Electronic Theses and Dissertations

Permanent URI for this collection

https://ir.vanderbilt.edu/handle/1803/9599

Electronic theses and dissertations of masters and doctoral students submitted to the Graduate School.

Browse

Now showing 1 - 20 of 7532

Deep-Learning-Enhanced Atlas-Based Preoperative and Intraoperative Registration for Cochlear Implant Surgery Navigation
(2025-05-22) Zhang, Yike; Noble, Jack; Noble, Jack
This dissertation provides the groundwork for intraoperative registration in cochlear implant surgery though the developed Vision6D pose annotation tool and a series of deep-learning-based methods. The primary contributions can be summarized as follows. First, self-supervised ossicles registration and segmentation, as detailed in Chapter 2. Second, the development of the Vision6D software and its comprehensive user study using two public 6D pose estimation datasets, as introduced in Chapter 3. Third, 2D monocular microscope views to 3D CT registration using the incus of the ossicles as a landmark, which is described in Chapter 4. Fourth, mastoidectomy shape prediction to extract the postmastoidectomy mesh directly from preoperative CT scans, as shown in Chapters 5, 6, and 7. Fifth, postmastoidectomy surface multi-view synthesis from a single microscope image is proposed in Chapter 8. Sixth, surgical scene completion for the synthetic postmastoidectomy surface multi-views through single-step denoising diffusion GAN, as illustrated in Chapter 9. Finally, Chapter 10 utilizes the prior contributions from Chapters 5 to 8 to perform the monocular patient-to-image intraoperative registration for cochlear implant surgery that leverages the synthetic surgical views. These combined components provide numerous opportunities for future intraoperative navigation systems and surgical applications.
Network Analysis and Visualization of Disease Multimorbidity Using Electronic Health Records and Genetic Biobank Data
(2025-05-28) Zhang, Siwei; Xu, Yaomin; Shotwell, Matthew S.
Disease multimorbidity, the co-occurrence of multiple diseases within an individual, presents complex challenges for both public health and precision medicine. Advancing our understanding of multimorbidity can illuminate disease mechanisms, reveal patient heterogeneity, and enable biomarker discovery and treatment repurposing. Large-scale Electronic Health Records (EHR) and EHR-linked genetic biobanks offer unique opportunities to quantify phenome-wide multimorbidity, uncover shared genetic mechanisms among co-occurring conditions, and define multimorbidity-based disease clusters. However, major analytical and methodological challenges remain. To address these, we present three key contributions. First, we introduce a phenome-wide multimorbidity network that quantifies nonrandom disease-disease co-occurrences while accounting for potential confounding factors. Second, we develop a genetic discovery platform that integrates polygenic scores for predicted transcriptomic, proteomic, and metabolomic traits with phenome-wide association studies (PheWAS) to uncover shared biological mechanisms among multimorbid conditions. To support exploration, we also develop an interactive network visualization tool featuring dynamic cluster analysis of biological pathways linked to diseases with similar multimorbidity patterns, enabling intuitive exploration of complex disease relationships and their shared biological mechanism. Third, we propose a model-based clustering framework using a bipartite stochastic block model (biSBM) with a stability-driven post-processing step to identify robust disease clusters and patient subgroups from individual-level EHR data. This framework demonstrates superior performance in simulations and replicates coherent, interpretable multimorbidity structures across independent datasets, including UK Biobank and Vanderbilt BioVU. A case study of JAK2V617F somatic mutation carriers reveals genetic heterogeneity across patient subgroups with distinct multimorbidity patterns, illustrating the potential of our data-driven approach to uncover mechanistic insights into patient heterogeneity through EHR-derived multimorbidity networks.
Constructing a Secure and Autonomous Infrastructure for Smart Cities
(2025-07-09) Yu, Alian; Lin, Dan; Lin, Dan
The rapid evolution of urban environments, propelled by advances in the Internet of Things (IoT) and autonomous vehicle technologies, has ushered in a new era of smart city infrastructure. This dissertation investigates four critical components that together form the backbone of a secure, efficient, and resilient urban digital ecosystem. First, this dissertation develops a flexible access control technique for large-scale public IoT services to manage vast, interconnected networks securely, ensuring data integrity and user privacy. Second, this dissertation proposes an AI-enabled efficient traffic scheduling for autonomous vehicles, which uses real-time decision-making to optimize routes, reduce congestion, and improve urban mobility. Third, to mitigate the risks of data exposure in centralized systems, this work extends to a privacy-preserving model that trains locally while sharing statistical traffic data for global optimization, ensuring secure and efficient decentralized traffic management. Finally, this dissertation proposes a neighborhood watch mechanism for attack detection and evacuation, which enhances intersection security by monitoring data for anomalies and triggering safety protocols to prevent cyber threats and system failures. These integrated solutions address critical challenges in modern urban environments. This dissertation is organized around the synergistic integration of these research areas. By establishing a secure IoT framework, enhancing autonomous traffic scheduling through AI, provide scalable and privacy-preserving traffic optimization, and implementing robust intersection safety measures, this dissertation provides a holistic solution to the challenges facing modern smart cities and illustrates how a multi-faceted approach can drive forward the next generation of urban infrastructure.
