203435 research outputs found
Sort by
Artificial Intelligence Approaches for Membrane Fouling Prediction
Get PDFAbstract
Membrane fouling is a persistent and critical challenge in membrane-based separation processes, resulting in reduced performance, higher energy consumption, and increased operational and maintenance costs. Addressing this issue is essential for improving the long-term sustainability and efficiency of membrane systems. This report provides a comprehensive overview of membrane fouling mechanisms—such as organic, inorganic, particulate, and biofouling—and explores the growing role of Artificial Intelligence (AI) in tackling this problem. Specifically, it reviews three widely adopted AI techniques: Artificial Neural Networks (ANN), Fuzzy Logic (FL), and Support Vector Machines (SVM). Each method is examined through a series of literature-based case studies, focusing on model structure, input parameters, data preprocessing strategies, and prediction performance.
The analysis highlights the unique strengths and limitations of each AI approach. ANN models are effective at capturing complex, nonlinear relationships in large datasets but may suffer from overfitting and require careful parameter tuning. FL offers interpretable rule-based systems and is well-suited for integrating expert knowledge, while SVM demonstrates superior performance in handling small and noisy datasets due to its robust margin-based optimization. The report also discusses how hybrid methods and the integration of fouling mechanism knowledge into AI models can further improve prediction accuracy and practical applicability.
Ultimately, this study shows that AI-based approaches have significant potential to support early fouling detection, optimize cleaning strategies, and reduce chemical consumption. The findings not only validate the effectiveness of AI in enhancing membrane fouling prediction but also suggest promising directions for future research, including real-time monitoring, hybrid modeling, and the incorporation of physical-chemical mechanisms into data-driven frameworks
Evaluation of Novel Extraction Strategies for the Mass Spectrometric Analysis of Drugs in Complex Biological Specimens for Clinical and Forensic Toxicology Purposes
No full textDrug abuse is a dynamic and widespread health issue that affects the quality and life expectancy of our society. Pain, an inherent aspect of life, is a complex phenomenon to treat humans that varies from patient to patient. The duress experienced by patients sometimes triggers the need to find alternative methods to cope with the pain, including alcohol, smoking, self-destructive behavior, or the abuse of recreational drugs. One of the most frequently prescribed medications for pain control is opioids. Investigating the prevalence of “hidden” drug entities in patients under rehabilitation therapy is challenging but can help in understanding adherence and abuse patterns. Long-term use of opioids may lead to tolerance in humans, which sometimes poses a risk of fatal overdoses. Recent drug reports highlight the increase in seized opioids, which is also reflected by the number of associated deaths, escalating to epidemic levels. According to the National Institute on Drug Abuse (NIDA), 81,806 opioid-related overdose deaths occurred in 2022. Therefore, developing efficient and novel approaches that offer superior sample preparation schemes is highly desirable in clinical and toxicology scenarios. To this end, this study aims to evaluate novel toxicology extraction strategies applied to oral fluid and submandibular salivary gland tissue specimens analyzed by LC-MS/MS to detect a large suite of drugs. The drug panel includes fentanyl and other synthetic opioids, cathinones, amphetamines, cocaine, PCP, and Δ9-tetrahydrocannabinol (THC) for a total of thirty-five analytes.
The validation of the novel extraction methods followed the current American National Standards Institute/American Academy of Forensic Science Standards Board ANSI/ASB Standard 036 (ASB 036) for method validation in forensic toxicology. The validation was completed for quantitative analysis, which includes the parameters of bias, precision, carryover, limit of detection, limit of quantitation, interference studies, ionization suppression/enhancement, one to four dilution integrity, and room-temperature processed sample stability.
Two extraction methods were compared to determine the efficacy of the extraction of a large suite of drugs from oral fluid. The first extraction method includes a solid phase extraction with a mixed-mode cation exchange sorbent cartridge (MCX). The second extraction method includes protein precipitation (PP) followed by filtration through a Nanosep® centrifuge device. Extraction protocols were compared for the efficiency of the extraction in the terms of matrix effects, recovery, dilution integrity, sample stability, time, and cost of extraction.
