findings suggest that both the type and source of administrative burden play a critical role in shaping political behavior
Development of Group III-V Quantum Confinement-Enabled Detectors: Bias-Tunable Quantum Well Infrared Photodetector (QWIP) and Quantum Dot Scintillation Detector (QDSD)
No full textThis dissertation discloses the physics, fabrication, characterization, and analysis of two novel types of group III-V semiconductor detectors relying on quantum confinement of carriers, namely voltage-tunable quantum well infrared photodetectors (QWIP) and a high-yield ultrafast quantum dot scintillation detector (QDSD). Both QWIP and QDSD heterostructures presented here were grown on 3” GaAs (001) substrates using molecular beam epitaxy (MBE).
A major part of the dissertation focuses on development of the voltage-tunable QWIPs targeting detection in the mid-wave infrared region (MWIR) (3μm -5μm) and long-wave infrared region (LWIR) (8μm -12μm) with control of sensitivity by the applied bias. The QWIPs utilize the design consisting of asymmetrically doped double QWs, where the external electric field alters the electron population in the wells and hence the spectral responsivity of the device. The design rules are obtained by calculating the electronic transition energies for symmetric and antisymmetric double-QW states using a 1D Schrödinger–Poisson solver. The MBE-grown semiconductor heterostructure contains 25 periods of coupled double GaAs QWs and AlxGa1-xAs barriers. One of the QWs in the pair is modulation-doped to provide asymmetry in potential between the two QWs. Standard fabrication techniques, containing UV lithography, contact metallization, and wet chemical etching are followed to build mesa structures of 2 different sizes. The QWIPs are characterized with blackbody radiation to evaluate the responsivity, detectivity, and with FTIR down to 77 K to evaluate the detection wavelength and its control with applied external voltage. From the ratio of the responsivities of the two prominent detection bands, voltage tunability of the detectors were evaluated. Depending on the growth parameters (QW width, Barrier width, Al concentration) of the detector, a QWIP with voltage-tunable detection band between MWIR (~5 μm) and LWIR (~8 μm) with an order of magnitude sensitivity control is demonstrated.
Voltage tunable QWIPs with detection in LWIR (8 -11 μm) show a sensitivity control of 2x with applied bias of +/- 4V, which is further optimized to gain a sensitivity control of 10x. Due to the voltage tunable IR detection capability, these QWIPs become a promising candidate for various applications such as Military and defense, environmental monitoring and AI-based object recognition.
The last chapter of this dissertation discusses the development of a novel InAs/GaAs quantum dot (QD) scintillation detector exhibiting high light-yield and high energy resolution. High speed and efficiency of this semiconductor QD scintillator makes it a promising alternative for medical imaging, nuclear security and other high-energy physics applications. The MBE-grown epitaxial semiconductor heterostructure contains self-assembled InAs QDs serving as artificial luminescent centers, converting the kinetic energy of incoming charged particles into photons, which are then collected by a monolithically integrated InGaAs p-i-n photodiode through waveguiding. These novel devices often suffered from poor energy resolution and light yield, exhibiting high variance responses to monoenergetic sources which significantly reduces detectors accuracy and precision. To mitigate this issue, we developed a 26-micron-thick scintillator demonstrating a yield about 35 electrons/keV (~15% of the achievable maximum), with an energy resolution of 4.4% using 5.5MeV alpha particle. The intrinsic resolution of the scintillating material is evaluated to be 1.9%. The collection of charges from different regions of the scintillator is studied using the multimodal response observed from flood exposure to 4.4 MeV alpha particle. The detectors response to gamma photons is presented using Ba-133 source. A hybrid response due to ionizing track share between scintillator and photodetector (PD) is observed with maximum yield of 45 electrons/keV. Noise-limited time resolution of 59ps is demonstrated by this high-yield detector which can be further improved with cleaner fabrication and better readout electronics
Dysregulation of the Transcriptome in Murine Models of CAG Expansion Spinocerebellar Ataxias
No full textSpinocerebellar ataxias (SCAs) are a group of more than 40 genetically heterogeneous neurodegenerative disorders, numerous of which are caused by CAG repeat expansions in coding regions that generate toxic polyglutamine (polyQ) tracts. Although these CAG repeat expansions drive progressive neurodegeneration, particularly in the cerebellum, the molecular mechanisms linking the mutation to disease pathogenesis remain poorly understood. Recently, dysregulation of alternative splicing has emerged as a transcriptomic hallmark of CAG expansion SCAs, yet most work to date has focused exclusively on the skipped exon (SE) event class and largely on SCA1 models.
