Digital Commons Network^™

Adversarial Robustness In Biomedical Time-Series Models, Rohan Tiwari

Bioengineering Theses

This study investigates adversarial vulnerabilities in deep learning models for biomedical time-series classification across two clinically important modalities: electrocardiography (ECG) and electroencephalography (EEG). Using the MIT-BIH Arrhythmia and CHB-MIT seizure datasets, I evaluate time-domain attacks (FGSM, PGD), Fourier-domain constrained attacks, and learned spectral perturbations designed to reveal modality-specific sensitivity patterns. Across both tasks, a consistent trend emerges low-frequency components (0–5 Hz) constitute a dominant axis of adversarial vulnerability, with perturbations in this range producing the steepest degradation in classification performance. In ECG models, protecting the physiologically relevant QRS band (5–20 Hz) significantly improves robustness, whereas EEG models remain highly sensitive …

Distilling The Complexity Of Agent-Based Simulations Into Textual Explanations Via Large Language Models, Noé Y. Flandre, Philippe J. Giabbanelli

VMASC Publications

Communicating the design and results of agent-based models (ABMs) to subject matter experts is challenging, which hinders participation and limits trust in simulation-based decision support. Large language models (LLMs) can communicate ABMs as textual summaries, thus complementing traditional disclosure through statistical and visualization techniques. While prior work translated the structure of conceptual models into narratives via LLMs, our extension covers the dynamics of simulation models via an automated simulation-to-text method that extracts contextual information from NetLogo ABMs, performs repeated simulations, and generates narrative descriptions (including the model’s purpose, parameters, and simulation dynamics) using mutimodal LLMs. Furthermore, four summarization algorithms spanning …

Advancing Task-Oriented Dialog Systems: Scalability, Generalization, And Evaluation, Adib Mosharrof

Theses and Dissertations--Computer Science

Task-oriented dialog (TOD) systems enable conversational interfaces for complex tasks like flight booking and restaurant reservations. However, deploying TOD systems at scale faces three critical barriers: scalability, generalization, and evaluation. Scalability is primarily restricted by the human-annotation bottleneck, as current systems depend on vast quantities of manually labeled data for every new domain, making deployment prohibitively expensive. Generalization remains a persistent challenge, as systems optimized for known domains often suffer significant performance degradation when encountering new, unseen ones. Existing evaluation metrics measure response quality and fluency, but fail to measure functional task success. As TOD systems are deployed across diverse …

Evaluating The Robustness Of Gnn-Based Vulnerability Detectors Under Semantics-Preserving Code Obfuscation, Jesse Ks Chumo

Computer Science and Engineering Theses

Graph neural network–based vulnerability detectors are typically evaluated on clean benchmark datasets, yet real-world code frequently undergoes semantics-preserving transformations such as identifier renaming, dead-code insertion, and control-flow restructuring. The extent to which such transformations affect detector reliability remains insufficiently understood. We evaluate ten vulnerability detectors from four architectural families across the Devign, Big-Vul, and DiverseVul datasets. To quantify robustness, we evaluate each model at three transformation budgets: one transform, two transforms combined, and all three together, finding that token-based models degrade under identifier renaming and compound transformations, while models that read only code structure are largely unaffected. We further evaluate …

A Productivity Rate-Based Comparative Carbon Footprint Cost Analysis Of Small To Large-Sized Open-Cut Pipeline Installation Activities For Sanitary Sewerage Construction: A System Boundary Concept, Amir Reza Zakeri

Civil Engineering Theses

Underground sanitary sewer pipelines are essential components of urban infrastructure; however, open-cut pipeline installation requires excavation, bedding preparation, pipe placement, backfilling, embedment, and compaction activities that rely heavily on construction equipment and fuel consumption. As a result, open-cut installation can generate measurable greenhouse gas emissions during the construction phase. With increasing attention to sustainable infrastructure delivery, there is a need for a consistent approach to quantify construction-phase carbon footprint and convert those emissions into a comparable economic indicator. Accordingly, this thesis aims to create and apply a productivity rate-based calculation framework for estimating and comparing construction-phase CO₂e emissions and carbon …

