18525 research outputs found
Sort by
Glofinder: Ai-Empowered Qupath Plugin for Wsi-Level Glomerular Detection, Visualization, and Curation
No full textRecent advances in medical imaging have underscored the superiority of circle-based object detection techniques for identifying spherical structures, such as glomeruli, cells, and nuclei, compared to traditional bounding box-based methods. Circle representations naturally conform to the inherent geometry of these spherical entities, resulting in more accurate and meaningful localization. In typical bounding box-based detection tasks, combining predictions from multiple models has become a prevalent strategy to enhance accuracy, particularly in contexts where real- time processing constraints are relaxed. However, such ensemble approaches have yet to be effectively translated to circle-based detections, presenting a noticeable gap in existing methodologies. To address this limitation, our study introduces Weighted Circle Fusion (WCF), a novel ensemble method specifically tailored for merging circle-based predictions. WCF utilizes the confidence scores assigned to each predicted circle, enabling a weighted combination of outputs from multiple detection models, thereby significantly enhancing detection accuracy and reducing false positives.
We validated the effectiveness of WCF using a proprietary dataset dedicated to glomerular detection within whole slide images (WSIs). Our results clearly demonstrate that the proposed method delivers superior performance in terms of precision and robustness compared to individual model predictions and conventional ensemble methods. Additionally, our investigation extended to evaluating annotation strategies, particularly exploring the benefits of incorporating a human-in-the-loop (HITL) approach. By integrating automated model predictions with human verification, we confirmed that HITL markedly accelerates annotation speed and improves data quality over traditional fully manual annotation methods. Capitalizing on these findings, we implemented WCF into a practical, user-friendly QuPath plugin named GloFinder, streamlining accurate glomeruli detection in WSIs and significantly optimizing medical imaging workflows
Robust and Reliable Deep-Learning-Based Techniques for Image-Guided Cochlear Implant Procedures
No full textCochlear implants (CIs) are neuroprosthetic devices that restore hearing in patients with severe-to-profound sensorineural hearing loss by bypassing damaged auditory pathways. Despite their success, CI outcomes vary considerably, highlighting the need for image-guided methods to optimize surgical planning and CI programming. This dissertation develops robust and reliable deep learning-based techniques for three critical aspects of image-guided cochlear implant procedures. First, it introduces a hybrid active shape model and deep learning (DL) method for robust and accurate segmentation of the intracochlear anatomy in clinical CT images. Second, it presents a unified DL framework for electrode array (EA) localization that can handle both closely- and distantly-spaced EAs, complemented by an automated image quality assessment network to ensure reliable results. Finally, it addresses the challenge of MR-only preoperative planning by developing methods to synthesize CT images from multi-sequence MRI, demonstrating robustness across multiple imaging sites and scenarios with missing modalities. As a whole, the methods in this dissertation achieve high accuracy while ensuring robustness and reliability, marking a significant advancement in improving image-guided cochlear implant procedures through automated image analysis techniques
Carbon Cowboys: Myth, Land, and Political Synthesis in the Rocky Mountain Energy Boom
No full textThis dissertation traces the Rocky Mountain energy boom, a period of intense political, economic, and environmental change in the Rocky Mountain states between 1973 and 1983 rooted in oil, gas, and coal extraction projects. It tracks key legal, environmental, political, ideological, and business interests that came together in a powerful regional configuration that would have national consequences for conservative politics and thought. Following oil executives and their political organizations dedicated to resisting federal regulation on public lands, it argues that the oil industry in Wyoming, Colorado, and Utah created powerful networks that molded the mountain landscape, recast western collective memory, redefined expectations of government intervention, and steered the priorities and power of the New Right
Edge-centric Network Analytics
No full textNetwork analysis has witnessed significant advancements in node-centric and graph-centric tasks, yet edge-centric analytics, crucial for understanding relationships within networks, remain underexplored beyond link prediction. This gap is particularly evident in domains such as social network analysis, cybersecurity, and bioinformatics, where edge dynamics play a pivotal role. To address this, my thesis systematically explores edge-centric analytics through four key chapters, each tackling unique challenges and opportunities.
The research begins with the challenge of imbalanced edge classification, where real-world networks often exhibit label and topological imbalances. To address this, we introduce topological entropy (TE), a novel edge-level measurement, and develop TopoEdge that combines novel reweighting and synthetic data generation strategies to improve imbalanced edge classification performance in real-world settings.
Building on this, we explore the problem of social tie strength, a specific and underexplored form of edge classification task where ground truth is often limited or unavailable. Through a review of pseudo-labeling techniques and empirical analysis, we identify dataset-specific inconsistencies in modeling, emphasizing the need for developing generalizing predictive methodologies.