White Matter-Associated Functional Connectivity in the Human Brain
(2025-06-11) Xu, Lyuan; Ding, Zhaohua
Functional magnetic resonance imaging (fMRI) has traditionally focused on gray matter (GM), while blood-oxygen-level-dependent (BOLD) signals in white matter (WM) were often dismissed as physiologically irrelevant. However, growing evidence indicates that WM exhibits coherent and meaningful neural activity. This dissertation systematically explores the spatiotemporal organization, modulation, and clinical relevance of WM-associated functional connectivity (FC) and proposes a novel framework to characterize the WM-GM connectome. We begin by examining the anisotropic structure of spontaneous BOLD correlations in WM, characterized through the construction of functional correlation tensors (FCTs). Leveraging a large normal aging cohort, we identify region-specific aging effects on FCT-derived indices, providing detailed maps of age-related alterations in local FCT metrics. We further examine seasonal modulations of WM-GM FC in a healthy population. By modeling sinusoidal trends across seasons, we detect significant periodic variations in low-frequency fluctuations, global and network-level FC strength, and topological properties of brain networks. These variations correlate with environmental factors like daylength and temperature, highlighting the dynamic influence of external cycles on intrinsic brain organization. Extending to clinical context, we analyze WM-GM FC and network properties in individuals with preclinical Alzheimer’s disease (AD) or AD dementia, compared to controls. Our findings reveal that the WM-GM functional connectome undergoes regional and systemic dysfunctions as early as in the preclinical stage, correlating with amyloid deposition and cognitive impairment. Finally, we introduce a hypergraph-based model to capture high-order WM-GM interactions, revealing group-level differences in clustering coefficients and centralities that may serve as novel AD biomarkers. Overall, this work establishes WM functional connectivity as a systematic and clinically informative feature, expanding our understanding of the functional role of WM in the human brain.
Discovery and Validation of Gut Microbiome and Circulating Metabolites Associated with Dietary Patterns and Cardiometabolic Diseases
(2025-07-02) Wang, Lei; Yu, Danxia; Yu, Danxia
Diet plays an important role in the development and progression of cardiometabolic diseases (CMD); however, the underlying biological mechanisms are not yet well understood. To address this gap, we integrated fecal metagenomics and blood metabolomics to identify and validate gut microbiome and circulating metabolites associated with different dietary patterns–Healthy Eating Index (HEI), Dietary Approaches to Stop Hypertension (DASH), Empirical Dietary Inflammatory Potential (EDIP), Empirical Dietary Index for Hyperinsulinemia (EDIH), and Ultra-Processed Foods (UPF)–and CMD. Microbiome analyses included 514 Black/African American participants from the Southern Community Cohort Study (SCCS) for discovery and 2,133 participants from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) for validation. Metabolomics analyses were conducted in the SCCS (N=1688) for discovery and in the HCHS/SOL (N=5,164) and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (N=2,315) for validation. Associations of five dietary patterns with microbiome abundance and circulating metabolites were assessed using linear regression after data transformation and adjusting for sociodemographics, lifestyles, and history of major chronic diseases. Logistic or Cox regression was used to further evaluate the associations of diet-related bacteria and metabolites with CMD. We identified and validated inverse associations between healthy dietary patterns (i.e., HEI and DASH) and abundance of Collinsella and C. aerofaciens, which were further linked to risk of obesity. Additionally, we validated a broad spectrum of metabolites across key biochemical pathways (e.g., amino acids, lipids, cofactors and vitamins, and xenobiotics) for those dietary patterns–HEI (n=89), DASH (n=52), EDIP (n=11), EDIH (n=60), and UPF (n=38); several of these diet-related metabolites were further linked to major health outcomes, including incident coronary heart disease (n=7), cardiovascular disease mortality (n=2), and total mortality (n=52). Importantly, we developed a metabolite signature for each dietary pattern, which demonstrated strong associations with disease outcomes and explained a significant proportion (>60%) of the diet-disease associations. These findings underscore the potential of multi-omics approaches to elucidate the complex interplay between diet and CMD, offering more precise insights for tailored dietary interventions.
