Digital Commons Network^™

Bridging Images And Language In Radiology: A Comprehensive Prisma Systematic Review Of Transformer Vision-Language Models And Clinical Readiness, Mohammad T. Khasawneh Dr., Sadaf Tabatabaee

Systems Science and Industrial Engineering Student Scholarship

Transformer vision-language models (VLMs) promise end-to-end automation of radiology reporting and related multimodal tasks. However, evidence remains fragmented across datasets, architectures, evaluation practices, and levels of clinical validation, limiting fair comparison and safe translation into practice. Following PRISMA 2020/PRISMA-S, search engines including PubMed, IEEE Xplore, Web of Science, and Google Scholar were systematically searched for peer-reviewed, English-language studies published between 2019 and 2025 that used paired radiology images and free-text reports. Dual reviewers screened records and extracted data using a locked schema covering datasets, modalities, architectures, training objectives, evaluation metrics, and indicators of clinical readiness. Free-text model descriptions were normalized …

Exploring Maximal Length Cellular Automata To Generate Primitive Polynomials In Gf(2), Sumit Adak, Subhrajit Deb, Anurag Ghosh, Angshuman Roy, Souvik Roy

Northeast Journal of Complex Systems (NEJCS)

We present a simple method that uses cellular automata (CAs) to find primitive polynomials over GF(2). We used maximal length CAs as tools to generate primitive polynomials. It is usually very difficult to find maximal length CAs or primitive polynomials since they require exponential time, and there is no linear time method. However, in our work, given an n-size specific sequence of CA with reasonable probability, our technique computes a cycle of length at most 2^n-1 (maximal length) in O(n) time. The characteristic polynomials of synthesized maximal length CAs are claimed to be primitive since it was previously established that …

Decision Making At Triage Classification Using Svm With Smote Technique, Mehanas Shahul, Pushpalatha Kp

Northeast Journal of Complex Systems (NEJCS)

The efficient functioning of triage gates in overcrowded emergency departments (EDs) occurs in the context of the complex adaptive system (CAS) framework, where diverse system elements – patients, medical personnel, resources, patients’ inflow patterns, and patients themselves – simultaneously and dynamically influence the decision process. This study addresses the automated incorporation of machine learning triage algorithms as part of the system triage process to support automated classified risk-level recognition based on a limited set of vital signs. Patients are dynamically subsumed under high and low-risk categories enhanced by sensitivity, which enables optimal diagnosis and triage response to the critical clinician …

Uncovering Discrete States From Multimodal Psychophysiological Data Using Gaussian Latent Dirichlet Allocation (Glda), Congyu Wu, Aaron Fisher, David Schnyer

Northeast Journal of Complex Systems (NEJCS)

In this article we explore and validate the utility of an unsupervised probabilistic model, Gaussian Latent Dirichlet Allocation (GLDA), for discovering discrete states from repeated, multimodal psychophysiological samples collected from multiple individuals. Psychology and medical research heavily involves measuring potentially related but individually inconclusive variables from a cohort of participants to derive diagnosis, necessitating clustering analysis for state identification. Traditional probabilistic clustering models such as Gaussian Mixture Model (GMM) assume a global mixture of component distributions, which may not be realistic for observations from different patients. The GLDA model borrows the individual-specific mixture structure from a popular topic model Latent …

The Reliability Of Yolo Object Detection On The Amd Versal In A Proton Radiation Environment, Jacob D. Brown

Theses and Dissertations

Object detection is an important operation for satellites to be able to perform, and in outer-space missions it is crucial that machine learning models perform inference on satellite images accurately and reliably. Where size, weight, power, and other constraints exist, meeting this goal for accurate and reliable object detection is challenging. Additionally, soft errors caused by radiation further disrupt and degrade the operation of object detection in satellites. This thesis studies the performance of a deep learning model on an embedded device, the AMD Versal, in the presence of soft errors. The well-known and high-performing YOLO convolutional neural network was …

Real-Time Fraud Detection, Haidi Aly Fahmy, Abdelhadi Nait-Zerrad, Shiva Krishana Reddy Ravuula, Sangwhan Cha

Harrisburg University Other Works

Financial fraud detection is a high-volume, high-velocity analytics problem. Traditional rule-based systems are often easy to deploy, but they are limited by static thresholds, delayed response, high false-positive rates, and weak explainability. This report presents a formalized end-to-end Big Data architecture for real-time fraud and anomaly detection in financial transaction streams.

