113588 research outputs found
Sort by
Improving Clinician Knowledge and Confidence in Esketamine-Assisted Psychotherapy: A Quality Improvement Project
No full textPurpose: The purpose of this quality improvement (QI) project was to increase mental health care professionals’ (MHCPs) knowledge, confidence, and motivation to recommend or initiate esketamine-assisted psychotherapy (EAP) through a targeted educational intervention in a large outpatient psychiatric clinic that provides esketamine therapy for treatment-resistant major depressive disorder (TR-MDD). Background: Major depressive disorder (MDD) is a leading cause of global disability, with many patients failing to achieve remission despite multiple treatment trials (Gutiérrez-Rojas et al., 2020; World Health Organization [WHO], 2020). Treatment-resistant depression (TRD) has been associated with significant functional impairment, increased suicide risk, and frequent relapse (Barbui et al., 2021). Esketamine, approved by the FDA in 2019, offered rapid antidepressant effects for TRD. When combined with psychotherapy as EAP, it has been shown to enhance neuroplasticity and treatment response (Canuso et al., 2018; Popova et al., 2019). Despite this potential, esketamine has rarely been integrated with psychotherapy in clinical practice. Methods: The project used the Plan-Do-Study-Act (PDSA) framework to guide implementation. Eligible participants included psychiatric prescribers, therapists, care coordinators, and administrators. A single 20-minute educational session was delivered during an interdisciplinary meeting, focusing on neurobiological mechanisms, psychotherapy timing, and workflow integration. Anonymous pre- and post-surveys assessed changes in knowledge, confidence, and perceived barriers. Descriptive statistics and paired t-tests evaluated results, and open-ended responses informed future training. Results: All 24 participants completed both surveys, yielding a 100% response rate. Knowledge of EAP psychotherapy modalities increased from 70.8% to 100% (p = .005). Confidence identifying integration strategies improved from 2.83 to 4.71, and belief in EAP’s clinical benefit rose from 3.67 to 4.79. Qualitative feedback highlighted improved interdisciplinary collaboration and readiness to implement EAP. Conclusions: The project effectively enhanced clinician understanding of EAP and improved interdisciplinary readiness. A brief, theory-informed educational intervention feasibly strengthened confidence, collaboration, and motivation to integrate psychotherapy with esketamine treatment within routine psychiatric care
Corpus-Based Pedagogy in Undergraduate L2 Writing Courses: Effectiveness, Engagement and Instructors’ Understanding
No full textCorpus-based pedagogy, or data-driven learning (DDL), aims to use large collections of real-world language examples to support students in analyzing language patterns and variations in authentic contexts, guiding them to inductively discover target language features, and develop their language awareness and knowledge (Friginal, 2018; O’Keeffe & McCarthy, 2010; Poole, 2022; Reppen, 2010). Recent studies found that corpus-based pedagogy has been increasingly adopted in writing education across different areas (Boulton & Cobb, 2017; Boulton & Vyatkina, 2021; Chen & Flowerdew, 2018). While professional and published-text corpora have been widely applied in corpus-based writing courses, the value of learner corpora is often overlooked in current writing instruction, despite the fact that they can provide more accessible and non-threatening texts that better align with genres students are expected to write (Seidlhofer, 2002). The three studies of the dissertation are centered around implementing corpus-based activities created using a learner corpus Corpus & Repository of Writing (Crow) (Staples & Dilger, 2018-), to enhance L2 students’ writing skills in first-year writing courses. Employing both quantitative and qualitative methods, students’ drafts, surveys from both students and instructors, semi-structured interviews, and teacher reflections were collected and analyzed. Article one explores the effectiveness of integrating texts and data derived from the learner corpus Crow in teaching evaluative language (evaluative adjectives and modal verbs) in an academic writing project. To make corpus-based pedagogy more engaging for students, Article two extends to integrating game elements, introducing a series of gamified corpus-based materials and investigating the students’ and instructors’ perceptions as well as addressing the benefits of gamifying corpus-based materials in their L2 writing classes. Finally, Article three examines non-corpus specialist instructors’ understanding and perceptions of corpus-based pedagogy before, during, and after implementing ready-to-use corpus-based materials in their own teaching contexts. Together, the three articles highlight the effectiveness of using learner corpus-based materials in teaching evaluative language, introducing practical strategies for increasing student engagement in corpus-based writing courses, and exploring non-corpus specialist instructors’ evolving understanding of corpus-based pedagogy throughout their first implementation. The dissertation contributes to addressing the pedagogical applications of learner corpora and offers insights for designing effective, engaging, and accessible corpus-based activities for L2 writing contexts
