Bioculture Journal
Not a member yet
1309 research outputs found
Sort by
Simulation Informed Machine Learning Interpretation of Electrochemical Measurements
No full textThesis (Ph.D.)--University of Washington, 2023This work resides at the intersection of rigorous physics-based simulation and machine learning. We seek to address problems that have complicated multiphysics and exist in vast parameter spaces. Where the data requirement for machine learning methods can make the experimental burden untenable and the time-sensitive nature or high computational cost limits the practicality of mechanistic physics model. Our overarching objective is to explore solutions to bridge these limitations in electrochemistry research through the integration of machine learning and mechanistic simulation.
Failure detection in solid oxide fuel cell (SOFC) is complicated due to the need to disentangle the failure response from the effect of degradation - gradual change in performance with aging. We used physics models to simulate the behavior of SOFC under three failures that could occur during its operation: fuel maldistribution, delamination and oxidant gas crossover. These simulations revealed deviations in electrochemical impedance spectroscopy (EIS) from behavior of standard circuit elements under failures, underscoring the significance of physics-based modeling in SOFC diagnostics. Leveraging synthetic data of a 6-cell sub-stack, we trained a support vector machine to identify failure modes with a 90% accuracy across degradation effects and operating conditions, discerning imperceptible differences in stack-level EIS responses. Investigation of synthetic data offered insights to failure diagnosis with EIS in determining most responsive frequency range and the efficacy of different machine learning methods.
In the second project, we reexamined the utility of a reference electrode positioned outside the current path on a thin solid electrolyte. Extensive prior research with this design had demonstrated significant polarization shifts and half-cell EIS distortions from minor electrode misalignment, limiting its usefulness in quantitative assessment of two half-reactions. Employing a physics model to simulate these behaviors in a proton-exchange membrane electrolyzer, we trained a neural network model with simulated data to deconvolute edge effects and determine the true oxygen kinetic overpotential with high accuracy (r2-score ≥ 0.96). Validation with experimental data from electrolyzer cells of varying membrane thicknesses and misaligned electrodes confirmed the breakdown of oxygen and hydrogen evolution reaction losses to align with literature values. These findings unveil the potential use of this straightforward reference electrode design and intentional anode-cathode misalignment for evaluating individual electrode kinetics in performance or long-term degradation studies
The Value of Teacher-Student Relationships: A Narrative Analysis of Teacher Evaluations in Washington State
No full textThesis (Master's)--University of Washington, 2023National and state initiatives to improve student academic outcomes in K-12 public schools have included efforts to reform teacher evaluation policies and practices. Prior research into teacher evaluation systems raises questions about how to accurately measure effective teaching practices. This study takes up these questions and uses Nel Noddings’ (2005) concept of an “ethic of care” to identify the limitations of a variety of metrics used in research on teacher-student relationships. This study examines the three instructional frameworks used in Washington State through the lens of narrative analysis and finds that the language addressing teacher-student relationships neglects the unique context of individual classrooms and presents the teacher-student relationship as one of managing classroom procedures and student behavior. This study argues that without explicitly incorporating teacher-student relationships into these policies and frameworks, not only is this aspect of effective teaching left to individual interpretation, but teacher evaluation systems neglect the vital role that teachers play in supporting students’ social, emotional, and academic growth
Model-driven DBTL cycle acceleration with broad-host-range bacterial CRISPRa/i circuits
No full textThesis (Ph.D.)--University of Washington, 2023CRISPR-Cas gene regulatory tools have revolutionized biological network programming. Recently, we developed CRISPR gene activation (CRISPRa) tools that demonstrate broad applicability across various bacteria. With thorough characterization of bacterial CRISPRa, we found that the design rules for effective CRISPRa are stringent and highly context-dependent. Thus, we developed numerous strategies to overcome existing limitations — including DNA context engineering, utilization of engineered proteins to bypass target sites requirements, and characterization of multiple bacterial CRISPRa systems for alternative design rules. Implementation of CRISPRa tools in chemical bioproduction enabled synthesis of p-aminocinnamic acid, a precursor to various functional polymer materials, which was previously difficult to synthesize through conventional routes. In combination with CRISPR gene interference (CRISPRi), we further explore the capability of a combined CRISPRa/i platform to regulate expression of both foreign genes and native genes intricately involved in bacterial metabolism. Since programmability of CRISPRa/i relies on guide RNA sequence, we found that multiple guide RNAs could be implemented simultaneously to regulate multiple genes in the same system. Engineering at the RNA level could also provide tunable regulation for each gene target. Furthermore, when combined with genome-scale metabolic models, this system accelerates the Design-Build-Test-Learn (DBTL) process for microbial strain optimization, bypassing stepwise genetic reconstruction through the use of trans-acting CRISPRa/i circuits. The resulting constructs can be comprehensively investigated using a multi-omics platform, providing detailed information to improve subsequent DBTL cycles. By coupling programmable and tunable gene regulatory tools with large metabolic models informed by omics data, our platform establishes a foundation for non-canonical microbial strain engineering that benefits diverse disciplines from industrial biotechnology to therapeutic discovery