The proposed architecture ingests transaction events through AWS Kinesis, enriches them through an Apache Flink stream-processing layer, scores them with an XGBoost classifier, explains model outputs using SHAP, and converts structured evidence into human-readable summaries through a controlled GPT explanation layer. Results are persisted through a hybrid storage strategy …

Sentience, Sheetal Agrawal

Masters Theses

The increasing urgency for sustainable and adaptive systems has driven research toward embedding intelligence directly into materials rather than relying solely on external sensing and control systems. This thesis explores how smart material1 embedded systems can be designed to recognize and respond to environmental signatures, defined as measurable patterns such as temperature fluctuations and mechanical forces. Central to this investigation is the integration of shape memory alloys, particularly Nitinol, with geometry-based actuation mechanisms that amplify material behavior into functional system responses.

The central argument is that designing with smart materials is a design problem, not primarily a materials science problem …

Somatic Prosthetics For Planetary Resonance, Disha Dharesh Kumar R

Masters Theses

This thesis investigates how industrial design can transcend extractive technological paradigms in favor of relational interfaces that foster planetary attunement. Framing the climate crisis as a 'crisis of imagination', the research challenges the Western bifurcation of nature and culture by drawing on Indic cosmologies- which recognize stones, plants, and ecosystems as conscious at different levels and participants in a shared cosmic field. By synthesizing research in Biosemiotics, Quantum Information Pansycishm, and Neuroscience, the project redefines intelligence as a distributed, more-than-human phenomenon.

The research materializes as a speculative design artifact: a device that functions as a somatic prosthetic for planetary resonance. …

Material Costs, Karima Weinman

Masters Theses

This thesis investigates how migration fatality and disappearance data can be reinterpreted through material craft to create a more reflective encounter with information. Working with the Missing Migrants Project's dataset, this project asks how design can communicate dimensions of human loss that conventional data visualization cannot reach.

The work situates contemporary border violence within a longer colonial history, arguing that the logics of surveillance and quantification that structured European imperial expansion persist in the databases that govern mobility in the Mediterranean today.

Terrazzo is a 15th-century Venetian flooring technique built from discarded fragments bound together into a unified surface. This …

Power Consumption Prediction And Energy Forecasting Using Machine Learning Models, Sheik Mohideen Shah S Mr

Theses and Dissertations

Power consumption trends are essential to be identified in the energy grid areas to analyze the utilization, deficiency, and the measures to be taken for an effective and comfortable usage of energy. There are two scenarios in which the power consumption can be analyzed, namely identification and prediction. Identification deals with the post-utilization analysis of energy trends, whereas prediction deals with prior analysis of various factors of energy utilization, including the cost, supply details, shortages, and the need for new energy resources. In the existing models, the power consumption-related data are collected through smart meters, and the energy forecasting methods …

The Analytic Hierarchy Process As A Tool For Architectural Quality Assessment: A Case Study On Municipal Facades, Gül Şebnem Tutal, Fatma Demet Aykal

HBRC Journal

This study evaluates the façade quality of three municipal buildings using the Analytic Hierarchy Process (AHP) to identify strengths, weaknesses, and improvement priorities in public architecture. Although façade design has been widely studied, quantitative and integrated assessment frameworks for municipal building façades remain limited. To address this gap, the study integrates the Design Quality Indicator (DQI) with AHP within a Multi-Criteria Decision Making (MCDM) framework, proposing a systematic and replicable evaluation model. The research assesses three main DQI criteria-functionality, building quality, and impact-subdivided into twelve sub-criteria and analyzed through pairwise comparison matrices. Data were collected through structured expert interviews with …

The Structural Limitations Of Mining Output Growth In Nevada, Mckinley Brown, Emma Kitchen

Undergraduate Economics Working Paper Series

This study compares short and long trends in Nevada to the production output within the mining industry. Nevada produces around 70% of the United States gold and about 15% of the world's critical minerals. From its position as one of the most resource rich mining regions globally to its emergence as a growing economic hub, mining played a central role in Nevada's economy. However, despite the industry’s importance, Nevada still faces inefficiencies in production output due to insufficient attention to key inputs such as labor, infrastructure, technology, and energy resources.

