University of Illinois at Chicago
University of Illinois at Chicago: UIC INDIGO (INtellectual property in DIGital form available online in an Open environment)Not a member yet
21439 research outputs found
Sort by
Study of Functional Complex Oxides at High Pressure-Temperature Conditions
No full textComplex oxides are a diverse family of
functional materials whose stability and
transformations under extreme conditions are central
to chemistry, energy research, and the design of
novel compounds. This dissertation focuses on
lithium cobalt oxyfluoride (Li2CoO2F), a Li-rich
disordered rock-salt cathode, together with other
representative complex oxides studied under high
pressures and temperatures.
Synchrotron X-ray diffraction, Raman spectroscopy,
neutron scattering, and first-principles calculations
were used to probe the structural response of
Li2CoO2F across multiple length scales. These
studies reveal how fluorination and Li enrichment
promote local spinel-like motifs, stabilize metastable
frameworks, and modify the elastic behavior under
compression and heating. Complementary work on
PbTiO3 demonstrates how kinetic barriers and
thermodynamic driving forces combine to stabilize
unexpected dissociation products, including a
previously unreported PbO polymorph at megabar
pressures. High-entropy oxide nanoribbons were
also examined, with high-pressure experiments
showing transformations from orthorhombic to
cubic and ultimately amorphous states, underscoring
their structural resilience and entropic complexity.
Overall, this work provides new insight into the
structural response of Li2CoO2F and related oxides
under extreme conditions, highlighting how
fluorination and Li enrichment influence their
stability
Geochemical Constraints of Porphyry Cu Deposits-Forming Magmas in Sonora, Mexico
No full textPorphyry Cu deposits (PCDs) represent the most significant source of economic-grade copper (Cu) on Earth (Sillitoe, 2010). PCD formation is intimately linked to the magmatic and tectonic evolution of convergent margins (Sillitoe, 2010). While the metallogenic characteristics of PCDs have been well-studied, the petrological and geochemical processes that govern magma fertility, essential prerequisites for PCD formation, remain less well constrained. Magma fertility is primarily controlled by variations in pressure-temperature (P–T) conditions, redox state, and volatile content during magmatic evolution (Richards, 2003; Audétat et al., 2008). These factors influence the capacity of magmas to concentrate and transport metals such as Cu.
This thesis investigates the geochemical signatures associated with fertile and barren magmatic systems in Sonora, Mexico. Using zircon trace-element geochemistry, U–Pb geochronology, and thermobarometry, I assess the magmatic conditions under which PCDs and barren rocks formed, and identify key proxies, such as REE anomalies, H₂O content, and oxidation state, that distinguish fertile magmas from barren magmatic systems. These findings provide new constraints on the petrogenetic evolution of magmas within the CEMMA and offer insights into the broader geodynamic and metallogenic processes that control porphyry Cu mineralization
Planning and Optimization of Charging Strategies for Electric Vehicles: From Infrastructure to Behavioral
No full textThe transition to electric vehicles (EVs), including both publicly operated battery electric buses (BEBs) and privately owned EVs, is a pivotal step toward achieving sustainable and efficient urban mobility. This dissertation addresses critical dimensions of this transition, integrating advanced optimization methods with empirical behavioral research across three interconnected studies.
The first study tackles the Problem of Locating and Allocating Charging Equipment for Battery Electric Buses (PLACE-BEB), a core challenge in planning for transit electrification. Recognizing the stochastic nature of charging demand across the network, a Mixed-Integer Non-Linear Programming (MINLP) model is developed to determine the optimal siting and allocation of slow and fast chargers. The framework explicitly incorporates queueing delays through an M/M/s system, treating the number of chargers as a decision variable to account for congestion and wait times. To improve computational tractability, especially for large-scale instances, Simulated Annealing (SA) and Genetic Algorithm (GA) heuristics are designed and benchmarked. A case study involving the Chicago public transit system demonstrates the model’s effectiveness and underscores the importance of jointly considering garage and on-route charging locations. The findings highlight that minimizing waiting time, through strategic charger placement and allocation, is crucial for ensuring the economic and operational viability of BEB deployment.
Expanding upon this planning foundation, the second study proposes a comprehensive Mixed-Integer Linear Programming (MILP) framework to optimize the scheduling and charging strategies of a mixed fleet of BEBs and Diesel Buses (DBs). The model determines the optimal fleet composition, trip assignments, and partial charging strategies at both garages and terminal stations, accommodating both slow and fast chargers. A dynamic queuing mechanism is introduced to replace traditional first-come, first-served protocols, improving charger utilization and overall system efficiency.
