Digital Commons Network^™

Brain-Adapter: Enhancing Neurological Disorder Analysis With Adapter-Tuning Multimodal Large Language Models, Jing Zhang, Xiaowei Yu, Yanjun Lyu, Lu Zhang, Tong Chen, Chao Cao, Yan Zhuang, Minheng Chen, Tianming Liu, Dajiang Zhu

Computer Science Faculty Research & Creative Works

Understanding brain disorders is crucial for accurate clinical diagnosis and treatment. Recent advances in Multimodal Large Language Models (MLLMs) offer a promising approach to interpreting medical images with the support of text descriptions. However, previous research has primarily focused on 2D medical images, leaving richer spatial information of 3D images under-explored, and single-modality-based methods are limited by overlooking the critical clinical information contained in other modalities. To address this issue, this paper proposes Brain-Adapter, a novel approach that incorporates an extra bottleneck layer to learn new knowledge and instill it into the original pre-trained knowledge. The major idea is to …

Exploring The Trade-Offs: Unified Large Language Models Vs Local Fine-Tuned Models For Highly-Specific Radiology Nli Task, Zihao Wu, Lu Zhang, Chao Cao, Xiaowei Yu, Zhengliang Liu, Lin Zhao, Yiwei Li, Haixing Dai, Chong Ma, Gang Li, Wei Liu, Quanzheng Li, Dinggang Shen, Xiang Li, Dajiang Zhu, Tianming Liu

Computer Science Faculty Research & Creative Works

Recently, ChatGPT and GPT-4 have emerged and gained immense global attention due to their unparalleled performance in language processing. Despite demonstrating impressive capability in various open-domain tasks, their adequacy in highly specific fields like radiology remains untested. Radiology presents unique linguistic phenomena distinct from open-domain data due to its specificity and complexity. Assessing the performance of large language models (LLMs) in such specific domains is crucial not only for a thorough evaluation of their overall performance but also for providing valuable insights into future model design directions: whether model design should be generic or domain specific. To this end, in …

Estimation And Model Misspecification For Recurrent Event Data With Covariates Under Measurement Errors, Ravinath Alahakoon, Gideon K.D. Zamba, Xuerong Meggie Wen, Akim Adekpedjou

Mathematics and Statistics Faculty Research & Creative Works

For subject i, we monitor an event that can occur multiple times over a random observation window [0, (Formula presented.)). At each recurrence, p concomitant variables, (Formula presented.), associated to the event recurrence are recorded—a subset ((Formula presented.)) of which is measured with errors. To circumvent the problem of bias and consistency associated with parameter estimation in the presence of measurement errors, we propose inference for corrected estimating equations with well-behaved roots under an additive measurement errors model. We show that estimation is essentially unbiased under the corrected profile likelihood for recurrent events, in comparison to biased estimations under a …

Smartsla: Enabling Quality Of Service In Blockchain-Enabled Iot Networks, Kyle M. Whitlatch, Asad Waqar Malik, Sanjay Madria

Computer Science Faculty Research & Creative Works

The significant advancement in Internet of Things (IoT) adoption has enabled Multi-access Edge Computing (MEC) to mitigate IoT sensors limited computational, transmission power constraints, and data distribution overhead. However, integrating MEC with the IoT ecosystem poses several challenges, resulting in integrity issues with the MECs, impacting their capacity to effectively serve users seeking data generated by IoT sensors. To address this, we propose SmartSLA, a blockchain based solution to ensure Quality of Service (QoS) from third party IoT devices. SmartSLA leverages the decentralized and immutable nature of blockchain to combat the shortcomings of MECs. Using smart contracts, we develop a …

Analyzing Micrometeoroid And Orbital Debris Shield Performance For Sample Return Spacecraft, William P. Schonberg, Michael Squire