THE TIRE PRESSURE MONITORING SYSTEM
No full textThe tire pressure monitoring system (TPMS) is a telemetry system used to monitor tire diagnos-
tics. In this paper, we analyze the transmission and reception of tire telemetry data and relate it
to aeronautical mobile telemetry (AMT). The comparison focuses on the encoding, modulation,
transmission, demodulation, and decoding of TPMS across various vehicle manufacturers.International Foundation for TelemeteringProceedings from the International Telemetering Conference are made available by the International Foundation for Telemetering and the University of Arizona Libraries. Visit https://telemetry.org/contact/ if you have questions about items in this collection
PILOT SEQUENCE DESIGN FOR TELEMETRY WAVEFORMS IN MULTI-ANTENNA TRANSMITTERS
No full textIn multi-antenna systems, accurate Channel State Information estimation is crucial for reliable
data decoding. In aeronautical telemetry, the spatial separation of antennas onboard aircraft, which
can span several meters, introduces non negligible differential delays between received signals at
receiver side which degrades channel estimation, thereby impacting the communication performance. To address this issue, we propose to use Space Time Coding to generate pilot sequences
that mitigate differential delay degradation. Thus leading to improved receiver robustness in multi-
antenna aircraft telemetry systems. Complete communication system simulation using AWGN
channel showcases gains in CSI estimation up to 6dB without any increase in pilot sequence size
nor in computational complexity.International Foundation for TelemeteringProceedings from the International Telemetering Conference are made available by the International Foundation for Telemetering and the University of Arizona Libraries. Visit https://telemetry.org/contact/ if you have questions about items in this collection
ENHANCING CUBESAT TELEMETRY SYSTEMS FOR AUTONOMOUS SPACE MISSIONS UTILIZING MACHINE LEARNING TECHNIQUES
No full textA CubeSat is a valuable tool used by many organizations, including NASA, who partners with universities to design and build satellites for data collection. A primary challenge for CubeSats is maintaining reliable telemetry during autonomous operations. The objective of this paper is to present a machine learning-driven approach to improve real-time data analysis and anomaly detection. The proposed algorithm has the potential to improve the decision-making and reliability of the CubeSat telemetry system, while addressing its unique constraints. The machine learning algorithm, incorporating data supplied by Attitude Determination and Control System (ADCS) components, could find new avenues to increase the efficiency of satellite reorientation based on supplied attitude determination data. Enhancements to the CubeSat operating system could allow for more effective research of autonomous space missions.International Foundation for TelemeteringProceedings from the International Telemetering Conference are made available by the International Foundation for Telemetering and the University of Arizona Libraries. Visit https://telemetry.org/contact/ if you have questions about items in this collection
Data Efficient Learning for Space: Neural Networks under Scarce Real Data
No full textDeep learning has demonstrated remarkable performance across diverse tasks, yet its success remains fundamentally limited by the availability of large labeled datasets. This constraint is particularly acute in space applications, where data collection is expensive, time-consuming, or physically impossible. While synthetic data generated from high-fidelity simulations offers an attractive solution, models trained exclusively on synthetic data often suffer significant performance degradation when deployed on real-world data due to distributional differences known as the "reality gap." This dissertation investigates domain adaptation techniques to bridge this gap across two critical space applications.The first application addresses satellite bus classification from photometric light curves for space situational awareness. With limited real light curves distributed across 17 satellites, we explore meta-learning techniques, specifically Proto-MAML, which enable rapid adaptation to new classification