Applied to real-world data from Chicago’s CTA and Pace systems, the model reveals that while BEBs have higher acquisition costs, they can effectively serve more trips due to their lower operational costs and optimized charging schedules. The results demonstrate that mixed-fleet electrification, supported by smart scheduling and flexible charging, can lead to substantial cost savings and improved service reliability.
While the first two studies employ optimization frameworks to resolve infrastructure and operational challenges, the third study complements these by investigating behavioral factors influencing EV adoption and usage, an essential yet often overlooked aspect of transportation electrification. Unlike the previous optimization-centric chapters, this ongoing research draws on original data from a comprehensive online survey we designed and administered, capturing public attitudes toward EVs, autonomous vehicles, and associated mobility trends. Preliminary analyses have employed Structural Equation Modeling (SEM) to explore how latent constructs such as ``Range Anxiety,'' ``Convenience,'' ``Environmentally Friendly,'' and ``Living Urban'' influence charging location choices. Ongoing efforts aim to refine and deepen this analysis by explicitly modeling ``Range Anxiety'' as a latent psychological construct using advanced methods (SEM or Hybrid Choice Modeling). This planned modeling will quantify the influence of range anxiety on EV adoption intentions and driving behaviors, providing critical behavioral insights to complement infrastructure and operational optimization.
Together, these three studies present a comprehensive framework for advancing sustainable transportation, bridging infrastructure planning, operations research, and behavioral analysis. By integrating optimization models with behavioral insights, this dissertation provides actionable tools and knowledge for transit agencies, policymakers, and urban planners aiming to accelerate electrification while ensuring reliability, cost-effectiveness, and user acceptance
Multiagent Approaches to Enhance Learning and Trust in AI Systems
No full textThis thesis addresses two central challenges in artificial intelligence: achieving scalable learning in interactive environments and ensuring trust in real-world deployment. Many domains involve multiple adaptive agents interacting under uncertainty and limited information. To be effective, agents must adapt continuously while operating efficiently in large and complex decision spaces. Using frameworks such as extensive form games and multiagent reinforcement learning, this work demonstrates how regret minimization can serve as a unifying foundation for scalable learning. Contributions include a new interpretation of an existing Multiagent Reinforcement Learning (MARL) as a regret-based method, improvements its learning through modified update rules, and the introduction of EINR, an algorithm that applies multiplicative weights in extensive form games to achieve faster and more reliable convergence.
The thesis also develops methods for trustworthy AI, where robustness, interpretability, and fairness are as important as accuracy. Building on cooperative game theory and social choice theory, it introduces Banzhaf indices to generate stable counterfactual explanations for graph neural networks and proposes voting based aggregation rules such as thresholded Borda count to defend against data poisoning in ensemble learning. Taken together, these contributions show how multiagent approaches can enhance both learning and trust, offering a pathway to AI systems that are capable, transparent, and reliable in practice
Investigating the Influence of Upper Limb Movements on Lower Limb Neuroplasticity in Humans
No full textThis dissertation investigates the influence of upper limb (UL) movements on lower limb (LL) neuroplasticity in humans, with a focus on both healthy individuals and those with chronic stroke. Using neurophysiological tools to assess spinal and cortical excitability, the studies examined whether UL activity modulates LL neuromotor function through interlimb neural coupling. In healthy adults, a novel UL movement paradigm was found to alter spinal excitability, suggesting shared pathways for coordinated motor control. While no uniform group-level effects were observed, unsupervised clustering revealed distinct facilitatory and inhibitory subgroups, highlighting consistent patterns of bidirectional modulation. These findings challenge the traditional assumption that UL activity uniformly suppresses LL reflexes, instead pointing toward individualized spinal network responsiveness.
Applying the same paradigm to individuals with chronic stroke revealed that residual interlimb connectivity can influence both spinal and cortical excitability, although with notable interindividual variability. Specifically, rhythmic, symmetric UL movements significantly enhanced paretic leg strength and prolonged cortical silent period duration, indicating supraspinal engagement with therapeutic potential. Although spinal effects were less consistent, a subset of participants demonstrated responsiveness, underscoring the importance of individualized assessment in clinical populations.