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Sample-return missions typically seek to collect and return samples from extraterrestrial locations to Earth for analysis. The return of samples from certain locations in the solar system represents a potential hazard to Earth's biosphere from foreign microorganisms. One way this could occur would be a break-up of the returning capsule during entry (which would disperse the samples through the air) due to undetected damage to the capsule thermal protection system (TPS). As a result, missions that plan to return payloads from some destinations are likely to have certain design requirements for the TPS surrounding their returning capsules, which may include …

Challenges In Numerical Simulation Of Frost Heave, Antai Dong, Xiong Zhang

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Frost heave of soil extensively exists in northern regions and poses a significant threat to infrastructure in cold regions. Despite over a century of research, challenges persist in numerically simulating frost heave. This study addresses two key issues: (1) What is the primary driving force for liquid water transfer during the freezing process? (2) How can we correctly represent unfrozen water content? Critical insights are derived from the theoretical analysis of coupled hydrothermal migration during soil freezing processes, followed by a case simulation using COMSOL Multiphysics. It concludes that of the water content gradient, suction gradient, and hydraulic gradient, only …

Floquet Theory For First-Order Delay Equations And An Application To Height Stabilization Of A Drone’S Flight, Martin Bohner, Alexander Domoshnitsky, Oleg Kupervasser, Alex Sitkin

Mathematics and Statistics Faculty Research & Creative Works

In this paper, we proposed a version of the Floquet theory for delay differential equations. We demonstrated that very natural assumptions for control in technical applications can lead us to a one-dimensional fundamental system. This approach allowed researchers to work with classical methods used in the case of ordinary differential equations. On this basis, new original unexpected results on the exponential stability were proposed. For example, in the equation x' (4)+a(t)x(t—-T(7)) = 0, t € [0, co), we avoided the assumption on the smallness of the product sup,j9,.) 41 SUP;< {9,00) TD) < 3/2 for asymptotic stability. We obtained that in the case of w-periodic coefficient and delay, the fact that the period w was situated in a corresponding interval can lead to exponential stability. We then applied our new tests of stability to the stabilization of a drone's flight, where smallness of the noted above product could not be achieved from a technical point of view. For an equation with periodic coefficient and delay, we got a formula of the solution's representation on the semiaxis.

Material Suppliers’ Perspective On Collaboration In Industrial Construction Projects, Seogjae Choi, William J. O'Brien

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Material suppliers' involvement in construction project planning has been recommended as an effective measure to improve project performance by using their expertise. However, the recommendation is not universal, and some researchers disputed the value of early involvement of suppliers depending on material type. Moreover, prior studies considered main contractors' and owners' perspectives mostly in examining how to increase the involvement of suppliers. To add material suppliers' perspective, this study conducted interviews with experienced professionals from 16 material suppliers including standardized, make-to-order, and custom products. The results are specified by separating the type of materials and sub-phases of planning. The results …

Forecasting State-Level Construction Labor Earnings For Enhanced Project Cost Control: An Econometric And Deep-Learning Analysis Of The Leading Economic Indicators, Ahmed Shiha, Islam H. El-Adaway

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

As a major input to several work packages, the labor element constitutes a critical component for successful performance of construction projects. Localized labor shortages, fundamental changes in prevailing wage laws, and historical shifts in the unionization rates of construction workers impair the adequate estimation of construction labor costs in diverse labor market dynamics. Meanwhile, existing studies have utilized national-level indicators to study the trends of construction labor costs, but the relationship between the multifaceted local economic factors and state-level construction labor costs remains understudied. This paper fills such a knowledge gap. A three-stage methodology is adopted: (1) data collection of …

Polarimetry Based Radar Estimation Of Extreme Rainfall: Case Studies, Bong Chul Seo, Witold F. Krajewski, James A. Smith

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

The study evaluated radar-derived polarimetric rainfall estimates for extreme rain events that occurred in the Kansas City Metropolitan area in the United States. To derive quantitative precipitation estimates (QPE), we implemented two polarimetric algorithms based on specific attenuation (A) and specific differential phase (KDP), along with the reflectivity (Z) based one using data from two radars in the study area. The analysis to assess radar-rainfall estimates (R) utilizes ground observations from a dense network of about 170 rain gauges. Based on our analysis results, the two polarimetric estimates from R(A) and R(KDP) outperform the conventional estimation R(Z). R(A) appeared to …