tasks with minimal examples. Our results demonstrate that meta-learning is able to generalize between the different buses; however, the limited number of available data and classes limits the overall performance. We compare this approach to transfer learning via fine-tuning a pre-trained model, and we perform a data analysis to understand which features is more important to differentiate between buses. The second application focuses on autonomous lunar hazard detection using both optical imagery and LiDAR-derived digital elevation models (DEMs). We evaluate transfer learning through fine-tuning and pixel-level domain adaptation via CycleGAN. For DEMs, we show that a model pre-trained on synthetic data achieved low accuracy on real LiDAR scans. But after a quick fine-tuning on a small real-world dataset, we achieve significant accuracy improvements. For optical images, we compare four approaches across multiple lunar terrains. Here we test and compare the CycleGAN technique. The results demonstrate that domain adaptation always brings improvements. We also show that terrain features influence the mapping between the simulated and real domains. This work provides practical insights into the applicability and limitations of domain adaptation for space applications. We demonstrate that while synthetic data alone is insufficient for real-world deployment, strategic use of limited real data through fine-tuning or integrated domain adaptation can bridge the reality gap. Our findings contribute to the growing body of evidence supporting the viability of deep learning for safety-critical space missions, while highlighting the continued need for careful validation and targeted data collection strategies
Design, Characterization, and Implementation of a Self-Aligned Focusing Schlieren System
No full textA self-aligned focusing schlieren (SAFS) system was developed to visualize changes in density gradients with a reduced depth of field (DOF) compared to traditional schlieren techniques, which visualize gradients along the entire light path and can make it difficult to isolate specific regions of interest. SAFS addresses this limitation while also offering simple setup and alignment. Multiple configurations were designed to explore how parameters such as field of view (FOV), DOF, and sensitivity influence system performance. These configurations were characterized through benchtop and wind tunnel testing to evaluate performance and image quality. Building on this foundation, the configurations, including a novel dual-camera design, were tested in the University of Arizona’s Indraft Supersonic Wind Tunnel (ISWT) to study flow over a 26 degree compression ramp at Mach 2.28. Each configuration successfully captured distinct regions of the flow field, demonstrating focusing capability. Systems with smaller DOF better resolved the three-dimensional structure of the separation shock, while the dual-camera setup confirmed the feasibility of simultaneous multi-plane imaging. These results show that SAFS can be used as a powerful tool for high-speed flow visualization
Alteration of Wingtip Vortices using Wingtip Jets on Fixed Wings and Rotating Blades
No full textThis dissertation presents a comprehensive experimental investigation into the dynamicsof wingtip vortices and the application of spanwise blowing jets as a method for active flow control on both fixed-wing and rotor blade systems. Wingtip vortices are a natural three-dimensional phenomenon of finite wings that reduces aerodynamic performance and create hazardous wakes. In the case of rotorcraft, the blade wake can be particular detrimental in the form of Blade Vortex Interactions (BVI). This work investigates the impact of wingtip jets on the aerodynamic performance and flow around both a small aspect ratio wing and a rotor. Wind tunnel experiments are conducted on a NACA 0012 fixed-wing model with jet slots at the wingtip, downstream of a built-in settling chamber. Force balance measurements determine the impact of active flow control on aerodynamic loads, while Particle Image Velocimetry (PIV) data are acquired at three chordwise locations to characterize the flow around the wingtip. The results demonstrate that the addition of wingtip jets increases both overall lift and drag, with the effects being more pronounced at lower freestream velocities for a given jet supply pressure ratio. PIV analysis reveals that the jets fundamentally alter the flow field, displacing the primary wingtip vortex and inducing a more complex flow field with secondary counter-rotating vortices. This outward movement of the primary vortex, combined with a reversal of the spanwise flow, directly correlates with the increased lift observed in the force balance experiments. Wingtip jet effects on the wingtip vortex are also investigated on a rotor with two blades. The rotor is set with a 12-degree blade angle of attack and a 6-degree disk tilt relative to a Reyonlds number of 2.45 × 10^4 based on chord length. PIV data acquired at four azimuthal blade positions confirm that, similar to the fixed-wing model, the jets displace the blade tip vortex and reverse the spanwise flow. The comparative analysis reveals that the jets are most effective at low Re, which are experienced by the retreating blade due to a reduced relative freestream velocity, where flow separation is a major concern. The application was less effective on the advancing blade, where the presence of BVI remains an issue