Together, these findings highlight the plastic potential of spinal and supraspinal pathways, suggesting that UL movement may serve as a novel priming strategy to facilitate LL motor recovery post-stroke. By advancing mechanistic understanding of interlimb interactions, this work underscores the clinical relevance of harnessing UL activity to promote neuroplasticity and functional rehabilitation. Importantly, the consistent observation of substantial individual variability across both healthy and stroke cohorts emphasizes the need for personalized approaches to neuromodulation. This dissertation thus provides proof-of-concept for integrating UL movement into rehabilitation frameworks and lays the groundwork for future studies on dosing, durability, and combination strategies to optimize recovery outcomes
Perceptions of Emergency Physicians in Japan in Managing Acutely Ill Pediatric Patients
No full textObjectives: The primary objective of this study was to delineate the self-perceived confidence levels of emergency physicians in Japan in managing acutely ill pediatric patients in the emergency department (ED). Based on the hypothesis that 50% or more participants did not feel confident, the secondary objective was to describe their underlying perceptions and barriers they faced in managing acutely Ill pediatric patients in the ED.
Methods: This study used a mixed quantitative and qualitative method. An anonymous on-line survey collected data on the demographics and the self-perceived confidence levels by Likert 5-point scale (quantitative data analysis). For those who responded “Neutral”, “Not confident”, or “Not confident at all” were eligible for the semi-structured individual interviews (qualitative data analysis).
Results: The study enrolled 54 emergency physicians in Japan Emergency Medicine Network (JEMNet). 50% of the participants responded either “Not confident” or “Not confident at all”. Fifteen subjects also participated in the individual interviews. The perceptions of those participants in managing acutely ill pediatric patients were uniformly negative due to factors including emotional distress, anxiety, and fear of worsening the clinical outcome by making inappropriate medical decisions and failing to successfully perform emergency procedures. The 3 key themes identified were lack of clinical experience in managing acutely ill pediatric patients, challenges with multidisciplinary team dynamics in pediatric resuscitations, and delivering bad news to family members after the death of a child in the ED.
Conclusions: 50% of the Japanese emergency physicians in the study did not feel confident in managing acutely ill pediatric patients in the ED and their perceptions were uniformly negative. A system-based approach is crucial to address this problem. The 3 key themes identified in the study could be reframed respectively as assurance of staff competencies in pediatric emergency care through education and training, allocation of appropriate institutional resources for pediatric resuscitations, and development of institutional policy and practice guidelines for pediatric resuscitations and family care in the ED. This comprehensive multi-level approach will most likely lead to the overall improvement in the quality of pediatric emergency care
The Impact of Depleting the Essential Bacterial Enzyme Peptidyl-tRNA Hydrolase on Translation
No full textPeptidyl-tRNA hydrolase (Pth) is an essential bacterial enzyme that recycles peptidyl-tRNAs (pep-tRNAs) by hydrolyzing the ester bond between the tRNA and the peptide. The best-characterized function of Pth is salvaging free pep-tRNAs that dissociate from the ribosome during translation to prevent starvation for translation-usable tRNAs. Depletion of Pth leads to the accumulation of pep-tRNAs, rapid inhibition of translation, and cell death. Recent studies have demonstrated that Pth can process pep-tRNAs associated with large ribosomal subunits and possibly even translating ribosomes. The discovery of these additional Pth substrates demonstrates that the physiological role of Pth may be more complex than was previously thought and brings longstanding beliefs about the essentiality of Pth into question.