Intact Audiovisual Spatial Integration in Autism: Lessons from Brain and Behavior
(2025-05-30) Vassall, Sarah Gibran; Wallace, Mark; Cascio, Carissa; Polyn, Sean
In addition to the core features of social communication impairments (SCI) and patterns of restricted and repetitive behaviors and interests (RRBIs), autism is characterized by changes in sensory processing, including in the integration of information across the different sensory modalities. While there is significant evidence that audiovisual (AV) integration is altered in the temporal domain in autism—and that these differences are associated with presentation of SCI and RRBIs—very little is known about how autistic individuals integrate AV information in the domain of space. We conducted a free-response AV spatial localization task in autistic and non-autistic children aged 7-17 years with EEG collection to understand AV localization abilities, as well as how these abilities relate to SCI, RRBIs, and other clinical features of autism. While we found no group differences in AV spatial performance, an exploratory analysis revealed three clusters of autism features that were significantly predictive of performance across groups: a motor/somatosensory cluster, a sensory responsivity cluster, and a social communication cluster. Similarly, a model comparing participant performance to optimal integration as measured by maximum likelihood estimation (MLE) revealed no group differences, but rather that children more broadly may be suboptimal in their integration of spatial stimuli. Interestingly, EEG power analyses did reveal differences in how autistic and non-autistic children may be processing AV spatial information, including different patterns of posterior and anterior activity in AV conditions. From our primary studies, we suggest that while autistic children with low support needs may have similar ability to integrate AV signals in space, they may be using alternative neural mechanisms to achieve this. Other chapters discuss autism in the context of predictive coding, as well as the lack of sampling diversity in multisensory research to date. We provide suggestions for more inclusive sampling and experimental approaches and possible directions for future research.
An Electrochemical Biosensor Array Platform for the Continuous, Dual-technique Monitoring of Multiple Biomarkers in Organ-On-Chips
(2025-07-15) Tuladhar, Pragun Ratna; Cliffel, David E
Electrochemical biosensors are ideal complementary tools to evaluate organ-on-chip (OoC) systems because they can be miniaturized, allow multiplexed measurements, have rapid response times, and require minimal sample preparation while being inexpensive. In this work, a microfluidic multianalyte electrochemical sensor array (µMESA) platform is presented for the continuous, dual-technique, multianalyte detection of biomarkers in OoCs. Surface and electrochemical characterization of the screen-printed electrode (SPE) revealed heterogeneous Pt surfaces that were stable and reproducible enough for sensing. The SPE with glucose, lactate, glutamate, and acetylcholine biosensors was integrated into the µMESA and simultaneous, selective detection of the four corresponding biomarkers under laminar flow was demonstrated. The µMESA was used to measure glucose consumption by bacterial cells. Osmium-based redox polymers (OsPVIAA) with various metal-loadings were synthesized, but higher loadings only correlated to larger oxidation peaks up to 32%. OsPVIAA was highly successful in preventing background interfering signals from cell media. Optimizing hydrogel crosslinking and incorporating a thin poly(3,4-ethylenedioxythiophene) (PEDOT) film produced an OsPVIAA-glucose sensor that minimized loss of the polymer to the bulk solution. Fluorosilicone-based ion-selective membranes for K⁺ and Ca²⁺ were also prepared on the SPE. However, they exhibited super-Nernstian responses during calibrations. Simultaneous amperometric (acetylcholine, glutamate) and potentiometric detection (K⁺, Ca²⁺) of analytes was also presented. This work explored the multianalyte detection, interference mitigation, and sensor stability aspects of electrochemical biosensing. By demonstrating its applicability in dynamic environments like flow conditions and biological media, the µMESA platform was shown to be well-suited to electrochemically analyze tissue functions in OoCs.
Deconvolution of Permeability-Activity Relationships for Verticilide-Inspired Cyclic (Depsi)Peptides and Progression of the Structure-Activity Relationship Campaign for ent- V erticilide
(2025-07-01) Thorpe, Madelaine Paige; Johnston, Jeffrey N
Through a cross-disciplinary collaboration, it was discovered that ent-verticilide is a selective inhibitor of RyR2-mediated calcium release, including a preliminary study of efficacy in vivo. As a 24-membered cyclic depsipeptide with a molecular weight of 853 Da, ent-verticilide falls outside of the category of a traditional small molecule drug. While “Beyond Rule of 5” compounds with in vivo activity are growing in number, an understanding of their pharmacokinetics (PK) has lagged, thereby requiring new chemical tools and creative tactics to advance the field. By methodical structural changes to ent-verticilide, we have created an SAR-based feedback loop between permeability, activity, structure, and conformation. A strength of this approach is the combination of rigorous tools to study passive membrane permeability, and cardiomyocyte-based functional studies using both permeabilized and non-permeabilized cells to achieve an overall hypothesis-driven approach to discover how analogues of ent-verticilide travel through cellular membranes and ultimately target RyR2. Another strength is our positioning to prepare diverse analogues that include ring-chain variants likely to exhibit contrasting permeability. With increased understanding of the mechanism of action, we hypothesize that we can design analogues of ent-verticilide with improved potency and selectivity, thereby providing a potential therapeutic against fatal ventricular arrhythmias. Key findings of this work include that a single ester-to-N-H- amide point modification can enhance cell permeability significantly but several backbone modifications exhibit low activity and permeability; N-permethylation of ent-verticilide is critical for activity and permeability but in less critical with the ent-B1 oligomer; backbone cyclization enhances permeability; nat-verticilide is significantly less active and permeable than ent- verticilide; and several side chain or backbone modifications results in loss of activity.