The applicability of the extraction strategies was assessed through the analysis of twenty-five authentic antemortem oral fluid specimens collected from deidentified individuals in a drug rehabilitation care facility over an eight-month period. These specimens were first analyzed through a previously validated method, QuEChERS, which stands for Quick, Easy, Cheap, Effective, Rugged, and Safe. Results were compared with those obtained through MCX and PP filtration extraction of the same twenty-five case samples. Buprenorphine, naloxone, and methamphetamine were the most frequently detected analytes throughout this dataset.
Another aspect of this project involves applying a modified QuEChERS method to analyze postmortem submandibular gland tissue. Twenty-two submandibular gland tissue specimens were collected from deidentified individuals to assess the utility of the QuEChERS extraction method on this biological matrix. Fourteen matched submandibular gland and postmortem whole blood were analyzed for comparison and to assess the validity of the submandibular gland specimens as an alternative matrix. In the matched blood and submandibular gland tissue specimens, amphetamine and methamphetamine, were the most frequently detected analytes ranging from 25.3 to 315.6 ng/mL and 6.9 to 2857.6 ng/mL, respectively, for the whole blood and ranging from 3.6 to 227.3 µg/kg and 4.1 to 2072.9 µg/kg, respectively, for the submandibular gland. The benefits of the QuEChERS extraction method on two alternative biological matrices, oral fluid and submandibular gland tissue, are demonstrated in this study for utilization in toxicology casework
Cleaving through the Chaos: RNase-III Drosha’s Cleavage and Cellular Translocation in Response to SARS-CoV-2 Infection
No full textSince its emergence, COVID-19—caused by the novel coronavirus SARS-CoV-2—has affected millions globally and led to over 1.2 million deaths in the United States alone. This global impact, coupled with the emergence of five new human coronaviruses over the past two decades, underscore es the urgency of understanding their pathogenic mechanisms at the molecular level—not only for managing the current pandemic but for preparing for future outbreaks. Small non-coding RNAs (sncRNAs) critically regulate host and viral gene expression, including antiviral responses. Among the molecular regulators implicated in antiviral defense, the microRNA-processing enzyme Drosha has emerged as a particularly intriguing factor. In addition to its canonical role, Drosha also exerts a non-canonical, interferon-independent antiviral function against several RNA viruses. We observed a striking shift in Drosha isoform expression following infection with multiple SARS-CoV-2 variants. This shift was absent following treatment with the viral mimetic poly(I:C) or infection with other RNA viruses, including the non-severe coronaviruses HCoV-OC43 and HCoV-229E. We also identified a distinct alteration in Drosha’s cellular localization post SARS-CoV-2 infection. Moreover, Drosha ablation led to reduced expression of SARS-CoV-2 genomic and sub-genomic targets.
While the cytoplasmic localization and presence of alternative Drosha isoforms have been well characterized under various cellular conditions, their relevance during SARS-CoV-2 infection remains unexplored. Upon infection, Drosha undergoes proteolytic cleavage and translocates to the cytoplasm. Functionally, this cytoplasmic shift appears to be significant, as Drosha depletion results in reduced expression of several key SARS-CoV-2 genomic and sub-genomic targets. This dissertation not only elucidates a novel aspect of Drosha’s antiviral role but also advances our understanding of SARS-CoV-2 host–pathogen interactions, details potential therapeutic avenues for future human coronavirus infections
Molecular Characterization of Fungus-Arthropod Associations