On the Development of a Robotic Biarticular Prosthesis Emulator
No full textThesis (Ph.D.)--University of Washington, 2023People with transtibial limb loss experience decreased walking ability and have long-term musculoskeletal issues, such as knee osteoarthritis. Currently available ankle-foot prostheses, which only actuate the ankle joint, do not replicate the biarticular nature of the gastrocnemius muscle, which may limit their clinical efficacy. It is possible that transtibial prosthesis designs that include knee flexion assistance through a knee exoskeleton or exosuit could improve gait and quality of life for people with transtibial limb loss. However, the design space of these devices is large, and it is difficult to determine how biarticular gait assistance from different device designs might help people with limb loss. Therefore, developing a method to rapidly test and compare a variety of biarticular prosthesis designs is crucial for improving the lives of people with transtibial limb loss. The objectives of this work were to design, build, and evaluate a biarticular prostheses emulator – a device that combines a robotic ankle-foot prosthesis and knee exoskeleton that can behave like a variety of hypothetical device embodiments to facilitate scientific exploration in a laboratory environment. This work was separated into three studies. The first study detailed the design, control, and evaluation of a uniarticular robotic ankle-foot prosthesis with offboard actuation and control. The prosthesis design incorporated novel design choices compared to other prosthesis emulators, including a commercial foot plate and parallel torsion spring to provide dorsiflexion moments during walking. The prosthesis was found to be able to emulate both idealized passive devices and powered devices in walking experiments. The second study explored the design and control of a robotic knee exoskeleton emulator specifically for people with transtibial amputation. The third and final study presented the design and evaluation of an integrated biarticular prosthesis emulator that combined the robotic ankle-foot prosthesis with a second-generation design of the robotic knee exoskeleton. The biarticular prosthesis emulator design and control were evaluated with benchtop experiments and a walking demonstration that showed the prosthesis emulator’s ability to provide simultaneous knee and ankle assistance during walking. The primary contributions of this work are the design and development of several novel assistive devices for people with transtibial limb loss. Due to our modular design approach, future research may use either the ankle-foot prosthesis or knee exoskeleton as stand-alone uniarticular assistive devices, or together as a biarticular prosthesis. The work in this dissertation lays the foundation for a future research program that uses these devices to explore biarticular gait assistance and compare different hardware embodiment for prostheses that improve health and quality of life for people with transtibial limb loss. The potential impact of this work is significant, as the development of more effective prostheses for people with transtibial limb loss can significantly improve their mobility and overall quality of life
Trajectories of Maternal Behavior, Infant’s RSA, and Emotional Recovery Following the Still Face Paradigm
No full textThesis (Master's)--University of Washington, 2023The development of emotion regulation is integral to children’s socioemotional adjustment. Respiratory Sinus Arrythmia (RSA) reflects parasympathetic regulation of cardiac arousal and is an indicator of emotion regulation. However, it is unclear how changes in RSA are associated with positive maternal behaviors and infant emotional recovery throughout the reunion following a social stressor. Using a series of autoregressive latent trajectory models, the current study aimed to elucidate the associations among maternal warmth, infant RSA, and infant negative affect across a 5-minute observation. Mothers and their 5-6 months old infants (N = 143) completed the Still Face Paradigm. Cross-lagged effects indicated that increases in RSA precede decreases in NA during the first 2.5 minutes of the reunion. Change in maternal warmth was associated with change in RSA partially supporting our hypothesis that maternal warmth supports infant physiological, and in turn, emotional recovery
Understanding methamphetamine and opioid co-use: national trends and local harm reduction strategies for overlapping illicit drug use