Perspectives on Magma Mush Deformation: Textural and Geochemical Constraints on Compaction and Melt Extraction in a Granitic Magma Body
(2025-07-09) Teeter, Elizabeth Clare; Gualda, Guilherme; Gualda, Guilherme
Magmatic textures and compositions illuminate magma dynamics, such as compaction and melt extraction. Preserved textures in plutons (crystallized magma bodies) or magma mush fragments (crystal-rich, melt bearing clasts potentially genetically related to the erupted magmas) can provide important perspective on mush processes in both inactive and active magmatic systems respectively. The emplacement of a large (2 m diameter) felsic enclave (LFE) within granitic mush in Aztec Wash pluton (15.7 Ma, NV-USA) is preserved within a granitic body. The LFE is an ellipsoidal microgranite mass inferred to have been nearly solid when it settled at the top of a crystal-rich zone of the magma body. Backscattered Electron (BSE) imaging, Energy Dispersive Spectroscopy (EDS), Electron Backscatter Diffraction (EBSD), and X-ray Fluorescence whole-rock data for samples from the granite surrounding the LFE show key textural and compositional differences between granite immediately beneath the LFE and granite located more distant and to the side (“far-field”). Alkali feldspar crystals beneath the enclave have well-defined euhedral rims, while those in the far-field have irregular overgrowths, suggesting continuing growth into larger melt pools. The samples underneath the enclave (69.6-70.0% SiO2, 1100-1170 ppm Ba, 400-430 ppm Sr) are less felsic than the far-field sample (72.1% SiO2, 750 ppm Ba, 330 ppm Sr), also suggesting less retained melt. We conclude that impingement of the LFE led to enhanced melt extraction. The composition of the far-field sample is typical of Aztec Wash pluton samples interpreted to represent cumulate mush (70.0-72.4% SiO2, 640-1100 ppm Ba, 250-440 ppm Sr; Harper et al. 2004 and our new data). For comparison, “non-cumulate” Aztec Wash samples that may represent input magma are distinctly more felsic (72.3-73.9% SiO2, 570-690 ppm Ba, 170-270 ppm Sr). Our beneath-LFE compositions indicate the greatest melt depletion identified among Aztec Wash granites, substantially more than in the typical cumulates. We calculate that emplacement of the LFE in a magma mush enhanced melt extraction, segregating 10-20% melt from the crystal-rich framework. Magma mush fragments – crystal-rich, glass bearing clasts potentially genetically related to the erupted magmas – can provide important perspective on mush processes in active magmatic systems. Granitoid clasts from the Rotoiti Ignimbrite, Taupō Volcanic Zone (TVZ), include magma mush fragments that can provide insights on magmatic systems of one of the most active areas of rhyolitic volcanism in the world. This study focuses on a large (35 cm) clast found within a lag-breccia of the 64 ka Rotoiti Ignimbrite, one of the largest eruptions from the Ōkataina Volcanic Center (TVZ). This fragment is layered and can be divided into 5 zones based on compositions and textures. Backscattered Electron (BSE), Cathodoluminescence (CL), and Energy Dispersive Spectrometry (EDS) techniques highlight the textural and mineralogical differences between the zones. In this study, we focus on a fine-grained, biotite granite zone with microcrystalline patches and sparse granophyric texture. Quartz grains range from euhedral to subhedral, with complex zoning patterns that are visible in both BSE and CL. Some feldspar grains are zoned, others are not. The microcrystalline patches make up less than 10% of the sample and they are surrounded by subhedral grains (both quartz and feldspar). The interstitial texture and distribution of the microcrystalline patches indicate that this fragment is representative of a magma mush with <10% trapped melt. The average compositions of this microcrystalline material range from 72-77% SiO2 and 4-6% K2O and yield rhyolite-MELTS model pressures of ~100 MPa. This melt composition does not correspond with Rotoiti eruptive products; this mush fragment may illuminate non-eruptive sequences of the TVZ. Tying textures and compositions in both active and inactive magmatic systems can provide key perspective on the generation of eruptible magma bodies.