Get PDFArthropods, or phylum Arthropoda, are the largest group of animals, accounting for over three quarters of all animal records on NCBI Taxonomy. And yet, only 40% of all deposited sequences on NCBI Nucleotide come from arthropods. Clearly, arthropods are an understudied group of animals: they are small, often drab, and lead lives very alien to our own. Arthropods have chitinous exoskeletons, segmented bodies, and jointed appendages, and include crabs, barnacles, insects, spiders, and millipedes. Since the era of DNA based studies took off in the 1980s, DNA and other small molecules have been integral to our understanding of the Earth and the organisms we share it with. Knowledge of evolutionary histories is driven by the number and quality of features examined, and using DNA or other molecular tools provides exponentially more information than even hundreds of physical traits. This knowledge has improved our understading of both cicadas and millipedes, the two arthropods discussed in this dissertation, but there is still much to learn about the even more neglected microbes that form relationships with them. Fungi, or mushrooms, are familiar to most people, but there is far more diversity in this kingdom of life than meets the eye. Not all fungi are saprotrophic, however: many interact with other living things as parasites, pathogens, or mutualists. In this dissertation, we examine the relationships between cicadas and their fungal pathogen Massospora, and the relationships between Colobognath millipedes and their broader community of fungal associates. In Chapter 1, we review the literature on fungi and their symbioses, and introduce cicadas, Massospora, and the Colobognatha. In Chapter 2, we examine the relationships among Massospora species, finding that of the 5 we could find, two of them were actually the same organism. We also provide detailed measurements of many features for these fungi, several of them for the first time. In Chapter 3, we explore the mycobiome of the Colobognath millipede Brachycybe lecontii, providing one of the first mycobiomes for all millipedes. This species alone associates with at least 620 genera of fungi in 9 phyla, and may have symbiotic relationships with a few of them. Their fungal diversity is only eclipsed by the diversity found in Chapter 4, where we sampled fungi from over twenty species of Colobognath millipedes from the United States. Among these millipedes, we recovered over 800 genera of fungi, and here too, we found possible evidence for fungal symbioses. These remarkable discoveries will fuel further studies that may lead to medically important chemical compounds, or even compounds that can be used to repel pests and manipulate pest behaviors. In addition, this work can fuel further studies of these organisms in their own right, which is critically important in a changing world
White Paper and Black Water: Environmental Control of Labor in Appalachian Paper Mill Communities
Get PDFIn several small Appalachian communities, the West Virginia Pulp & Paper Company dominated the local landscape. The company stripped the mountainsides of spruce trees for pulpwood and used rivers for industrial processes and waste disposal. The small towns along these rivers and between these mountains relied on paper mills for employment. Corporate control over the natural environment, and water more specifically, was a necessary requirement for paper production, but also acted as a source of power over employees and their communities. When paper workers sought to organize, the company responded by leveraging their control over the environment against them to prevent unionization. In Davis, West Virginia, a prolonged strike led to the company pulling out, taking the community’s water supply with them. In Luke and Westernport, Maryland, and Piedmont, West Virginia, company officials reframed their environmental control to portray themselves as environmentally conscious, in hopes that employees would vote for the company union. After unions were established, the company still took on ecological projects to foster goodwill, as they did in Covington, Virginia, all while threatening to leave if seriously threatened by regulation. Using union, company, and newspaper records as well as oral history sources, this thesis seeks to examine how paper companies solidified this environmental power as well as how workers were affected by and responded to that power dynamic over the twentieth century in Appalachia
Cultural Immersion and Practices: Ellie Mannette’s Steel Pan Building Processes in Morgantown, West Virginia.