Robust Two-Stage Mimo-Ofdm Channel Estimation Method Against Sensing Errors, Yi Peng, Jun Wang, Qingqing Yang, Jianming Wang, Hui Li

Journal of System Simulation

To address the challenges of performance degradation, high pilot overhead, and high computational complexity in traditi onal channel estimation methods for integrated sensing and communication (ISAC) assisted MIMO-OFDM systems when radar sensing information contains errors, this paper proposes a robust two-stage sparse channel estimation framework designed to be tolerant of sensing errors. In the first stage, a residual energy weighted simultaneous orthogonal matching pursuit (REW-SOMP) algorithm is designed. Leveraging locally adaptive dictionary expansion and a residual- weighted path selection mechanism, it accurately captures communication-associated paths even under sensing errors. The second stage introduces an adaptive penalty factor alternating direction method …

Asymmetric Opinion Formation Of Emotional Excitable Agents, Irene Ferri, Emanuele Cozzo, Aleix Nicolás-Olivé, Albert Díaz-Guilera, Luce Prignano

Northeast Journal of Complex Systems (NEJCS)

The bounded confidence model represents a widely adopted framework for modeling opinion dynamics wherein actors have a continuous-valued opinion and interact and approach their positions in the opinion space only if their opinions are within a specified confidence threshold. Here, we propose a novel framework where the confidence bound is determined by a decreasing function of their emotional arousal, an additional independent variable distinct from the opinion value. Additionally, our framework accounts for agents' ability to broadcast messages, with interactions influencing the timing of each other's message emissions. Our findings underscore the significant role of synchronization in shaping consensus formation. …

Build Digital Annual Survey 2025 And Trends & Comparisons 2022 - 2025, Clare Eriksson, Robert Moore, Bilal Succar

Reports

The Build Digital Annual Survey 2025 provides an overview of digital transformation across Ireland’s construction and built environment sector, drawing on responses from 205 participants across industry, government, and academia. The findings show continued sector engagement with digital transformation, driven by government mandates, project benefits, and partner expectations. Common digital deliverables, such as model-based design, documentation, coordination, and clash detection, are now widely used, while more advanced uses linked to operations, sustainability, interoperability, and AI remain less mature. The report also highlights ongoing challenges, including skills gaps, uneven training provision, limited OpenBIM adoption, and inconsistent implementation of ISO 19650 and …

Characteristics And Implementation Paths Of Goal For Building Energy Powerhouse, Liye Xiao, Jiaofeng Pan, Xiaojiong Wang, Deqiang Sun, Mingliang Qi, Jianlei Mo, Huimin Li, Ting Wang, Jie Yang, Jie Lin, Yuchao Wang, Haiting Chen

Bulletin of Chinese Academy of Sciences (Chinese Version)

China is a major energy-consuming country, and ensuring effective energy supply is one of the core tasks for promoting national development and national rejuvenation. To guarantee national energy supply and energy security, and to vigorously develop and utilize clean and low-carbon energy, the Outline of the 15th Five-Year Plan for National Economic and Social Development of the People’s Republic of China clearly states: “We will thoroughly implement the new energy security strategy, accelerate the construction of a clean, low-carbon, safe, and efficient new energy system, and build a strong energy nation. We will promote the safe, reliable, and orderly replacement …

Seizing Strategic High Ground Of Space Computing Power: Global Competition Landscape And China’S Path, Yan Chen, Wenbin Song, Ping Zhang

Bulletin of Chinese Academy of Sciences (Chinese Version)

The deep integration of artificial intelligence and commercial aerospace is accelerating the transformation of space computing power from conceptual exploration to engineering verification, becoming a key direction for building an integrated space-air-ground information infrastructure. This study delves into its strategic value, global landscape, industrial chain bottlenecks, and advancement paths. The research reveals that the core value of space computing power does not lie in replacing ground data centers, but rather in focusing on network coverage blind spots, data transmission limitations, and high-timeliness scenarios, providing a new supply model of “in-orbit computing + space-ground collaboration”. Currently, the world has entered a …

Visualization And Marker-Less Tracking Of User-Defined Pre-Processed Mri Articulator Data Using Deep Learning, Michael De George

Student Theses

This thesis presents a comprehensive framework for the automated tracking and visualization of articulatory movements based on magnetic resonance imaging (MRI) data. A well-known data analysis tool for markerless pose estimation, known as DeepLabCut, is investigated for this purpose. The performance of this tool is enhanced through the design and implementation of a pre-processor. DeepLabCut is a markerless pose estimation toolbox based on deep learning, which overcomes the issue of making manual annotations frame-by-frame. Limitations from manually marking the MRI images are addressed by implementing transfer learning with convolutional neural networks to achieve accurate, user-defined articulator tracking without markers. Current …

The Impact Of Optimization Approximation Algorithms On The Performance Of The Bht-Qaoa, Ali Al-Bayaty, Marek Perkowski