No full textThesis (Ph.D.)--University of Washington, 2023Introduction: In the past 20 years, the United States has seen a remarkable increase the use of both methamphetamine and opioids, used concurrently or simultaneously. A more detailed picture is needed to understand where methamphetamine-opioid co-use is increasing nationally, as well as the characteristics of people who co-use. Methamphetamine-opioid co-use has become particularly prevalent in Seattle, WA, yet the rationale for co-use among people who use drugs in this important region is not yet clear. More harm reduction tools to serve the growing number of people who co-use in Seattle are needed. Methods: First, we used data from the 2012, 2015, and 2018 cycles of the National HIV Behavioral Surveillance (NHBS) project in people who inject drugs (PWID) to describe trends in methamphetamine-opioid co-use over time and in different Census regions. We also compared the demographic, socio-economic, sexual health, and drug use behavioral characteristics of people who co-used compared to people who primarily used one drug (Chapter 1). Second, we conducted in-depth semi-structured interviews with people who regularly used both methamphetamine and opioids recruited from a syringe services program (SSP) in downtown Seattle (N=21). We conducted an interpretive descriptive analysis of the data informed by the social-ecological framework to identify themes in the rationale behind methamphetamine-opioid co-use for our participants (Chapter 2). And last, we evaluated access to and interest among Seattle-area PWID in a potential harm reduction strategy to promote safer consumption by facilitating a switch from injection to safer routes such as smoking or oral consumption with free safer smoking equipment. Using data from the Seattle 2018 NHBS survey of people who inject drugs (N=555), we described whether respondents had access to safer smoking equipment, whether they were interested in getting it, and if they thought access did or would reduce their injection frequency (Chapter 3). Results: In the national data, we found that methamphetamine-opioid co-use increased from 14.0% in 2012 to 26.3% in 2018 in the overall NHBS sample. Co-use was most prevalent in the West and increased the most in the Northeast. Younger age, opioid overdose in the past year, sharing syringes, and sharing other injection equipment were significantly associated with methamphetamine-opioid co-use compared to all other drug use patterns. In our qualitative study, we identified two overarching themes in the rationale of methamphetamine-opioid co-use: availability and function. For many, methamphetamine and opioids were readily available in their social networks, in community sources, and through the fluctuating illicit drug market. Methamphetamine and opioids served a number of functional uses individually and in families and communities. We also identified that houselessness was an environment in which the availability and function of methamphetamine and opioids were uniquely elevated. And last, we found that among Seattle-area NHBS-PWID participants, just 12% reported access to free safer smoking equipment. Between one third and half of respondents were interested in getting free safer smoking equipment, depending on the drug. A large number of participants reported that access did or would reduce their frequency of injection. Conclusions: The widespread change in drug use patterns and the higher-risk behavior associated with co-use nationally signal the need for swift, coordinated public health action to expand harm reduction and treatment services and to develop data-informed clinical guidelines to serve this growing population. Locally, methamphetamine-opioid co-use was influenced by complex personal, social, and societal factors. Public health policy to address the needs of people who co-use through treatment, harm reduction, and other social programs must support individuals, their communities, and the broader structural environment. Harm reduction strategies like provision of free safer smoking equipment may be an important tool to reduce risks from opioid and stimulant injection
Biomimetic Active Chemical Separation
No full textThesis (Ph.D.)--University of Washington, 2023This dissertation embarks on an exploration of nature's design principles in the realm of chemical separation and transfer, with the overarching goal of applying these insights to address pressing real-world challenges. Our journey begins with a comprehensive review of prevailing industrial chemical separation techniques, revealing their remarkable energy demands and the unfortunate squandering of unrecoverable energy. This realization serves as a catalyst for our quest to draw inspiration from the intricate, yet efficient solutions nature offers, particularly exemplified by the cell membrane's remarkable ability to discern among myriad compounds using a seemingly straightforward design.The cell membrane in living organisms serves as an emblem of nature's ingenuity. Equipped with delicate membrane proteins, each devoted to specific functions vital for the cell's existence, it operates with a sophistication that continues to elude our full understanding. Yet, our quest is not to replicate nature but to draw inspiration from it. We take the humble dialysis machine as an exemplary case, a technology characterized by its relatively simple design. This apparatus, functioning as an artificial kidney, relies on precision-controlled transmembrane separation facilitated by hollow fiber membranes with precise pore sizes, complemented by mechanical pumps and dialysate. Despite its apparent simplicity, this technology remains a lifeline for patients awaiting kidney transplants. Nevertheless, we acknowledge the imperative to enhance its long-term survival rates.