Towards Application-Driven Optimal Memory And Storage Management, Venkata Naga Prajwal Challa

Computer Science and Engineering Dissertations

Modern computing systems increasingly run on diverse hardware platforms and support applications with widely different access patterns, performance goals, and data lifecycles. In this setting, traditional one-size-fits-all approaches to memory and storage management are often inefficient because they apply fixed policies regardless of application behavior, workload context, or hardware asymmetry. Such generic designs can lead to unnecessary data movement, wasted bandwidth, excessive rewriting, poor resource utilization, and degraded user-perceived performance. This dissertation is motivated by the view that optimal memory and storage management should be application-driven: instead of treating all data uniformly, systems should adapt their decisions to how applications …

Development Of Multimodal Measurements And Analysis For Early Detection Of Alzheimer’S Disease, Fiza Saeed

Bioengineering Dissertations

Alzheimer's disease (AD) is the leading cause of dementia, and existing diagnostic methods such as PET scans, cerebrospinal fluid sampling and biomarker quantification, and gene sequencing are all either invasive, costly, or not sensitive enough for early detection. This dissertation introduces three different studies that develop a novel multimodal, non-invasive approach to diagnosing AD at its early stages by combining broad band near infrared spectroscopy (bbNIRS) and electroencephalography (EEG) technologies.

The first study showed cerebrovascular-cerebrospinal fluid coupling (CBV-CSF), which is measured by using 2-channel bbNIRS as an indicator of brain aging and early AD. Linear correlations between total blood (Δ[HbT]) …

Aeroelastic Simulation Of Shape Adaptive Wing, Allan Alfred Kozich Iii

Honors Undergraduate Theses

Morphing wings provide aerodynamic qualities that normal fixed wings cannot, such as the ability to improve endurance yet maintain maneuverability, overcome strong gusts, vibrations, and shocks, and handle both ideal flight for both high and low speeds. A critical application of the new generation of morphing wings is the ability to overcome and affect the onset flutter, a self-excited oscillatory instability that has led to the destruction of aircrafts. This work investigates aeroelastic behavior and measurement of a meta-material structured "smart" wing and its attempt to delay the effect of flutter. The variable wing tip model is analyzed through finite …

Modeling Rank Distribution And The Relative Importance Factor Index In Discrete Power-Law Models: Application To Social Resilience Using The Scopus Database, Brian Llinas, Jose Padilla, Humberto Llinas, Erika Frydenlund, Katherine Palacio

VMASC Publications

Prior research on power-law distributions has primarily focused on modeling frequency patterns, with less attention given to rank distributions and how ranked positions reflect relative importance among elements. In discrete power-law distributions, frequency-based metrics often provide limited discrimination in the tail, where elements may exhibit similar counts but differ in relative dominance. These patterns are especially evident, for instance, in academic publishing, where keywords, affiliations, and citations commonly exhibit power-law behavior. To address this limitation, we introduce the Relative Importance Factor (RIF) Index, a statistical measure derived from the estimated discrete power-law rank distribution rather than an additional independent parameter. …

Exploring Undergraduate Women’S Experiences Of Collaborative Learning In Engineering Education: A Phenomenological Study Of The Social World, Sandra Ireri Cruz Morena

Doctoral

The persistent underrepresentation of women in engineering education remains a concern across global and national contexts, despite rising overall enrolment of women in higher education. In Ireland, the national average of women entering Engineering, Manufacturing, and Construction (EMC) programmes has gradually increased; however, the rate at Technological University Dublin (TU Dublin) has stayed at approximately 15% since 2016. This study seeks to better understand the gendered challenges underlying such trends by exploring the lived experiences of female undergraduate engineering students at TU Dublin, particularly in relation to collaborative learning environments such as project- and problem-based learning (PBL).