Electrical and Computer Engineering Faculty Publications and Presentations

This article investigates the performance impact of five classical optimization approximation algorithms on our previously introduced quantum search algorithm, termed the Boolean–Hamiltonians Transform for Quantum Approximate Optimization Algorithm (BHT-QAOA), to effectively search for all best-approximated solutions for Boolean-based problems. These optimization approximation algorithms are BFGS, L-BFGS-B, SLSQP, COBYLA, and COBYQA. Their performance impact is evaluated and compared using two proposed performance metrics—(i) the final number of function evaluations (the lower numbers denote the best optimization approximation algorithms) and (ii) the final quality of qubit measurements (the higher values indicate all best-approximated solutions were found for a problem). Arbitrary classical Boolean …

Optimization And Energy Efficiency Analysis Of An Automatic Feed Mixer With A Rotating Drum Mechanism, Kris Witono, Talifatim Machfuroh, Nurlia Pramita Sari, Lisa Agustriyana, Aini Lostari

Journal of Mechanical Engineering Science and Technology (JMEST)

Energy-efficient feed mixer machines are important for improving the productivity and sustainability of small and medium-scale livestock farms. Previous studies primarily focused on either structural performance or mixing efficiency, with limited studies integrating both aspects. Therefore, this study evaluated an automatic rotating-drum feed mixer by combining Finite Element Method (FEM) analysis and energy modeling. The study used FEM simulations for different materials, namely A36 steel alloy, stainless steel 304, aluminium 6061, and galvanized steel, with thicknesses of 3 mm and 4 mm, as well as different drum systems. The FEM results showed that all evaluated materials met the minimum safety …

The Heterogeneous Response Of Astrocytes In Different Brain Regions Under Metabolic Stress, Pawan Kumar Thapaliya

USF Tampa Graduate Theses and Dissertations

Astrocytes play a central role in maintaining ionic homeostasis, neurotransmitter clearance, and brain en-ergy metabolism. Their excitability, reflected in activity-evoked intracellular sodium (Na+, [Na+]i) and calcium (Ca2+,[Ca+]i) dynamics, is increasingly recognized as a critical determinant of neuronal function and vulnerability, particularly under metabolic stress. Despite their importance, the mechanisms underlying astrocytic ionic signaling and its regional heterogeneity remain incompletely understood. In this thesis, I developed detailed computational and biophysical models to investigate astrocytic Na+ and Ca2+ dynamics, their bioenergetic consequences, and the roles of key ion transporters under physiological and metabolically stressed conditions, validating the models against experi- mental data. …

Millimeter-Wave Antenna Gain Enhancement Through Stacked Planar Substrates, Zachary Bergstedt

Electrical and Computer Engineering ETDs

This work presents a new wideband millimeter-wave (mmWave) and sub-terahertz antenna design with flexible directivity through the integration of stepped horn antennas and transverse substrate integrated waveguide (SIW) slots for radar and communication applications. The work gives a theoretical and analytical basis for this filter-inspired approach to improving bandwidth and directivity, and presents design and results for a standalone stepped horn, a Ka-band antenna with a solid stepped horn and SIW feed, and W-band antennas with empty SIW feeds and stepped horns manufactured out of multiple planar layers. The realized antennas show bandwidth up to 40% and gain up to …

Hierarchical Safe Reinforcement-Learning Framework For Mission-Aware, Edge-Enabled Multi-Uav Iot Networks, Abee F. Alazzwi

Electrical and Computer Engineering ETDs

This Ph.D. dissertation presents a unified Hierarchical Safe Reinforcement Learning (HSRL) framework for mission-aware, edge-enabled multi-UAV Internet of Things (IoT) networks. The work addresses the need for autonomous aerial infrastructures capable of delivering low-latency communication, scalable edge computation, and provably safe operation in dynamic environments. The dissertation develops three primary contributions. First, it formulates longhorizon drone base station placement and load balancing as a strategic actor–critic learning problem, enabling proactive adaptation to spatiotemporal demand variations. Second, it introduces a mission-aware multi-agent reinforcement learning controller for coordinated mobility, sensing, and computation offloading under latency and energy constraints. Third, it integrates a …

Predictive Natural Language Metrics Of Alzheimer's Disease And Cognitive Decline Trend Analysis, Zerui Ma