Individual Design Task Final Report Security Design, Ciarrah Bell

Honors Academy

"The Environmental Services Campus project is a comprehensive development designed to support the City of Springfield’s Environmental Services operations, with a focus on functionality, sustainability, and safety. The campus includes administrative office space, staff parking, a maintenance vehicle garage, and educational facilities aimed at community outreach. A component of the project is establishing site-wide security that not only protects personnel and assets but also integrates seamlessly with the architectural and environmental goals of the facility.

The primary focus of the Security IDT is the development and implementation of a comprehensive security strategy for the entire site. This includes designing and …

Lidar From The Skies: A Uav-Based Approach For Efficient Object Detection And Tracking, Baya Cherif

Masters Theses

"Recently, there has been a growing interest in deploying the Light Detection and Ranging (LiDAR) technology to gain traction in the autonomous vehicle industry, its applications are expanding into areas like smart cities, agriculture, and renewable energy. This work proposes an advanced approach to enhance aerial traffic monitoring using Li- DAR. We aim to provide accurate, real-time object detection and tracking from an aerial perspective by integrating Unmanned Aerial Vehicle (UAV) with LiDAR, culminating in a smart UAV-integrated LiDAR (A-LiD) sensor for traffic surveillance. We introduce an adapted version of one of the newest methods of the cutting-edge 3D object …

Long Short-Term Memory (Lstm) -Based Neural Network Model For Optimizing Composite Manufacturing Process Using Autoclave, Sourav P. Bolar

Masters Theses

The combination of high pressure and controlled heat plays a critical role in ensuring the uniform curing of composite materials, leading to parts with superior mechanical properties. In this study, three composite samples of IM7/CYCOM 5320-1, each cut into 12x12-inch squares, were placed in an autoclave at three different locations, spaced 6 inches apart. Sixteen thermocouples were randomly distributed across the setup to monitor the curing process as the autoclave temperature was systematically ramped up and down while maintaining constant pressure, creating a fully controlled curing environment. The primary objective was to optimize the curing locations to reduce machine runtime …

Constraint Programming For Optimized Degree Paths, Mitchell Lee Skaggs

Masters Theses

This work presents a degree planning tool developed as part of the Pervasive Cyberinfrastructure for Personalized eLearning and Instructional Support (PERCEPOLIS) project which generates complete, valid, and personalized degree paths at any point from admission to graduation. This eliminates tedious calculation and double-checking, allowing advisors to focus on a student’s long-term plans and students to proactively explore potential degree paths. The original research contribution of this work is the use of a unified model for academic requirements to automatically translate complex, real-world curricula into a constraint programming model that can be quickly optimized based on personalized student criteria.

Automatically translating …

Identifying Early Warning Signs Of Construction Labor Shortages, Ahmed Shiha, Islam H. El-Adaway

Civil, Architectural and Environmental Engineering Faculty Research & Creative Works

Construction labor shortages constrain project-level objectives and national development plans. The goal of this study is to utilize the lagged effects of macroeconomic conditions as early warning signs of construction labor shortages. To this end, the authors adopted a methodology, encompassing (1) retrieval of publicly available data and preprocessing of construction labor shortage as the target variable and macroeconomic measures as the explanatory variables, (2) identification of short-term associations between shortages and economic cycles using the Granger causality test, (3) examination of long-term relationships between labor shortages and economic conditions using the Johansen cointegration test, and (4) estimation of the …

Visual Understanding Of Rock Wettability Distribution To Contact Angle Regions And Oil Displacement Patterns In The Silty Sand Reservoir Via 2d Pore-Scale Modeling, H. Al-Ajaj, W. Al-Bazzaz, Ralph E. Flori, S. Alsayegh, H. Almubarak, D. S. Ibrahim, H. Al-Saedi