We used genome-wide techniques to characterize the effects of the loss of Pth on translation and searched for factors that can compensate for its depletion in Escherichia coli to expand our understanding of the physiological role of Pth. In a thermosensitive Pth mutant, pth(Ts), we found that Pth depletion promotes ribosome stalling at many codons, especially at the early codons of ORFs. This broad stalling leads us to conclude that the cell may starve for many tRNAs rather than just tRNALys, as was previously envisioned. We discovered that the overexpression of a protein of unknown function, YajQ, partially restores the growth of cells depleted of Pth. YajQ overexpression globally reduces the ribosome stalling caused by Pth depletion at most codons, suggesting a possible role for YajQ in translation. To disentangle the effects of heat shock in investigating the functions of Pth, we created a heat shock-independent model of Pth depletion, in which Pth can be depleted at the temperature optimal for cell growth. Using this model, we found that Pth depletion still leads to widespread ribosome stalling, though it occurred at a few specific codons. Intriguingly, we found that overexpression of tRNAs that recognize the codons at which stalling occurred does not improve survival of cells during Pth depletion, suggesting that additional functions of Pth beyond recycling free pep-tRNAs may play a key role in the essentiality of this enzyme
Localizing Strain in a 2D in vitro Stretch System to Model Regional Injury Dynamics
No full textTraumatic brain injury (TBI) is a disruption in normal brain function caused by mechanical insult to the head. Nonuniform brain tissue deformation complicates TBI diagnosis and treatment. Capturing this spatial heterogeneity in vitro is important for studying region-specific injury responses under controlled conditions. This work presents the development and validation of a two-dimensional (2D) in vitro stretch injury model that produces bimodal strain within a single culture well, creating coexisting populations of injured and uninjured cells. The previously established system uses a flexible polydimethylsiloxane (PDMS) membrane indented by a rigid post to deliver a controllable, clinically relevant stretch to 2D cultures. To induce strain heterogeneity, the system was modified to increase friction at the membrane–post interface, reducing strain in the center of a well, and redistributing deformation to the periphery. Experimental strain measurement confirmed that central strain is reduced under high-friction conditions compared to low-friction. However, experimental measurement of peripheral strain was not feasible due to out of plane deformation of the periphery. A finite element analysis (FEA) was developed to address this limitation, incorporating membrane thickness and hyperelastic behavior. The FEA model, validated against experimental data, accurately predicted strain distributions and identified an indentation depth that produced the desired bimodal strain pattern— a protected center, and injurious periphery. Human induced pluripotent stem cell (hiPSC)-derived astrocytes were cultured on polydopamine and Matrigel® coated PDMS membranes and subjected to the high-friction indentation. Calcein AM staining before and after injury revealed decreased cell viability in the periphery, and maintained viability in the center, consistent with predicted strain fields. This study demonstrates, for the first time, a 2D in vitro stretch model generating spatially heterogeneous strain and biological responses within a single well. This work advances TBI modeling, demonstrates a novel method for culturing hiPSC-derived astrocytes on PDMS, and expands opportunities for mechanobiological studies of neural cell injury and repair
Blood Bank / Transfusion Medicine and Hemostasis RISE Reviews
No full text2025, 2026 RISE review slides for pathology residents.</p
A Pilot Clinical Trial of SunnysideFlex to Treat Posttraumatic Stress Disorder Symptoms During Pregnancy
No full textThe perinatal period is associated with a heightened risk of psychopathologies, including affective distress, which in turn is associated with both mother and infant adverse health outcomes. Although extensive literature exists on perinatal depression and anxiety, other perinatal mental health challenges that often co-occur with depression, such as posttraumatic stress disorder (PTSD), have received less attention. Approximately 4-8% of pregnant women meet criteria for PTSD during pregnancy, while it can also negatively impact forming of a secure relationship between mother and infant, from pregnancy through postpartum. Despite scientifically supported interventions for treating PTSD in the general population, treating PTSD during pregnancy has remained an understudied area and there is limited clarity of treatment guidelines for prenatal PTSD. Available research on perinatal interventions for PTSD more broadly suggests that addressing trauma during pregnancy is safe, feasible, and acceptable. In this pilot study, we assessed the feasibility, acceptability, and preliminary effectiveness of SunnysideFlex, a web-based cognitive behavioral therapy (CBT) intervention tailored for targeting PTSD symptoms during pregnancy.
This study had three aims. First, we analyzed individual-level changes in PTSD symptoms pre- to post-intervention. Second, we assessed the individual-level changes in self-reported measures of mother-infant bonding and attachment pre- to post-intervention. Third, we conduced individual-level analyses focusing on intervention adherence, engagement, and acceptability (i.e., overall usefulness, ease of use, ease of learning, and satisfaction), and their potential association with the individual’s changes in PTSD symptoms, as well as mother and infant bonding and attachment, utilizing a descriptive approach.
The primary clinical outcome of this study was a reduction in PTSD symptoms, with 85% of participants showing clinically meaningful reductions in PTSD symptoms from pre- to post-intervention. The results related to mother-infant bonding and attachment from pre- to post-intervention timepoints were more nuanced, such that only 5.5% of participants demonstrated a clinically significant improvement in bonding. Similarly, 4% showed a clinically significant increase in attachment. Overall, the individual-level analysis of engagement and clinical outcomes suggested an association between higher program engagement and improvements in key study outcomes, particularly PTSD symptoms. However, the relationship between tool usage and level of improvement in PTSD symptoms was not consistently clear. On the other hand, intervention adherence was high as all participants completed the 6-week SunnysideFlex program and engaged with an average of 10.75 out of 13 lessons. Overall, this pilot study provided preliminary support for the efficacy, feasibility, and acceptability of SunnysideFlex program as an evidence-based online intervention for improving PTSD symptoms in pregnant women
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