Impact Of Bubble-Induced Turbulence On Two-Phase Flow Dynamics At High Void Fraction In A Large Diameter Channel, Sungje Hong, Joshua P. Schlegel, Subash L. Sharma

Nuclear Engineering and Radiation Science Faculty Research & Creative Works

This study represents the first investigation into the influence of bubble-induced turbulence (BIT) on interfacial area transport mechanisms in gas–liquid two-phase flows under conditions of high void fraction and high velocity in a large diameter channel. Given the unique characteristics of bubble flow in larger channels, the turbulent effects induced by bubbles differ from those observed in smaller channels. However, limited research exists regarding the impact of BIT beyond bubbly flows in large-diameter channels. To address this gap, two approaches for implementing the BIT model are explored: a direct method and an indirect method. This paper assesses both the general …

Electronic Bills Of Lading And Blockchain Technology: A Regulatory Perspective, Hsin-Hua Tsai

Journal of Marine Science and Technology–Taiwan

Information and communications technology (ICT) systems are increasingly relevant to the maritime sector. Regarding the revolution of the bill of lading, many shipping lines and other technology startups have been looking into developing viable electronic transport documents that have the same functions as paper bills of lading to reduce supply chain risks. The limitations of the paper-based process have become increasingly visible in the COVID-19 crisis. The main purpose of the article is to analyze the application of blockchain bills of lading, which utilizes computational logic to create the digital ledger, and users can set up algorithms and rules to …

Autonomous Uav Mission Planning Under Threat Using Model Predictive Control With Proportional-Navigation Pursuers, Mehmet B. Ozcelik

Mechanical and Aerospace Engineering Theses

Autonomous unmanned aerial vehicles (UAVs) operating in contested environments must

complete mission objectives while avoiding restricted regions, radar exposure, and pos-

sible interception. This thesis develops a MATLAB-based simulation framework for

two-dimensional UAV mission planning under threat using model predictive control and

proportional-navigation chasers. The mission requires the UAV to travel from a start

location to a goal while visiting required checkpoints and avoiding no-fly zones and radar

regions. A chaser attempts to intercept the UAV using either a basic pure-pursuit-style

law or a proportional-navigation guidance law.

The framework integrates environment generation, augmented visibility-graph rout-

ing, waypoint management, UAV kinematic …

Llm-Driven Closed-Loop Uav Control With Obstacle-Aware Model Predictive Control, Halimcan Yasar

Mechanical and Aerospace Engineering Theses

This thesis presents a closed-loop control architecture for uncrewed aerial vehicles (UAVs) in which a large language model (LLM) serves as a high-level decision module operating over a persistent, metric 3D world model.

Rather than generating low-level commands or open-loop plans, the LLM selects one parameterized maneuver per decision step from a small, verified library of flight primitives conditioned on a structured representation of the drone state, tracked object positions, and mission specification.

Translational motion is executed by a planar model predictive controller (MPC) with soft obstacle avoidance, using obstacle hypotheses provided by the LLM, so that safety-critical constraint handling …

The Composition Of Gases From A Diffusion Flame Above Longleaf Pine Needle Fuel Beds, David R. Weise, Thomas H. Fletcher, Timothy J. Johnson, Weimin Hao, Russell G. Tonkyn, Catherine A. Banach, Javier Palarea-Albaladejo, Mahsa Alizadeh, Stephen Baker

Faculty Publications

The gas and tar composition of a wildland fire diffusion flame from longleaf pine needles is currently relatively unmeasured and more data are needed to fill in the gap between pyrolysis data and smoke plume data, thus improving physical and chemical modeling of wildland smoke formation. A pilot experiment to measure light gas and tar composition of such a flame is described for three flame zones: persistent flame (flame base), intermittent flame, and smoke plume. Flame gases from 24 experimental fires were collected in canisters and analyzed for CO2, CO, H2, CH4, and C2 to C7 hydrocarbon gases. Other light …