Magnetic Resonance Imaging of Physiologic Sodium (23Na) for Human Lymphatic Disease
(2025-06-11) Taylor, Shannon Leigh; Crescenzi, Rachelle L; Smith, Seth A; Crescenzi, Rachelle L; Smith, Seth A
Despite the prevalence of lymphedema, few objective tools exist to quantify the severity of lymphatic disease in a way that is sensitive to the complex physiologic changes occurring in affected tissues. Current clinical tools are not sufficient to capture disease features, especially when adipose tissue remodeling is prominent. Tissue sodium could be a relevant physiologic indicator of lymphatic disease, based on preclinical models showing that lymphatics participate in sodium storage and clearance to maintain homeostasis. However, the importance of sodium to lymphatic physiology in humans is not well-characterized, nor exploited for clinical applications, largely due to a lack of methods to observe both sodium and lymphatics together in vivo. Sodium (23Na) magnetic resonance imaging (MRI) is a translational imaging modality capable of noninvasively measuring tissue sodium. We hypothesize that sodium content is elevated in tissues that exhibit lymphatic insufficiency and tissue remodeling in human lymphatic disease. In this dissertation, we implemented sodium (23Na)-MRI together with hydrogen (1H)-MR lymphangiography and anatomical imaging in patients with localized leg lymphedema to study disease characteristics and the relationship between lymphatic function and tissue sodium. Results found that tissue sodium content, measured with 23Na-MRI, increases with lymphedema disease severity and decreases following lymph-mobilization therapy. Sodium accumulation was observed in distinct locations in the adipose tissue, suggesting a link between sodium storage, lymphatic anatomy, and tissue remodeling. We also evaluated the relationship between local tissue sodium measurements and systemic sodium excretion to offer study design guidance for future 23Na-MRI studies in lymphatic disease, further promoting the clinical translation of this technique. Altogether, this work implemented objective, physiologic imaging methods to demonstrate an anatomical and functional relationship between lymphatics and tissue sodium in human lymphedema. This research addresses a clinical unmet need of patients with lymphedema by introducing a non-invasive tool sensitive to underlying disease mechanisms and laying foundations for clinically translatable ²³Na-MRI in lymphatic disease. More broadly, this work supports the emerging role of lymphatic circulation in human diseases involving sodium storage, and it demonstrates translational imaging approaches necessary to explore lymphatic health and disease.
Impact of Amino Acids on Host Susceptibility to the Colonization and Expansion of Salmonella enterica serovar Typhimurium
(2025-07-17) Shealy, Nicolas Gray; Byndloss, Mariana X.; Zhu, Wenhan
Salmonella Typhimurium represents one of the four leading causes of foodborne illnesses worldwide. While studied for many decades as a model intracellular pathogen, there exists a gap in knowledge on the role of the gut microenvironment in influencing Salmonella colonization and expansion in the inflamed gut. One major factor influencing pathogen success is the resident collection of commensal organisms, known as the gut microbiota. The microbiota hinders Salmonella through several protective mechanisms, like sequestration of nutrients, i.e., amino acids (colonization resistance). However, which amino acids are important for Salmonella in the mammalian gut remained unknown. Using bacterial mutagenesis in combination with in vitro and in vivo studies leveraging conventional and germ-free animals, we identify and characterize the critical role of two S. Tm aspartate interconverting enzymes, AspA and AspC. We find that aspartate enables Salmonella to overcome microbiota-mediated colonization resistance by fueling anaerobic respiration during gut inflammation. Additionally, we develop a novel model for the study of colonization resistance against Salmonella in the terminal ileum, through the manipulation of dietary amino acids, overcoming host genetics-based resistance to severe disease and mortality. Lastly, we identify and characterize key alterations in pathogen metabolism that support colonization and expansion within the ileal lumen. Altogether, this work highlights the importance of amino acids in dictating Salmonella colonization, expansion, and the resultant disease severity.
Navigating the Cross and Crossing the Sea: Collaborative Knowledge-Production, Visual Literacies, and Racial Formations Across the Spanish Empire
(2025-05-23) Serratos García, Brayan; Hill, Ruth
This dissertation titled “Navigating the Cross and Crossing the Sea: Collaborative Knowledge-Production, Visual Literacies, and Racial Formations Across the Spanish Empire” puts indigenous codices, Spanish manuscripts—some translated from Portuguese and others integrating Chinese vocabularies—transcontinental correspondence, cartographic representations, illustrations, original archival research, and published literary works into conversation with each other to illuminate how colonial knowledge-production intersected each other in Spanish America and Spanish Asia during the Early Modern Period. The project underscores the interconnectedness of the early modern world and challenges the conventional narratives of colonial history by highlighting the collaborative and contested nature of knowledge creation and dissemination. This work contributes to the historical understanding and enriches the appreciation of the cultural, political, and intellectual exchanges that shaped the modern world. It is a testament to the impact of visual and written culture in bridging diverse civilizations under the expansive umbrella of the Spanish Empire, revealing a world where boundaries were both drawn and blurred through the acts of seeing, writing, and mapping.