From a materials science perspective, our pursuit of advancement in these technologies hinges on the development of superior functional materials. For dialysis, this translates into better control of pore size distribution and porosity, a journey we embark upon through the development of hollow fiber membranes. However, the intrinsic limitation of passive diffusion in dialysis technology necessitates a shift towards biomimetic active high-selectivity functionalization. In response, we design, propose, synthesize, and validate selective surface chemistry modifications for serum albumin. This groundbreaking surface chemistry enhancement facilitates the adsorption and desorption cycling of albumin, concurrently removing protein-bound uremic toxins that defy conventional dialysis processes. Moreover, our journey extends to the scaling-up of this technology, underpinned by systematic design implementation and continuous refinements guided by computational models.
Likewise, in our pursuit of dialysis technology enhancement, we explore the modification of highly selective functional groups onto low-cost silica adsorbents. This endeavor aims to efficiently remove potassium ions enriched in dialysates during the development of mobile dialysis techniques. Despite the availability of various adsorbents and natural materials, we opt for silica adsorbents due to their stability, cost-effectiveness, and ease of preparation. The modified crown ether-silica adsorbents exhibit remarkable potassium ion adsorption capacity, promising applications beyond conventional adsorbents.
Our expedition also ventures into the domain of carbon material-based membranes, particularly carbon nanotubes and reduced graphene oxide membranes. These materials hold the potential to revolutionize separation processes by compensating for membrane structure surface area limitations. Under nanoscale confinement, they achieve high-speed transmembrane fluid transport, outperforming conventional adsorbent-based methods. We make history by directly observing high-speed transmembrane electroosmotic transport within carbon nanotubes, marking a significant milestone in our journey.
The observations from this study underscore the promise of carbon nanotube membranes, especially when functionalized with high-selectivity functional groups. We validate the concept of transmembrane drug delivery on negatively modified carbon nanotube membranes, showcasing their potential in various applications.
In our exploration of graphene oxide membranes, we unearth a fascinating catalytic reaction, offering unique selectivity. However, this discovery comes with the challenge of reduced interlayer spacing, limiting transmembrane separation rates. We surmount this challenge by ingeniously reintroducing substances that reverse the catalytic reaction, enabling high-speed fluid passage and salvaging the potential of graphene oxide membranes in alcohol dehydration applications.
In conclusion, this dissertation's comprehensive exploration of materials science and biomimetic high-selectivity surface chemistry modification yields profound insights. Our research validates scientific theories and overcomes engineering constraints, heralding a promising future for separation technologies. As we continue to draw inspiration from nature's design, the possibilities for addressing real-world challenges through biomimicry are vast and exciting
A Multi-Domain Trojan Detector for Deep Neural Networks
No full textThesis (Master's)--University of Washington, 2023Backdoor attacks have been demonstrated to compromise the functioning of machine learning models that utilize deep neural networks (DNNs). An adversary carrying out a backdoor attack embeds a predefined perturbation called a Trojan trigger into a small subset of input samples. The DNN can then be trained in a manner such that the presence of the trigger in the input results in an output label that is different from the correct label. At the same time, outputs of the DNN corresponding to inputs without the trigger remain unaffected. Backdoor attacks, where an attacker can negatively affect the DNN's behavior, might have severe repercussions in safety-critical applications. Existing defenses in the literature against backdoor attacks involve pruning or retraining DNN models, which can be computationally expensive. In addition, researchers have demonstrated the success of these solutions on input domains based on images. The performance of such defenses on other inputs needs to be understood better. In this thesis, we propose and develop MDTD, a multi-domain Trojan detector. MDTD for DNNs has several distinguishing characteristics, including (i) not requiring retraining DNN models (ii) not requiring knowledge of the trigger or the embedding strategy of the attacker, (iii) is computationally inexpensive (iv) capable of being applied to image and graph-based inputs. To the best of our knowledge, MDTD is the first Trojan detection mechanism proposed for graph-based inputs. MDTD uses the insight