Geologic Map of the Safford 30' x 60' Quadrangle, Graham and Greenlee Counties, Arizona, Grant and Hidalgo Counties, New Mexico
No full textThe Safford 1:100,000-scale quadrangle encompasses a broad swath of diverse geology in the Mexican Highland subsection of the Arizona Basin and Range Province in southeastern Arizona. The geology within the quadrangle is diverse, including Paleoproterozoic and Mesoproterozoic metamorphic and igneous rocks, Cretaceous to Paleogene (Laramide) intrusive and volcanic rocks, Neogene volcanic rocks, intrusive rocks and basin deposits, and extensive Quaternary surficial deposits cover much of the map area.The quadrangle includes the growing community of Safford (about 10,000 people) and numerous smaller communities. The Safford Mine in the northern part of the quadrangle is an open-pit mine operated by Freeport-McMoran, extracting primarily copper from the San Juan, Dos Pobres, and Lone Star ore bodies. Deposits of the Gila River supply abundant, high-quality aggregate for the region.A variety of published map data was utilized to construct this Safford 1:100,000-scale quadrangle. Substantial portions of the quadrangle had been mapped previously at 1:24,000 or 1:48,000 scale by geologists with the Arizona Geological Survey or the U.S. Geological Survey. In these areas, the large-scale mapping was generalized and modestly reinterpreted. Other parts of the quadrangle had been mapped at smaller scale by U.S. Geological Survey geologists. This mapping was used for the bedrock compilation, but surficial and basin deposits were remapped as the current map was developed.
This geologic map was funded in part by the U.S. Geological Survey National Cooperative Geological Mapping Program under STATEMAP award G23AC00556. The views and conclusion obtained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Government.Documents in the AZGS Documents Repository collection are made available by the Arizona Geological Survey (AZGS) and the University Libraries at the University of Arizona. For more information about items in this collection, please contact [email protected]
Next-Generation Computational Phenotyping with Large Language Models
No full textIn this dissertation, we presented a critical re-examination of computational phenotyping, a foundational activity in biomedical informatics that supports cohort discovery, observational research, and clinical quality improvement. Despite the development of numerous computable phenotypes across a wide range of clinical outcomes and conditions, the field continues to rely on labor-intensive methods involving manual review and algorithm design. In response, we introduced novel phenotyping methods using Large Language Models (LLMs) to reduce human burden and achieve synergy between human expertise and machine intelligence. These methodological enhancements enabled successful application of LLMs to phenotyping processes previously requiring substantial human oversight. Our work lays the groundwork for the next-generation of computational phenotyping methods, redefining how clinical knowledge is extracted and applied in the era of artificial intelligence. Each of the studies presented in this dissertation supported the progression of next-generation phenotyping methods by assessing the application of LLMs to computational phenotyping tasks. In the first study, we presented PHEONA (Evaluation of PHEnotyping for Observational Health Data), an evaluation framework specifically for LLMs. The components of this framework allowed us to thoroughly evaluate the suitability and feasibility of LLMs for various computational phenotyping tasks. In the second study, we developed a companion framework, SHREC (SHifting to language model-based REal-world Computational phenotyping), that outlined both an end-to-end phenotyping pipeline and the steps necessary to advance next-generation phenotyping methods. Using this framework, we assessed LLMs for concept classification and phenotyping of encounters, which were both individual steps within the end-to-end pipeline. Finally, in the third study, to further evaluate performance deficiencies in applying LLMs to these tasks, we enhanced PHEONA to include an assessment of faulty reasoning within LLM responses.Release after 09/05/202