On Behavior Trees and their Verification
(2025-07-14) Serbinowska, Serena Serafina; Johnson, Taylor T; Karsai, Gabor; Dubey, Abhiskeh; Koutsoukos, Xenofon D
Behavior trees are high level controllers that have gained popularity in a variety of safety critical domains such as robotics and medicine. Because of this, it is important that we be able to formally verify that behavior trees work as intended. To that end, we created a formal model for behavior trees called stateful behavior trees. A stateful behavior trees includes a behavior tree and the environment that behavior tree operates in. Furthermore, we created a tool named BehaVerify for the verification of stateful behavior trees. BehaVerify takes as input a stateful behavior tree specified using a domain specific language we created. The domain specific language allows for the use of invariant specifications, linear temporal logic specifications, and computation tree logic specifications. As output, BehaVerify produces a model for verification with nuXmv, a Python implementation, or a Haskell implementation. We compare BehaVerify against various competing tools and find that BehaVerify generally outperforms. We have expanded BehaVerify to use contingency runtime monitors. To create a contingency runtime monitor, a linear temporal logic specification is included in the input. Additionally, actions to be taken in the case of a violation are also included; this allows the behavior tree to respond to violations. With the rising power and popularity of neural networks, it is also important to be able to handle behavior trees that use neural networks. To that end, we expanded BehaVerify to handle neuro-symbolic behavior trees. We tested BehaVerify on neuro-symbolic behavior trees navigating a complex grid world environment and on a simplified version of ACASXu, an aircraft collision avoidance system. In the case of ACASXu, we verified a neuro-symbolic behavior tree that made use of 5 neural networks, each of which had 6 layers of 50 neurons each. We included a comparison of various methods to encode the neural networks in nuXmv.
Perception of Auditory, Visual, and Audiovisual Motion in Macaque Monkeys
(2025-07-12) Schoenhaut, Adriana; Wallace, Mark T; Ramachandran, Ramnarayan; Constantinidis, Christos
The ability to perceive moving objects, which tend to engage multiple sensory systems, is critical for navigating everyday environments. This dissertation examines how motion cues are processed and integrated across auditory and visual modalities in rhesus macaques. In Chapter 1, macaques performed a direction discrimination task using auditory motion stimuli that systematically varied in displacement, duration, and velocity. Behavioral analyses revealed that displacement had the strongest influence on motion direction sensitivity, with duration exerting a moderate effect (particularly at short timescales) and velocity contributing the least. These findings support a “snapshot” model of auditory motion perception, in which motion is inferred from sequential spatial samples, rather than encoded directly through dedicated velocity-tuned mechanisms. Chapter 2 builds on this by presenting experiments utilizing matched auditory, visual, and audiovisual motion stimuli. In most conditions, audiovisual stimulation produced multisensory gain, i.e., enhanced motion sensitivity relative to the unisensory condition eliciting the highest sensitivity. However, these effects were often suboptimal relative to predictions from the Maximum Likelihood Estimation (MLE) model, which assumes cues are integrated based on their relative reliability. These results suggest that crossmodal motion integration in macaques does not consistently follow MLE predictions and may reflect more flexible strategies, such as those described by Bayesian causal inference models. Taken together, this work provides a detailed behavioral account of motion processing and multisensory integration in a nonhuman primate model. The results also lay a foundation for future neurophysiological studies,and highlight the importance of naturalistic sensory stimulation in studies of auditory and visual perception.
Examining General Education Access for Students with Extensive Support Needs in Kindergarten Through Third Grade
(2025-07-09) Rossi, Elizabeth Burnett; Hodapp, Robert; Hodapp, Robert
Special education researchers argue about the provision of general education access for students with disabilities. In the United States, parents and school-based professionals collaboratively make two important decisions that relate to general education access. These decisions deal with ensuring a child's free and appropriate public education and the child's least restrictive environment. This study aimed to gain insight into the processes used to construct individualized educational programs (IEPs) while making these decisions. Respondents included 115 special education teachers of students with extensive support needs in kindergarten through third grade, each of whom filled out an anonymous, web-based survey about themselves, the students on their caseloads, their schools-classrooms, and school practices concerning IEP meetings. Specific information was collected about the extent of conversations around four items dealing with selecting educational placements, as well as about 15 types of supplementary aids and services. Findings revealed that the common use of draft IEPs related to greater levels predetermination (i.e., placement decisions made prior to the IEP meeting itself). Greater levels of predetermined placements, in turn, were tied to educational placements that were more restrictive. Although greater use of draft IEPs did not correlate with conversations that should be happening in IEP meetings, such general education-oriented discussions were related to greater sharing of different views about educational placements, about supplementary aids and services, and about the general education curriculum. As opposed to mandating inclusive placements, this study's recommendations aim to reform problematic practices. As such, this study's findings support additional oversight mechanisms for districts, schools and teachers, as well as free special education trainings, by state or district entities, for parents and school personnel.