Geosciences and Geological and Petroleum Engineering Faculty Research & Creative Works

This study, conducted using a silty sand reservoir rock extracted from a Kuwait-producing oilfield, provides crucial insights into fluid distribution patterns and mineralogy, as well as wettability contact angle preferences. The method used for visual identification not only captures rock physics but also suggests effective oil recovery displacement strategies. Visual identification is presented using 2D image technology and a Scanning Electron Microscope (SEM). Analytical data are presented from electron bombardments and backscattering reflections collected in the BSE detector. These analyses were used to characterize the various surface boundary morphology parameters of the silty sand mineral surfaces, including area, perimeter, mineral …

Integration Of Physics-Informed Neural Networks And Transfer Learning For Rainfall Induced Landslide Forecasting, Shian Cao, Weibing Gong

Geosciences and Geological and Petroleum Engineering Faculty Research & Creative Works

Rainfall-induced landslides are a significant geological hazard, causing severe economic losses and casualties. Accurate forecasting of these events is particularly challenging due to the complex interactions of spatial and temporal factors governing slope stability. The Iverson model, which uses the Richards equation to describe water infiltration in unsaturated soils, is a widely adopted framework for analyzing rainfall-induced landslides. However, its reliance on traditional numerical methods limits its scalability and efficiency, particularly for complex boundary conditions and transient behaviors near slope failure. To address these limitations, we propose a physics-informed neural network (PINN) enhanced with transfer learning (TL-PINN) to solve the …

Fast Demodulation Of Ofdr-Based Distributed Sensing Based On Enhanced Buneman Frequency Estimation, Zhaopeng Zhang, Bo Liu, Xiao Liu, Caiyun Li, Osamah Alsalman, Chen Zhu

Electrical and Computer Engineering Faculty Research & Creative Works

Aiming at realizing high-efficiency distributed strain sensing through optical frequency domain reflectometry (OFDR), this paper introduces a fast demodulation algorithm to determine strain-induced spectral shifts from coarse Rayleigh backscattering (RBS) cross-correlation spectra. The proposed approach employs an enhanced Buneman frequency estimation (BFE) algorithm, enabling direct spectral shift analysis across coarse signals. By applying this algorithm, the need for dense interpolation in the conventional cross-correlation demodulation process - typically required for a finer spectral sampling interval but at the cost of demodulation efficiency - can be eliminated. Both theoretical analysis and experimental investigation reveal the equivalence of the BFE and conventional …

Analytical Model Of The Ac-Ac Dab Converter In The Egam Framework, Arnold A. Fernandes, Kartikeya Jayadurga Prasad Veeramraju, Jonathan W. Kimball

Electrical and Computer Engineering Faculty Research & Creative Works

This paper develops an analytical model of the bidirectional AC-AC Dual Active Bridge (DAB) converter. The passive components of the AC-AC DAB are subject to grid, switching, and sideband harmonics. Thus, it is impossible to model via the conventional Generalized Average Method (GAM). It has been numerically shown that Extended GAM (EGAM) can be used to model the AC-AC DAB converter. In this paper, an analytical sixteenth order EGAM-model has been developed that considers only grid harmonics at the filter components and only sideband harmonics for the transformer leakage inductor. A closed-form expression is developed for the 2D convolution product. …

Ms-Yolo: Infrared Object Detection For Edge Deployment Via Mobilenetv4 And Slideloss, Jiali Zhang, Thomas S. White, Haoliang Zhang, Wenqing Hu, Donald C. Wunsch, Jian Liu