Efficiently Stowed Flashers With Uniform-Thickness Panels Based On Non-Developable Origami, Larry L. Howell, Robert J. Lang, Specer P. Magleby, Davis Wing

Faculty Publications

We present the mathematical construction of a deployable flasher pattern with near-uniform nonzero thickness in the deployed form and efficient face-to-face packing in the stowed form. We demonstrate its fabrication, deployment, and stowage. The results facilitate the design and manufacture of deployable systems with panels whose substantial thickness is dictated by the end application, with particular utility for space systems.

Buckling Analysis Of Auxetic Composite Laminates And Optimal Design Using Lamination Parameters And Machine Learning, Hans Bendon Maria Tamil Selvan

Mechanical and Aerospace Engineering Theses

Composite materials are widely used as structural panels in aerospace, automotive, and civil engineering applications, where buckling is often a critical failure mode. This thesis focuses on the analysis and design of composite laminates that maximize buckling performance under prescribed stiffness and thickness constraints.

The first part of the study investigates the buckling behavior of auxetic laminates, which exhibit a negative Poisson's ratio. While previous studies suggest that auxetic laminates can achieve higher critical buckling loads than non-auxetic laminates under simply supported boundary conditions with lateral restraint, the influence of other boundary conditions and plate aspect ratios has not been …

Sewer Pipe Condition Assessment Using An Ensemble Machine Learning Framework For Infrastructure Decision Support, Mahnaz Rouhi

Civil Engineering Dissertations

Aging wastewater infrastructure presents significant challenges for municipalities across the United States, with many sewer networks approaching or exceeding their design life. Conventional inspection methods, such as closed-circuit television (CCTV), are limited by subjectivity, cost, and inefficiency. To address these challenges, this study develops an ensemble machine learning framework for assessing the structural condition of sewer pipelines using inspection records enriched with geospatial attributes.

The primary objective of this research is to enhance predictive accuracy, interpretability, and decision-support for risk-based asset management. The scope of the study encompasses 4,802 CCTV inspection records from Dallas, TX, and Tampa, FL, integrating physical, …

Videoscoop: A Non-Traditional, Domain-Independent Framework For Video Analysis, Umme Hafsa Billah

Computer Science and Engineering Dissertations

Due to the proliferation of cameras in handheld devices and the widespread use of CCTV, images and videos have become a preferred alternative for capturing and disseminating information. Automated analysis for understanding image or video contents (e.g., objects, activities, backgrounds, situations of interest, etc.) is critical for many applications such as Civic Monitoring, Surveillance (in general), monitoring activities in Assisted Living environments, and many more. Image and Video Analysis (IVA) research has been ongoing for several decades, resulting in numerous techniques for algorithmically analyzing and understanding image and video contents.

Image Analysis (IA) has advanced in several areas, including object …

Deep Learning For Wireless Communications, Swarada Ajit Kulkarni

Electrical Engineering Dissertations

The rapid evolution of wireless communication imposes stringent requirements for ultra-reliable, low-latency transmission in dynamic, interference-prone environments. Traditional model-driven signal processing struggles to adapt to nonlinear hardware effects, time-varying channels, and complex interference patterns. Deep learning (DL) offers a transformative, data-driven alternative, enabling end-to-end optimization and robust adaptation under uncertain propagation conditions.