Overall, this research provides a comprehensive exploration of edge-centric tasks, highlighting their importance across diverse applications. By addressing these critical challenges and proposing innovative methodologies, this work aims to advance the field of edge-centric network analysis and stimulate further research in this essential yet underexplored domain
Refining Shockley’s Electron-Hole Pair Equation Using Phonon Dispersions
No full textWe investigate refining Shockley’s electron-hole pair model by incorporating contributions of optical and acoustic phonons. Using full-band Monte Carlo software, called Anduril, we determine ionization energies in silicon (3.47 ± 0.05), germanium (2.53 ± 0.05), and gallium arsenide (3.98 ± 0.05), as well as their energy loss mechanisms via phonon emission. When compared to phonon dispersions, there is relationship between the energies from the acoustic and optical branches to the preferred energy loss mechanism for hot carriers. This model eliminates the need for expensive software simulations and provides a more physical basis to Shockley’s fitting parameters
A hierarchical model of youth internalizing symptoms and associated neurostructural differences
No full textInternalizing symptoms often have their onset during youth and are associated with
differences in neurostructural development. However, most studies conceptualize internalizing disorders categorically, which does not adequately capture the dimensional, transdiagnostic, highly correlated, and hierarchical nature of these symptoms. Prior studies investigating dimensions of internalizing psychopathology in youth are limited, and few have related dimensional models of youth internalizing symptoms to brain structure. The present studies leverage a large sample (N = 11,868) of 9-to-10-year-old children and hierarchical modeling techniques to delineate the underlying structure of youth internalizing problems (Study 1) and identify associated neurostructural differences using regional gray matter volume (GMV), cortical thickness (CT), and cortical surface area (SA) (Study 2). The hierarchical modeling in Study 1 revealed factors for general internalizing, distress, cognitive, fear, and somatic symptoms. After controlling for age, sex, income, parental education level, and MRI device, Study 2 results showed that general internalizing symptoms, distress symptoms, and cognitive symptoms were each associated with relatively widespread smaller regional GMV and SA. Fear symptoms were associated with a more localized pattern of smaller SA in some regions within the parietal, temporal, and insular cortices. No associations were found for CT regions or the somatic symptoms factor. Together, these results suggest that at 9-10 years of age, GMV and SA may be more relevant to internalizing psychopathology than CT; general internalizing, distress, and cognitive symptoms may have a relatively widespread inverse relationship with GMV and SA; and that general internalizing, distress, and cognitive symptoms may have stronger relationships with brain structure than somatic and fear symptoms
Master Mercenaries: Moroccan Regulares in the Spanish Civil War
No full textHistory Department Honors ThesisCollege of Arts and ScienceDepartment of Histor
Secondary Educators' Confidence in Using Research-Based Inclusive Practices: A Comparison of General and Special Education Educators Across Tennessee
No full textTrustees: Exploring an Elite Student Leadership Program
No full textLeadership and Learning in Organizations capstone projectThe focal organization of this study is a small,Christian liberal arts university located in a rural city in the Midwest. The Presidential Fellows Program at the university is an annual, competitive leadership program open to junior and senior class students. The university invests approximately $100,000 into the Fellows Program per year, in addition to the Fellows’ and Principals’ time.
This capstone project explores the Fellows’ experiences and perceived increase in leadership competencies. The study used a mixed-method design, a Student Leadership Competencies Inventory (SLCI) and semi-structured interviews. The study suggests that 1) Fellows most value the networking and mentorship aspects of the program; 2) confidence and appropriate interaction increased most among Fellows’ SLCIs; 3) variances were slight among the Fellows’ SLCI gains; however, some Fellows of color experienced higher tensions as leaders; and 4) all Fellows prized cohort relationships and community
Beyond Access: Understanding Financial, Cultural, and Structural Barriers to Persistence for 21st Century Scholars
No full textLeadership and Learning in Organizations capstone projectThe 21st Century Scholars Program has dramatically increased college access for low-income and first-generation students in Indiana by covering full tuition at public institutions. Despite an 88% matriculation rate, fewer than half of Scholars complete a post-secondary credential. This mixed-methods work explores the cultural, structural, and economic factors influencing college persistence among Scholars. Quantitative data from administrative records and student surveys, alongside qualitative interviews, reveal three primary barriers: non-tuition financial stressors, inconsistent institutional support, and limited cultural capital among first-generation students. While persistence rates have improved modestly over time—particularly at four-year institutions—racial disparities remain significant. Based on these findings, we recommend increasing flexibility in enrollment requirements, offering support to offset non-tuition expenses, and enhancing navigational and community-building supports for first-generation students