To address this gap, I performed the first comprehensive investigation of all five major alternative splicing event classes skipped exon, (SE), retained intron (RI), mutually exclusive exons (MXE), alternative 5′ splice site (A5’SS), and alternative 3′ splice site (A3’SS), across multiple CAG expansion SCA mouse models. This analysis revealed widespread missplicing across all splicing types and demonstrated that alternative splicing dysregulation is a shared transcriptomic feature of SCAs. Splicing alterations were enriched in pathways relevant to SCA pathogenesis, including neuronal structure and function, cytoskeletal processes, and ion channel regulation. Importantly, dysregulated events were highly expressed in cerebellar Purkinje neurons, supporting a direct link between alternative splicing defects and selective neuronal degeneration.
To further evaluate the contribution of alternative splicing to disease progression and therapeutic response, I characterized the transcriptome of SCA3 Q84 mice in the v cerebellum and brainstem following treatment with an antisense oligonucleotide (ASO) targeting ATXN3. SCA3 mice exhibited robust dysregulation of alternative splicing and differential gene expression in both brain regions, with SE events representing the most abundant splicing alteration. ASO-5 treatment partially restored normal splicing patterns, demonstrating that alternative splicing represents a molecular marker of therapeutic target engagement.
Collectively, this work establishes that dysregulation of multiple classes of alternative splicing is a transcriptomic feature of CAG expansion SCAs, broadening the current mechanistic understanding of disease pathogenesis and highlights the potential of alternative splicing as a promising biomarker
Pathways to Disclosure and Treatment among Individuals with an Attraction to Minors
No full textChild sexual abuse prevention is a significant social concern in the United States and internationally. While there have been some successful efforts to reduce recidivism risk among individuals who have committed sexual offenses, we fall short when preventing first time offenses. Therefore, looking at ways to provide treatment to those who are more at-risk for a first offense seems like a potential solution to this problem. One such relevant risk-factor is attraction to minors. This study is a qualitative examination using Interpretive Phenomenological Analysis of how people who are attracted to minors make the decision to disclose their attraction, seek treatment and experience treatment. Non-offending minor-attracted persons (N=17) were interviewed in an effort to understand how they come to their decisions to disclose and seek treatment, as well as their experiences once they are in treatment. Findings suggest that there are multiple pathways to disclosure, treatment, and transitioning out of treatment and that feelings of acceptance play a significant role in the decision to disclose and to continue working with a treatment provider. The experiences of this population are discussed in detail
Advancing 4H-Silicon Carbide Power Mosfets through Three-Dimensional Technology Computer Aided Design Optimization
No full textThis research contributes to the advancement of 1.2 kV 4H-Silicon Carbide (SiC) Metal-Oxide-Semiconductor Field-Effect Transistors (MOSFETs) through the development and implementation of sophisticated three-dimensional Technology Computer-Aided Design (3D TCAD) methodologies. These advanced simulation techniques enable comprehensive evaluation of innovative unit cell architectures that remain inaccessible to conventional two-dimensional (2D) TCAD approaches. By leveraging these 3D simulation capabilities, this work facilitates significant improvements in both the performance and reliability of next-generation wide-bandgap power devices.
Power devices convert electrical energy between different forms throughout the power grid making them essential in applications ranging from personal electronics, electric vehicles, renewable energy systems, industrial equipment, and data centers. Growing demand for these applications, combined with the drive toward a more sustainable electrical grid, requires developing increasingly efficient power devices that minimize resistance and energy losses during voltage control, current regulation, and switching operations. Historically, power devices were primarily fabricated using silicon, but the push for greater efficiency has driven exploration of wide bandgap materials offering superior performance characteristics. Among these alternatives, 4H-SiC has emerged as one of the most promising candidates, demonstrating exceptional capabilities that have already enabled its successful commercialization, especially 1.2 kV rated MOSFETs.
Persistent challenges hinder 4H-SiC MOSFET advancement, most notably the poor channel mobility stemming from defects at the 4H-SiC epitaxial layer/SiO2 gate oxide interface. These defects create electron trapping sites and scattering centers that significantly degrade carrier transport, increasing specific on-resistance (Ron,sp) and switching losses. Despite extensive efforts to improve this interface, channel mobility remains below 30 cm²/V·s.
Alternative approaches to improve the electrical characteristics of 1.2kV 4H-SiC MOSFETs involve device and layout optimization. However, these approaches were previously constrained by the limitations of 2D TCAD simulations, which cannot fully evaluate novel device architectures with non-linear device geometries. The development and deployment of 3D TCAD simulation now enables the comprehensive evaluation of these novel device designs prior to fabrication, accelerating the optimization of 1.2kV SiC MOSFETs.
This work represents a critical step forward in power device optimization. By deploying 3D TCAD simulations, researchers can now evaluate numerous device architectures to identify and better optimize critical regions in these devices prior to fabrication. This capability enables the development of more efficient power devices essential for meeting the growing energy demands of modern society while advancing sustainability goals across multiple technological sectors
Opioid-Associated Out-of-Hospital Cardiac Arrest
No full textOut-of-hospital cardiac arrest (OHCA) remains a leading cause of mortality in the United States, with opioid-associated OHCA (OA-OHCA) presenting unique challenges, including during the COVID-19 pandemic. Understanding individual-, community-, and policy-level factors shaping risk for OA-OHCA, as well as survival and resuscitation outcomes after OA-OHCA, is essential in improving care and guiding interventions.