that input samples containing a Trojan trigger are located relatively further away from a decision boundary than clean input samples. Initially, MDTD estimates the distance to a decision boundary using adversarial learning methods. These methods estimate the smallest magnitude of noise required for the model to misclassify a sample. MDTD uses this information to infer whether a given sample is Trojaned or not. More precisely MDTD learns a threshold for the distance to the decision boundary using a small set of clean labeled samples and uses this threshold to flag a sample as possibly Trojaned. We evaluate MDTD against state-of-the-art (SOTA) Trojan detection methods across five image-based datasets - CIFAR100, CIFAR10, GTSRB, SVHN and Flowers102- and four graph-based datasets - AIDS, WinMal, Toxicant and COLLAB. Our results show that MDTD effectively identifies samples that contain different types of Trojan triggers. We also show that an adversary who trains robust DNN models using a combination of clean and Trojaned samples does not cause a significant deterioration in MDTD performance without significantly reducing the classification accuracy of the DNN model
The effect of Temperament on Outcomes of Opioid and Non-Opioid Pediatric Dental Sedation
No full textThesis (Master's)--University of Washington, 2023Background: Procedural sedation is frequently used to safely and effectively complete pediatric dental treatment. However, there is no standard regimen or patient assessment used among pediatric dentists and sedation outcomes vary widely. Purpose: The primary objective of this randomized trial was to assess the effects of oral sedation using midazolam and hydroxyzine with and without meperidine on sedation outcomes in pediatric dental patients. The secondary objective was to assess the relationship between child temperament and sedation outcomes.
Methods: This pilot study recruited 37 children between the ages of 3-7 years who met study eligibility criteria and were planned to undergo dental treatment with oral sedation at the University of Washington Center for Pediatric Dentistry. The children were randomly assigned to receive a regimen of midazolam and hydroxyzine with or without meperidine. Parents completed the Child Behavior Questionnaire Short Form (CBQ-SF) to assess temperament.
Results: There were no significant differences in sedation outcome with age, sex, insurance status, sedation regimen, isolation method, or duration of procedure. In general, children with high pre-operative Frankl behavioral ratings were more likely to have successful sedation outcomes. Children who displayed high soothability experienced higher rates of success, and this effect was more pronounced in the non-opioid regimen group.
Conclusions: Overall, there was a low rate of success in this study and a relatively small sample size. However, the results suggest that pre-procedure behavior and the temperament characteristic of soothability may warrant more exploration as predictors of sedation success
The Design, Characterization, and Deployment of a Bipolar ±0.5-mV-Minimum-Input DC-DC Converter with Stepwise Adiabatic Gate-Drive and Efficient Timing Control for Thermoelectric Energy Harvesting
No full textThesis (Ph.D.)--University of Washington, 2023This work presents a step-up DC-DC converter that is optimized to extract power from thermoelectric generators that generate extremely low voltages from small temperature differentials. The DC-DC converter uses a stepwise gate-drive technique to reduce the power FET gate-drive energy by 82%, allowing positive efficiency down to an input voltage of ±0.5 mV—the lowest input voltage ever achieved for a DC-DC converter as far as the author knows. Below ±0.5 mV the converter automatically hibernates, reducing quiescent power consumption to just 255 pW. The converter has an efficiency of 63% at ±1 mV and 84% at ±6 mV. The input impedance is programmable from 1 Ω to 600 Ω to achieve maximum power extraction. A novel delay line circuit controls the stepwise gate-drive timing, programmable input impedance, and hibernation behavior. Bipolar input voltage is supported by using a flyback converter topology with two secondary windings. A generated power good signal enables the load when the output voltage has charged above 2.7 V and disables when the output voltage has discharged below 2.5 V. The DC-DC converter was used in a thermoelectric energy harvesting system that effectively harvests energy from small indoor temperature fluctuations of less than 1°C. To aid in efficiency optimization, an analytical model with unprecedented accuracy of the stepwise gate-driver energy consumption was developed. A test fixture that can accurately measure the efficiency, input impedance, and quiescent power consumption of the DC-DC converter was designed and utilized. Lastly, a method for characterizing the leakage current of capacitors was developed so that capacitors of the lowest leakage can be selected for the DC-DC converter output energy storage