SRSF2P95H splicing factor mutations impair emergency hematopoiesis and elicit context-specific splicing programs
(2025-04-17) Robb, Abigail Rich; Zinkel, Sandra S; Boothby, Mark R
SRSF2P95H (P95H) splicing factor mutations are initiating mutations in myelodysplastic syndromes (MDS) that are associated with myeloid-biased hematopoiesis in both mouse models and human disease. Mechanistically, P95H substitutions induce transcriptome-wide alterations in RNA splicing that vary by cell type and context. The relationship between physiologic, context-specific P95H splicing and the characteristics of P95H-associated hematopoietic disease is poorly understood. This study characterizes P95H-directed splicing and hematopoiesis in two different physiologically relevant contexts: myeloid differentiation and inflammation. To evaluate splicing changes, we performed bulk RNA-seq of purified bone marrow promyelocyte, myelocytes, and neutrophil populations from Vav-Cre Srsf2P95H/+ (P95H) and Srsf2+/+ (WT) mice. Cluster-based differential splicing analysis was applied to these data, RNA-seq of P95mut MDS, and P95H mouse progenitors. To evaluate changes associated with acute inflammation, we induced emergency hematopoiesis in P95H mice. P95H splicing varied by cell type, differentiation state, and inflammatory status. The most prominent conserved event across myelopoiesis, in human MDS, and during inflammatory hematopoiesis impacted Csf3r, encoding the G-CSF receptor. Notably, P95H mouse bone marrow exhibited impaired activation of G-CSFR-dependent pathways (STAT3/5) that are G-CSFR specific in myelocytes and neutrophils. Human P95mut CMML also exhibited impaired G-CSF dependent activation. During short-term inflammatory hematopoiesis, P95H mutant myelo/erythro hematopoietic progenitor marrow cells fail to expand. This inflammatory hematopoiesis program is complemented by inflammation-specific P95H directed differential splicing within the marrow affecting TNF production, regulation of mRNA stability, and neutrophil function . Our findings implicate acute inflammation as a key driver of P95H hematopoietic dysfunction which impacts both hematopoiesis and splicing.
Life of the Mined: Gender and Race in the Long Coal Century
(2025-06-11) Reich, Nicholas Tyler; Fay, Jennifer
This project examines how literary and cultural texts have mediated the global history of coal extraction through the intertwined lenses of gender and race. While much scholarship in energy humanities and environmental history focuses on material infrastructures and ecological consequences, this dissertation foregrounds literary analysis to trace how extractive capitalism becomes thinkable, desirable, and durable through narrative form, aesthetic conventions, and symbolic investments. Spanning the middle of the nineteenth century to the present, and focused primarily on the extraction zone known as “Appalachia,” this project conceptualizes “the long coal century” as a transnational and transhistorical period shaped by the cultural labor of mining fictions. Each chapter pairs literary and cinematic texts with theoretical insights from feminist, Black, Indigenous, postcolonial, and queer and trans studies to uncover how extraction structures not only economies and ecologies, but also imaginaries of personhood, community, and survival. By reading literature as both a witness to and agent of extractive modernity, Life of the Mined argues for the centrality of narrative in shaping the historical and affective contours of energy regimes, and reimagines literary analysis as essential to understanding the uneven violences and speculative futures of the Anthropocene.