Using the 2018-2023 National EMS Information System (NEMSIS) dataset, I conducted multilevel logistic regression models and comparative interrupted time series (CITS) analyses to evaluate differences between OA-OHCA and non-OA-OHCA cases. Patient-, incident-, EMS-, community-, and state-level covariates were examined, including demographic characteristics, resuscitation practices, county-level social vulnerability (SVI), and state naloxone laws. Random intercepts accounted for clustering at the EMS agency level. Outcomes included return of spontaneous circulation (ROSC) and positive end-of-event status (presumptively alive). Time-series analyses assessed weekly rates of OHCA, ROSC, and survival before and during the COVID-19 pandemic.
Age was the strongest individual-level factor distinguishing OA-OHCA patients from non-OA-OHCA patients. Unadjusted ROSC was higher in OA-OHCA than non-OA-OHCA patients (28.3% vs. 25.3%), though this difference was largely explained by initial rhythm and bystander intervention. In fully adjusted models, OA-OHCA patients had similar odds of survival to non-OA-OHCA patients (aOR = 1.03; 95% CI [1.01-1.06]). Among OA-OHCA patients, witnessed status, shockable rhythm, and shorter EMS response time strongly predicted ROSC and survival. Arrests that occurred in counties with higher uninsured rates had lower survival odds. Comparative interrupted time series showed that both OA-OHCA and non-OA-OHCA rates increased significantly during the COVID-19 pandemic, while ROSC and survival declined modestly.
Both individual clinical characteristics and community-level disadvantage significantly influenced outcomes. The COVID-19 pandemic disrupted OHCA incidence, resuscitation, and survival trends, with greater negative impact on non-OA-OHCA cases. Policies supporting bystander intervention, EMS response, and naloxone access may mitigate disparities and improve outcomes for OHCA cases, especially in the event of future public health emergencies
Sulfur Chemistry for Protein Synthesis and Modification: From Thiolactone to Bicyclic Sulfoxide Frameworks
No full textThis dissertation explores two intersecting themes in modern chemical biology: (1) the design of novel ligands for transition-metal catalysis, and (2) the development of thiolactone-mediated strategies for protein synthesis and modification in aqueous media. Together, these studies advance chemical methods for precise molecular transformations under mild, biocompatible conditions.
In the first chapter, a geometrically constrained bicyclic [3.3.1] sulfoxide framework was developed as a new class of bissulfoxide ligands for palladium-catalyzed C–H functionalization. The ligand was synthesized concisely from thiophenol through a three-step sequence and shown to promote allylic C–H amination and alkylation reactions with high efficiency. The catalytic system tolerated diverse nucleophiles, including nitrogen- and carbon-based donors, and demonstrated broad functional-group compatibility. Mechanistic analysis suggested that the unique geometry of the [3.3.1] scaffold and hydrogen bonding between the sulfoxide and nucleophile synergistically facilitate metal activation and turnover. As a proof of concept, the methodology enabled a concise four-step synthesis of the antifungal drug Naftifine. The modular architecture of this ligand framework also offers potential for future incorporation of oligopeptide motifs, bridging organometallic catalysis with biomolecular design.
The second chapter focuses on expanding the scope of native chemical ligation (NCL) through thiolactone chemistry. A β-thiolactone-based activation strategy was developed to form native amide bonds directly from unfunctionalized peptide C-termini in water, bypassing the need for pre-installed thioesters or external thiol additives. Using isonitrile-mediated intramolecular activation, this method enabled efficient peptide segment condensation and iterative elongation cycles under mild aqueous conditions. The approach was further extended to cysteine-free ligations via β-thiolated amino acids and N-thiol auxiliaries, and applied to side-chain cyclization of the protein kinase inhibitor peptide IP20 to yield a conformationally constrained variant. Mechanistic and kinetic studies elucidated key structural determinants governing thiolactone formation, including the position of cysteine residues and the influence of terminal amino acids.
Finally, in chapter 3, the iterative thiolactone-mediated ligation platform was demonstrated through the total synthesis and oxidative folding of short neurotoxin 1 (SNT1), a 61-residue three-finger toxin, entirely in aqueous phase. Building on this foundation, future efforts will focus on incorporating sulfur handles or thioauxiliary motifs onto small-molecule and peptide modifiers—such as glycans, lipids, and monoamine donors—to enable chemoselective installation of post-translational-modification-like moieties under mild aqueous conditions.
Collectively, this work introduces a unified framework that integrates organometallic ligand design and peptide chemistry, offering versatile tools for C–H activation, bioconjugation, and protein synthesis in water. The methodologies developed herein expand the synthetic accessibility of both small-molecule catalysts and cysteine-rich proteins, contributing to the convergence of synthetic organic chemistry and chemical biology