Mathematics and Statistics Faculty Research & Creative Works

Infrared imaging has emerged as a robust solution for urban object detection under low-light and adverse weather conditions, offering significant advantages over traditional visible-light cameras. However, challenges such as class imbalance, thermal noise, and computational constraints can significantly hinder model performance in practical settings. To address these issues, we evaluate multiple YOLO variants on the FLIR ADAS V2 dataset, ultimately selecting YOLOv8 as our baseline due to its balanced accuracy and efficiency. Building on this foundation, we present MS-YOLO (MobileNetv4 and SlideLoss based on YOLO), which replaces YOLOv8's CSPDarknet backbone with the more efficient MobileNetV4, reducing computational overhead by 1.5% …

Influence Of Alloy Composition On The Process Robustness Of Steels Consolidated Via Laser-Directed Energy Deposition, Jonathan Kelley

Masters Theses

"To ensure consistent quality of additively manufactured parts, it is advantageous to identify alloys which can meet performance criteria while being robust to process variations. Toward this end, this work investigated the effect of alloy composition on the robustness of steels consolidated via laser-directed energy deposition (L-DED). Ultra-high-strength low-alloy steel (UHSLA) and pure iron powders were mixed in-situ to produce 10 compositions containing 10-100% UHSLA by mass. JMatPro material simulations roughly predicted phases and mechanical properties. Two sets of experiments were used to evaluate the sensitivity of as-built hardness (all 10 compositions) and tensile properties (5 select compositions) to process …

Computation Of Natural Relative Trajectories In The Elliptic Restricted Three-Body Problem, Dane Huck

Masters Theses

"Humankind’s presence in deep space is certain to increase throughout future decades. Future missions will continue to exploit the complex dynamics making use of liberation point orbits. Along with this trend is the rise of small satellites ("small sats") and fractionated spacecraft in general. Compared to traditional monolithic spacecraft, small sats can often have higher maneuverability, be a fraction of the cost, and be more expendable in nature. The growing interest in these two areas drives the need for improvements in the area of relative motion in deep space. One obvious use of small sats in deep space is for …

Automated Generation Of Malware Metadata Signatures, Joel Schott

Masters Theses

In advanced, targeted malware attacks, the custom software tools used to package and send malicious files and messages can lead to distinctive metadata values that facilitate creation of a malware metadata signature. Manual creation of these signatures requires expert domain knowledge and is time-consuming and error-prone. Our goal is to automate this process. We created several methods of automatically generating malware metadata signatures for ZIP files and emails. We evaluated these methods by comparing signatures generated with these methods to existing expert-created signatures. We found automated methods for ZIP files and emails that are capable of generating metadata signatures that …

Fetal Acidosis Prediction Using Attention Enhanced Convolutional Neural Networks, Anusha Adhikari

Masters Theses

This study explores the integration of spectral mixtures of fetal heart rate (FHR) and uterine contraction (UC) signals to enhance the prediction of fetal acidosis, utilizing the CTU-CHB dataset. Several classification models were trained using two distinct oversampling techniques and inputs, demonstrating that models incorporating spectral mixtures significantly outperform those using raw signals. These models, particularly when combined with convolutional neural networks (CNNs) and attention mechanisms, achieved a notable F1-score of 0.98, with the highest model achieving an area under the Receiver Operating Characteristic (ROC) curve of 0.95. The research employs a variety of techniques including short-time Fourier transform and …

Emi Mitigation, Material Characterization, Transformer Equivalent Circuit Modeling, Reza Vahdani

Masters Theses

Modern high-frequency electronic systems demand precise characterization and modeling techniques to ensure signal integrity and electromagnetic compatibility. This thesis presents three core studies focused on real-world challenges in high-speed and power electronics: EMI mitigation using 3D printed absorbers, wideband liquid dielectric characterization, and accurate transformer modeling.