Get PDFEllie Mannette, among others, was one of the first builders of the modern steel pan. Born and raised in Trinidad, Mannette started building steel pans at a young age. Eventually, he gained employment with the United States Naval Steel Band, which opened opportunities for him to grow in his field. Through these opportunities and connections, Mannette arrived in the United States permanently in 1967. In the later part of his life, he moved to Morgantown, West Virginia, where he continued building, teaching his building processes, and tuning steel pans. The study aims to serve as an aid for other scholars who are filling in similar gaps related to early steel pan builders. The primary method of research is grounded in ethnographic field work. Specifically, a vast majority of the information was gathered through interviews, video recordings, and photographs. This investigation sheds light on how cultural and locational factors influenced Ellie Mannette’s building processes and examines the development of those building processes. Cultural and locational factors influence a vast majority of topics similar to this one and understanding how those factors affect the development of history is an important element to focus on through this document. Broadly, this study examines pan-building processes within the United States through the lens of Ellie Mannette. More specifically, this research follows Mannette’s time from Trinidad to Morgantown, West Virginia, investigating the socio-cultural and environmental dynamics at play as he navigated building steel pans within a new cultural environment
La2NiO4+δ-based Solid Oxide Electrolysis Cell (SOECs) Electrodes Enhanced with Complex Perovskite Nanocatalyst Processed by Surfactant-Enabled Infiltration
No full textAs the world seeks to reduce its reliance on hydrocarbons, the demand for sustainable hydrogen production methods has become increasingly critical. Hydrogen is a versatile energy carrier that can be produced from various sources, including water, natural gas, and biomass. Replacing the burning of hydrocarbons with hydrogen could significantly reduce greenhouse gas emissions, contributing to global climate goals like those set by the US H2NEW program. The joint technologies of solid oxide fuel and electrolysis cells present the ability to both produce and process clean hydrogen to meet these goals. Development into solid oxide fuel cells has been met with success, with several on-market fuel cell stacks having been developed and utilized. However, the development of solid oxide electrolysis cells has lagged, leaving a critical gap in the production of clean hydrogen gas. Though SOECs offer several advantages over traditional low-temperature electrolysis methods, including higher efficiency and the potential for integration with renewable energy sources and industrial processes, the technology faces challenges. The presence of steam and the common use of stainless-steel interconnects results in the known issue of Cr migration, where high-temperature gaseous chromium hydroxide’s high vapor pressure enables it to react with many components within the cells. This causes insulative solid phases and altered electrode chemistries, as well as blockage of active sites and Cr migration to the electrode-electrolyte interface, ultimately leading to delamination of the cells and long-term degradation of the system. This issue and others result in the need for chromium-resistant electrode materials that must have strong electrochemical properties. Commonly used SOFC materials have been shown to struggle with chromium poisoning in part due to strontium content, while newer developed SOEC materials may have suboptimal electrochemical properties.
In this work, the process of surfactant-enhanced wet impregnation of nanoparticulate oxides was studied as a potential nano-coating method for the SOEC material La2NiO4+δ. The overarching goal of the research was to improve the surface oxygen exchange coefficient by addition of an ionically conductive oxide surface layer, with the effect being the improvement of the overall electrochemical performance of the cell. Three nanocoating compositions were studied, including ternary oxide LaCoO3 (LCO), quaternary oxide La2MnNiO6 (LMNO), and high-entropy particle (HEP) La0.2Sr0.2Pr0.2Y0.2Ba0.2Co0.2Fe0.8 (LSPYB). Initially, the work studied the LCO composition, and the information from these experiments was then applied to the infiltration of the chosen quaternary and HEP compositions.
Investigation into the LCO composition as a coating candidate began with the selection of a variety of surfactant molecules as chelating agents for the infiltration process. A powder study was completed to determine the most suitable candidate for chelation of the oxides in aqueous solution and enabling correct phase formation when calcined at temperatures ≤900 °C. This low temperature was chosen to maintain the high surface area of the nanocoating by reducing nanoparticle densification, a property that contributes to more efficient oxygen exchange properties. The molecules studied were catechol and catechol-like chemistries, including poly-norepinephrine (pNE), caffeic acid, dihydroxybenzoic acid (DHBA), and gallic acid. Previous work from Ozmen et al. and Wang et al. showed the success of bio-inspired catechols such as pNE, caffeic acid, and DOPA to deposit single-component nanoparticles [1], [2]. This work expands on this premise to include both new catechol-family molecules as chelating agents, and to attempt deposition of more complex oxides (≥3 different cations within the perovskite). For the purposes of nanocoating, a successful chelating agent will be defined to serve three purposes: 1) act as a
complexing agent and in the formation of complex, high-order oxides, 2) assist in the deposition of these chemistries in distinct, nano-range structures, and 3) assist in controlling the homogeneity of these coatings, with optimal coverage statistics.