Genotype-Specific Effects of Clonal Hematopoiesis of Indeterminate Potential (CHIP) on Breast Cancer
(2025-07-14) Reed, Sarah Camille; Park, Ben; Fingleton, Barbara
Clonal hematopoiesis of indeterminate potential (CHIP) is characterized by expanded blood cell clones containing somatic mutations in leukemia-associated genes in patients without hematologic malignancies. A growing body of research indicates that CHIP is associated with aberrant inflammatory signaling, and studies have identified high rates of CHIP in patients with solid tumors. CHIP has been associated with adverse outcomes in some solid tumor settings, but its impact on breast cancer remains unclear. This dissertation leverages orthogonal approaches with clinical data and mouse models to investigate the impact of CHIP on breast cancer progression and the breast tumor microenvironment. We identified a retrospective cohort of 125 patients presenting with primary breast cancer and used targeted sequencing on peripheral blood to identify CHIP. Metastatic outcomes were curated via chart review, and distant metastasis-free survival probability was analyzed. In parallel, we used bone marrow transplantation to develop novel chimeric mouse models of CHIP. Sublethally irradiated mice received mixtures of wildtype cells and CHIP-mutant hematopoietic cells representing the two most common CHIP genotypes observed in patients (Dnmt3a and Tet2). After engraftment, CHIP and control mice were injected with syngeneic breast cancer cells into the mammary fat pad. Tumor growth was measured regularly, and at tumor endpoint, peripheral blood and breast tumors were harvested for analysis of immune cell abundance via mass cytometry. Patients with high-burden CHIP (variant allele frequency > 10%) and non-DNMT3A CHIP had significantly shorter distant metastasis-free survival. In vivo, mice with Tet2-CHIP developed larger primary tumors and were more likely to experience lung metastasis, while Dnmt3a-CHIP did not differ from controls. The general immune subsets observed in both CHIP models were similar, but immunophenotyping revealed clonal expansion and immune cell subset skewing specific to the Tet2-CHIP model. This work demonstrates a genotype-specific impact of CHIP on breast cancer across human and mouse data. Further, the chimeric models described offer a clinically relevant tool to study solid tumors in a CHIP background. These results underscore the need for further functional studies and personalized risk assessment to clearly define the impact of various CHIP genotypes on solid tumors
Fabrication and Application of Thermoresponsively Patterned Microvasculature
(2025-06-27) Rector, John Albert; Bellan, Leon M
Hydrogels are a potent platform for developing medical technologies: their biocompatibility, viscoelastic structure and high-water content make them ideal for applications ranging from artificial tissue scaffolds to medicated wound dressings. Critically, they act as a medium in which water-soluble molecules, including nutrients, growth factors and cellular waste products, may move via diffusion or advection. These mass transport properties can be directly affected, without significantly compromising the mechanical or chemical properties of the polymer network itself, by creating micrometer scale cavities throughout the hydrogel. Modern methods of fabricating such features are limited by any combination of feature size, lumen geometry, scalability, interconnectivity or 3D complexity. Such limitations restrict the potential for hydrogel-based technologies, especially with regards to patterning microvasculature in engineered tissues. Using microfibers made with a nontoxic, thermoresponsive polymer, I patterned complex, tortuous, microscale channel networks in hydrogels that enable fluidic access to the full gel volume. Central to my research has been the application of 3D microchannel networks as exchange vessels, a necessary advancement on the path to artificial tissues that are not limited by diffusive transport. I patterned hydrogels using sacrificial, thermoresponsive templates to form channel networks whose size and architecture approached those of natural capillaries. Within these microchannels, I cultured endothelial monolayers that remained viable for over three weeks and exhibited functional barrier properties. Endothelialized microchannels were also cultured within hydrogels containing suspended fibroblasts, which remained viable at 14 days post-seeding, thereby demonstrating their potential for integration into more complicated artificial tissue systems. The highly anisotropic geometry of the microchannels is also advantageous for designing drug delivery vehicles, as it creates fluidic connections throughout the gel volume which can tune its transport properties while sacrificing very little solid material. By patterning interpenetrating polymer hydrogels with full-thickness microchannel networks, I created devices with accelerated drug-release times without sacrificing payload. Initial data suggests that patterned interpenetrating hydrogels loaded with therapeutics can be used to effectively treat partial-to-full thickness burn wounds. This work represents early steps in overcoming the mass transport barriers preventing the development of thick artificial tissues and the advancement of hydrogels as regenerative technologies.
Common Cause versus Dynamic Mutualism: Empirical Insights into ADHD’s Placement Within the Psychopathology Hierarchy
(2025-05-29) Peng, Zheyue; Watts, Ashley L.; Kaczkurkin, Antonia N.; Christensen, Alexander P.
Contemporary psychopathology classification systems disagree on how ADHD should be conceptualized. One system regards ADHD as an externalizing condition because it argues that it is primarily comorbid with conditions characterized by poor impulse control. Another system regards ADHD as a neurodevelopmental condition because it is primarily comorbid with conditions characterized by early onset and cognitive deficits. Despite these disagreements, both perspectives assume that ADHD’s comorbidity arises because ADHD shares a common set of causes, or shared etiology, with other forms of psychopathology. Still, emerging models posit that comorbidity arises from dynamic interactions among forms of psychopathology, a theory known as dynamic mutualism. We pitted the common cause theory against the dynamic mutualism theory in explaining the developmental ties between ADHD dimensions (i.e., ADHD, inattention, hyperactivity/impulsivity, cognitive disengagement) and three psychopathology spectra (i.e., externalizing, neurodevelopmental, internalizing). Results support the common cause theory, with the strongest evidence for the neurodevelopmental and externalizing spectra. Taken together, our findings support conceptualizing ADHD as a condition that spans both the externalizing and neurodevelopmental spectra as opposed to fitting exclusively within either category.

Browse

Recent Submissions