This dissertation investigates deep learning architectures for intelligent and resilient wireless communication through three complementary contributions. The first introduces a Vision Transformer (ViT)-based modulation classification framework that leverages self-attention to capture local and global dependencies in spectrogram representations of Quadrature Amplitude Modulation (QAM) signals. The ViT achieves superior …

A Transdisciplinary Approach To Advancing Water Security And Public Health: The “Nurse + Engineer” Model For Achieving Sustainability In Wash Systems Using Community-Based Participatory Research, Daniel B. Oerther, Sarah Oerther

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Human health is linked to the health of the environment, a concept referred to as planetary health. As environmental challenges become increasingly complex, single-disciplinary approaches are insufficient to achieve sustainable solutions. This article presents a transdisciplinary framework, termed the "Nurse + Engineer" model, as a novel methodology for addressing complex problems at the nexus of environmental engineering and public health. The model integrates the technical design and systems-level thinking of environmental engineering with the community-based, patient-centered care of nursing. This work details the key lessons learned from the retrospective application of this framework to a series of multi-year, community-based participatory …

Environmental Engineers Develop Solutions To Problems Of Planetary Health: The Legacy Of Professor P. Aarne Vesilind, Daniel B. Oerther

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

The principal ethical obligation of the engineering profession, to "hold paramount the health, safety, and welfare of the public," has been challenged for decades. Professor P. Aarne Vesilind identified two primary deficiencies: an anthropocentric bias that values nature only instrumentally and the use of non-mandatory, aspirational language for environmental protection, which resulted in the exclusion of enforceable environmental canons. This paper argues that two recent developments provide a comprehensive response to Vesilind's recommendations for engineering ethics. First, the United States Bureau of Labor Statistics has redefined environmental engineering as "developing solutions to problems of planetary health." Second, Oerther proposed a …

Fairness Without Demographic Attributes In Medical Vision–Language Models, Ahsan Habib Akash

Graduate Theses, Dissertations, and Problem Reports (ETD)

Vision–language models (VLMs) and the embedding models that underlie them have become the default representation layer for multimodal artificial intelligence. Trained by contrastive alignment over enormous, loosely curated image–text corpora, they inherit the demographic skew of that data, and they encode it in the geometry of the shared embedding space itself. The consequence is a model that performs unevenly across demographic groups even when no protected attribute is ever supplied as an input. In consumer applications this is an equity problem; in medicine it is a safety problem, because an embedding that is systematically less discriminative for one subpopulation translates …

Cats 2.0: Leveraging Large Language Models And Graph Databases For Robust Arabic Sms E-Commerce Systems, M. Samir Abou El-Seoud, Daoud M. Daoud 2

Computer Science

Arabic SMS-based e-commerce platforms pose unique challenges due to the spontaneous and noisy nature of user-generated text (e.g., abbreviations, dialectal Arabic, or “Arabizi” transliterations). In this paper, we present Classified Ads Text Service (CATS) 2.0, an improved classified ads system that combines probabilistic large language models (LLMs) with deterministic graph-based knowledge representations to achieve robust understanding and matching of Arabic SMS content. Building on earlier work that emphasized the importance of integrating sublanguage analysis with content-oriented methods, our approach uses a hybrid pipeline: an LLM interprets free-text messages and extracts structured information, which is then inserted into a Neo4j graph …

Novel Methods For Environmental Fluoride Measurement, Cable Warren

Chemistry & Biochemistry Dissertations

Fluoride analysis has been an important focus of analytical analysis for many years and will continue to be so going forward. Today, fluorinated compounds, specifically per/polyfluoroalkyl substances (PFAS), better known as “forever chemicals” have captured the moment and are the subject of vast research and regulation. Under this context, I have researched new methods of fluoride separation and concentration in complex media as well as a novel method of PFAS destruction and total organic fluorine analysis for screening of PFAS in aqueous samples. Through research and work on micro-scale detection methods, an understanding of the state of the art and …

Climate And Seasonal Influences On Indo-Pacific Sea Level: Insights From 30 Years Of Satellite And Tide Gauge Data, Karlina Triana