The first study demonstrates a targeted approach to mitigating electromagnetic interference (EMI) in a commercial router. By using holography imaging to identify radiation hotspots, custom absorber structures were designed with commercially available materials, fabricated via 3D printing, and applied directly to emission sources. Radiated emission tests in a reverberation chamber showed up to 9 …

Ground Testing Of Cryogenic Environments In Vacuum Conditions, Lucas A. Scott

Masters Theses

The environment in the vacuum of space is unforgiving and ever-changing. Not only facing the lack of a protective atmosphere, but powerful radiation and extreme temperature ranges as well. Equipment destined for space must be hardened to withstand them. Just launching satellites that are theoretically able to survive these conditions is unfeasibly costly in the event of failure. Because of this, spacecraft must be tested in a simulated environment that can match the harsh climate in space. Therefore, it is the goal of the Gas and Plasma Dynamics Lab's (GPDL) to test and prove specific methods of recreating these conditions. …

Frequency-Tracker And Power Supply For Piezoelectric Lunar Dust Removal Actuator, Praneeth Uddarraju

Masters Theses

"This work presents a reconfigurable phase-locked loop (PLL)-based control system for the resonant excitation of piezoelectric actuators aimed at automated removal of particulate contaminants from photovoltaic (PV) surfaces—particularly in extraterrestrial environments such as the lunar surface. Regolith or lunar dust buildup on solar panels is a serious hazard to the effectiveness of energy harvesting on extended missions. By using high-frequency structural excitation and inertial forces the suggested system removes surface impurities. Tunable Sallen-Key low-pass filters for reliable feedback conditioning are used in conjunction with a digitally implemented PLL on an FPGA (XLR8 platform) to precisely lock the drive frequency to …

Performance Of Standard Medical Mllms On Ecg Image Data, Prisha Anil

Masters Theses

This work presents a structured benchmarking study of multimodal large language models (MLLMs) applied to electrocardiogram (ECG) interpretation tasks. We evaluate three representative architectures: MedGemma, HuatuoGPT-Vision, and LLaVA-Med, across progressive experimental stages involving text-only structured prompt normalization, text–image fusion with ECG plots, and full multimodal fusion incorporating time-series signals. A standardized five-section cardiology prompt was designed to enforce consistent output structure and SCP-code alignment, enabling reproducible metric computation across models. Quantitative evaluation using BERTScore, token-level F1, and diagnostic accuracy demonstrates that HuatuoGPT-Vision achieves the highest semantic and diagnostic alignment, while MedGemma exhibits superior formatting stability and reproducibility. In contrast, LLaVA-Med …

Raft-Based Polymer And Nanoparticle Materials For Traumatic Brain Injury Treatment And Diagnostics, Aaron Priester

Doctoral Dissertations

"Traumatic brain injury (TBI) is a leading cause of death and disability worldwide. Neurodegenerative diseases that develop post-TBI can be, in part, attributed to DNA and cell-damaging reactive oxygen species (ROS) and lipid peroxidation products (LPOx). This thesis focused on overcoming the limits of current TBI material treatment approaches by employing a RAFT (reversible-addition fragmentation chain transfer) polymer approach that incorporated novel therapeutic, diagnostic and peptide-targeting monomers. An improved nanoparticle synthesis approach was also developed. Thiol and thioether-containing monomers neutralize both ROS and LPOx while Gd-containing monomers with enhanced magnetic resonance imaging (MRI) contrast provide diagnostics and material tracking in …

Development Of Dual-Scan Nuclear Magnetic Resonance (Nmr) Pulse Programs For Spin-Lattice Relaxation Measurements, Zachary Mayes

Doctoral Dissertations

This research develops and refines the Split Inversion Pulse and Recovery (SIP R) methodology for spin-lattice relaxation (T₁) measurements in NMR spectroscopy. SIP R introduces a two-scan difference technique using a split inversion pulse sequence, where the 180° pulse is split into two 90° pulses phase-shifted with respect to each other. This approach simplifies data fitting, requiring only two parameters to extract T₁ values, compared to the traditional inversion-recovery method, which needs three. The SIP-R-DS and SIP-R-S adaptations extend this framework by incorporating selective NMR resonance excitations, enhancing its applicability to cross-polarization experiments and enabling unobstructed NOE measurements. These adaptations …