Initial experiments focused on the formation of the binary LCO composition. Complexing properties of the chosen chelating agents were analyzed using X-ray diffractometry (XRD) to determine the impact of the molecules on the formation phase pure nanoparticles. LCO nanoparticles were deposited on highly polished and flat single-crystal YSZ substrates, then analyzed by atomic force microscopy (AFM) to determine the impact of deposition parameters on the nucleation and growth rate of the samples and the final microstructure of the nano-coatings. These samples were then evaluated using X-ray photoelectron spectroscopy (XPS) for a second level of validation for the results of the phase-purity testing by analysis of binding energies. Results from these experiments were implemented on the heterostructured LNO – LNO/GDC porous electrode microstructures to determine their impact on the performance of the coated cell. The infiltrated LNO microstructures were studied by scanning electron microscopy (SEM) and electron diffractometry spectroscopy (EDX), and the polarization resistance of the LCO-coated LNO was evaluated by symmetrical cell electrochemical impedance spectroscopy (EIS) testing and compared to baseline LNO performance.
Using the methods and understanding gleaned from the LCO nano-deposition study, selected experiments were completed for higher complexity solid solution nano-catalysts. For both LMNO and LSPYB, a representative powder study (with samples analyzed using XRD) was completed with each of the previously attempted chelating agents to determine any changes in effectiveness, after which infiltration into symmetrical LNO cells was completed. LMNO samples were analyzed using EIS and SEM to determine effectiveness of nanocoating. The same methods and understanding were utilized to successfully deposit the more complex high-entropy perovskite (HEP) LSPYB. Despite success in ex-situ analysis, low pH as a complication from Fe-nitrate in solution resulted in destruction of the LNO backbone. Alternate infiltration procedures and materials were investigated and implemented, resulting in a high phase purity nano-coating that provided electrochemical improvements at lower infiltration concentration and deposition times.
From all tested nanocoatings, all three showed at least some performance improvements, with the most successful LCO nano-catalyst improving the performance across all infiltration times and concentrations
Customer Segmentation and Fuel Economy Prediction using Telemetry Data from Heavy-Duty Trucks
No full textHeavy-duty trucks constitute only a modest fraction of on-road vehicles, yet their intensive duty cycles and high fuel demands yield a disproportionately large share of transportation fuel use and greenhouse gas emissions. Addressing this imbalance requires data-driven tools that capture the realities of fleet operation and translate complex telemetry into actionable insight.
This dissertation introduces a unified machine-learning framework that operates exclusively on high resolution time-series data collected from fifty-nine diesel trucks deployed across Southern California. It begins by constructing a multi-modal feature space that blends statistical summaries of key engine signals, static vehicle descriptors, and Mel-Frequency Cepstral Coefficients, thereby providing a compact yet expressive encoding of temporal and structural behavior. An unsupervised clustering pipeline based on Ward-linkage agglomerative clustering then partitions the fleet into semantically coherent operational segments, revealing groupings such as long-haul freight, urban delivery, and construction service trucks.
Within these segments, the study investigates several predictive architectures, including feed-forward neural networks, Long Short-Term Memory models, and a hybrid design that couples LSTM-derived temporal embeddings with Gradient-Boosting regression. Empirical evaluation shows that aligning model training with the discovered behavioral clusters substantially improves fuel economy prediction accuracy, particularly under high-variance duty cycles, while preserving the interpretability required for operational decision making. Beyond methodological innovation, the proposed framework supports practical benefits: fuel-use forecasting, scheduling and maintenance planning, and establishes a scalable foundation for anomaly detection and emissions reporting as environmental regulations tighten. Collectively, the work advances intelligent transportation analytics by bridging complex real-world telemetry and sustainable fleet management, charting an interpretable and extensible path toward lower emissions and higher operational efficiency in heavy-duty trucks
Distress Tolerance and Sleep Quality: Potential Moderating and Mediating Factors
Get PDFResearch has shown that insomnia and poor sleep quality are related to a variety of negative physical and mental health outcomes. There is emerging evidence that the psychological construct of distress tolerance is related to the quality of one’s sleep, including the presence of insomnia. However, few studies have explored how additional factors may be related to this association. This study aimed to replicate previous research supporting the association between distress tolerance and sleep and exploring this relation by examining three different hypotheses (e.g., health behaviors, perceived stress, and emotion regulation) that could potentially explain how these two variables are related.