ASEAN Journal on Science and Technology for Development

We analyze Indo-Pacific sea-level variability using monthly satellite altimetry (1993–2025) validated against records from 16 tide-gauge stations. Cross-comparisons show strong agreement, especially in the western Pacific, confirming the reliability of altimetry for regional assessments. Seasonal SLA variability is largest in the Bay of Bengal and South China Sea and reflects monsoonal forcing, whereas interannual fluctuations in the eastern Indian and western Pacific oceans are dominated by ENSO and modulated by PDO. Harmonic decomposition isolates annual and semi-annual cycles, and an EOF/PCA framework identifies the leading modes: EOF1 (30.8%) captures basin-scale interannual variability and EOF2 (20.9%) reflects the seasonal cycle. Spectral …

Metagenomic Polymorphic Toxin Effector And Immunity Profiling Predicts Microbiome Development And Disease-Related Dysbiosis, Hunter W. Schroer, Francesco Beghini, Juan Antonio Raygoza Garay, Nicholas A. Christakis, Dustin E. Bosch

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Bacteria use antagonistic interbacterial weapons, such as polymorphic toxin secretion systems (TSS), to compete for niches in the human gut microbiome. We hypothesized that TSS influence gut microbiome development and disease-related dysbiosis. We developed a bioinformatic marker gene approach (PolyProf) to quantify TSS including ~200 effector and immunity genes and applied it to ~15,000 publicly available human metagenomes. PolyProf alpha and beta diversity readily distinguished 12 different human disease states and enabled the construction of highly accurate linear regression classifier machine learning models. Elastic net machine learning models integrating bacterial taxonomy with PolyProf had strong predictive value for 12 disease …

Sme Ai Outreach In Finland—A Case Study, Kaj Mikael Björk, Anton Akusok, Amaury Lendasse, Leonardo Espinosa-Leal

Engineering Management and Systems Engineering Faculty Research & Creative Works

This paper presents a project (work in progress) where entrepreneurship and higher education in AI (from Master level to postdoc level) are integrated in order to produce a dual effect; helping SMEs to gain insight in how AI can aid in the corporate environment and to expose AI researchers to the real-life situations in the company world. If successful, the companies are made ready for the AI revolution and the researchers more equipped for corporate settings. The project is ongoing, so this paper addresses a work-in-progress project. The paper reflects on the project as well on some aspects that need …

Design Of Novel Gating Systems For Steel Castings, K. Balasubramanian, Laura Bartlett, M. Xu

Materials Science and Engineering Faculty Research & Creative Works

Gating systems play an important role in determining the quality and mechanical properties of castings. To understand the efficiency of gating systems, four systems, namely pressurized system, non-pressurized system, naturally pressurized system with a side riser and a naturally pressurized system with a top riser, were studied. The naturally pressurized systems were provided with overflows which collected the incoming metal swirl. Parameters like velocity of metal flow, air entrapment, microporosity and Niyama criterion were considered, and a design was developed with a common pouring basin. 8630 alloy was poured into two molds using a teapot ladle. The inclusion analysis revealed …

Microstructure And Properties Of Oxide Dispersion-Strengthened Alloys, Ertugrul Demir, Seung Min Ha, Anish Ranjan, Xingshuo Zhang, Aaron Penders, Mukesh Bachhav, Xiaochun Li, Lin Shao, Alexander Demblon, Haiming Wen, Enrique Lavernia

Materials Science and Engineering Faculty Research & Creative Works

Oxide dispersion-strengthened (ODS) alloys are a critical class of structural materials for extreme environments, owing to their unique combination of high-temperature strength, thermal stability, and radiation tolerance, enabled by a very high density of nanoscale oxide dispersoids. These features make ODS alloys attractive for advanced nuclear systems, aerospace applications, and other harsh-service conditions where conventional alloys rapidly degrade. Despite decades of development, key challenges remain in understanding how nanoscale oxides interact with matrix microstructures, alloy chemistry, and irradiation-induced defects to control macroscopic performance. This review provides a focused, mechanism-based synthesis of the microstructural features that govern the properties of ODS …