A sample of West Virginia University college students (N = 226) and a sample of adult participants recruited from Amazon Mechanical Turk (MTurk) (N = 186) participated in this research. A variety of multi-method screeners were implemented to eliminate AI-generated responses from being included in these samples. These participants completed a set of demographic questions, the Distress Tolerance Scale (DTS), the Pittsburgh Sleep Quality Index (PSQI), Insomnia Severity Index (ISI), Tobacco, Alcohol, Prescription Medication, and other Substance Use (TAPS) Tool: Part 1, The Leisure Time Exercise Questionnaire (LTEQ), Perceived Stress Scale (PSS), and the Emotion Regulation Questionnaire (ERQ). A series of linear regression analyses were conducted to examine the relation between distress tolerance and sleep outcomes (e.g., sleep quality and insomnia severity). For health behaviors only, mediation analyses were run to examine if physical activity, tobacco use, alcohol use, or other illicit drug use explained the association between distress tolerance and sleep outcomes. Additionally, a series of moderation analyses were conducted to examine these health behaviors, as well as perceived stress and emotion regulation (e.g., cognitive reappraisal and expressive suppression), as moderators of the distress tolerance – sleep outcomes association.
Results from this study confirmed that lower distress tolerance was associated with poor sleep outcomes and that this relation was diminished among college students with high levels of physical activity engagement. Alternatively, the association between distress tolerance and sleep outcomes became stronger as levels of physical activity increased among the MTurk sample. Regardless of DTS levels, MTurk participants engaging in higher levels of physical activity exhibited decreased sleep quality scores than those with lower physical activity. Perceived stress was only found to be a moderator for the college student sample, and results showed that the distress tolerance – insomnia severity relation was only significant for individuals with students with high levels of perceived stress. Lastly, one emotion regulation strategy, cognitive reappraisal, moderated the association between distress tolerance and sleep outcomes among college students but not the MTurk sample wherein it only moderated the association at low and moderate levels but not at high levels of cognitive reappraisal. This research highlights the importance of examining psychological and behavioral factors that help explain the association between distress tolerance and sleep outcomes
Authoritarian Control Over Election Administration
Get PDFStreet-level election bureaucracy (SLEB) remains one of the least studied aspects of authoritarian elections. Our current understanding of micro-level election administration in autocracies is fragmented and unsystematic. In this dissertation, I develop and test a theory of authoritarian SLEB that comprehensively addresses key elements of election administration: the incentivization and mobilization of election commissioners, the establishment of control over election commissions, the monitoring of their performance, and the post-election reshaping of their personnel composition. Drawing on rational choice theory, I propose a framework in which autocrats are seen as outcome-maximizers who strategically design SLEB to achieve their electoral goals. The empirical tests and their results support the idea that autocrats act strategically in shaping SLEB and consolidating control over it. To illustrate this, I examine both modern Russian elections and elections in the Soviet Union, showing how authoritarian regimes adapt their SLEB strategies to varying structural conditions and constraints. The analysis is based on several sources: polling data from Russian election experts and members of precinct election commissions (PECs); occupational and socio-demographic data on PEC members in selected regions of modern Russia and the USSR; archival materials on Soviet election administration practices; and data on PEC compositions before and after the 2018 Russian presidential election, combined with precinct-level election results. The key findings demonstrate that the Russian authoritarian regime relies more on positive than negative incentives in its design of SLEB. It mobilizes diverse professional and social groups, adapts its strategies based on the availability of personnel, and strategically reshapes PEC compositions according to their electoral utility. Overall, this dissertation offers a comprehensive and alternative perspective on street-level election bureaucracy in authoritarian regimes – one in which autocrats are viewed as clients seeking contractors – the election commissioners – who voluntarily participate in election administration and contribute to the regime’s electoral success in exchange for positive incentives