Computer Science and Engineering Theses and Dissertations

Inspired by Dr. David Snowden's Nun Study, which linked early-life Propositional Idea Density (PID) to later-life Alzheimer's disease, this thesis investigates two questions: whether fine-tuned Transformer-based large language models (LLM) can detect cognitive decline from patient speech transcripts with meaningful feature attribution, and whether longitudinal PID trends are observable across large-scale internet and academic text corpora. We evaluate dementia prediction on the DementiaBank Pitt Corpus and conduct an exploratory longitudinal PID analysis across seven diverse datasets spanning up to 29 years and over 12.6 million documents. This work suggests that linguistic ability metrics, traditional PID metrics and novel LLM-based analysis, …

Technical Advancements In Single-Molecule Spectroscopy For High-Throughput Measurement Of Long-Time Dynamics, Quyen B. Le

Chemical and Biological Engineering ETDs

Single-molecule fluorescence spectroscopy is a powerful technique for resolving transient biomolecular dynamics and quantifying free energy landscapes, kinetics, and binding interactions. However, two fundamental limitations restrict its broader application: slow, labor-intensive data acquisition and the limited observation time imposed by fluorophore photobleaching. These limitations hinder its use in drug discovery and biomedical engineering applications that require both high-throughput and access to long-time dynamics. In this work, both challenges are addressed through technical advancements. First, a simple, generalizable approach is introduced to automate data acquisition, eliminating manual intervention during experiments. This increases the acquisition rate by more than an order of …

Nuclear Deterrence: Enhancing The Mission Through Smart Factory Predictive Solutions, Jarrod Matthew Ronquillo

Chemical and Biological Engineering ETDs

To meet the stewardship and modernization initiatives set by the Department of Energy new technologies must enter the manufacturing facilities within the nuclear deterrent complex. Predictive solutions begin with collecting data. An equipment health monitoring device was created to streamline data collection and organization for equipment and processes. A predictive and process performance dashboard was developed for a deionized water system. The dashboard used Western Electric statistical process rules and a machine learning regression algorithm to predict when the resistivity would fall out of specification. Lastly, a remaining useful life calculation was developed for all equipment related to nuclear deterrent …

Novel Algorithmic Methods For Random Telegraph Noise Detection And Characterization In Electronic Devices, Victor Darie Pepel

Electrical and Computer Engineering ETDs

Random telegraph noise (RTN) produces discrete stochastic fluctuations in nanoscale semiconductor devices and increasingly limits performance and reliability as dimensions scale. This dissertation introduces three algorithmic contributions enabling automated and accurate RTN characterization across diverse devices and operating conditions. First, a computationally efficient histogram-based detection algorithm enables rapid identification of RTN in large focal plane array datasets for statistically robust defect analysis. Second, a frequency decomposition framework separates slow and fast RTN components, extending the range of extractable time constants and reducing estimation error in multi-trap signals obscured by background noise. Third, to address the lack of standardized RTN metrics, …

Applications And Comparisons Of Machine Learning Methods In Ultra-Fast Laser Control, Aasma Aslam

Electrical and Computer Engineering ETDs

This dissertation demonstrates the applications and comparative analyses of machine learning methods in ultrafast laser control. By learning the relationship between the system’s input parameters and output pulse characteristics, the performance of a laser can be significantly improved. In this work, the results are presented in two stages by utilizing data from the femtosecond laser system. The first stage concerns two neural networks, named NN1 (fitrnet) and NN2 (feedforwardnet). The second stage, which extended with five different models, namely the linear regression (fitlm), the support vector machine (SVM), the Gaussian process regression (GPR), the boosted tree (fitrensemble), and LASSO (fitrlinear), …

Self-Supervised Spoofing Detection, David S. Choi

Electrical and Computer Engineering ETDs

Global Navigation Satellite Systems (GNSS) are vulnerable to spoofing attacks that can mislead receivers with counterfeit signals. Traditional detection techniques, such as antenna-based, encryption based, and signal processing approaches, often face limitations in adaptability, computational cost, or reliance on predefined thresholds. Supervised machine learning models, while powerful, require large labeled datasets and struggle to generalize to unseen spoofing scenarios. In this work, we propose a self-supervised spoofing detection framework based on Adaptive Sparse Gaussian Processes (ASGP). The method predicts incoming GNSS features using past observations and identifies spoofing as anomalous deviations in the prediction residuals. Unlike supervised approaches, ASGP adapts …

Quantum Control Protocols For Robust Quantum Computing, Leeseok Kim

Electrical and Computer Engineering ETDs

The fundamental goal of quantum computing is to precisely control quantum systems to perform meaningful tasks, including implementing high-fidelity quantum gates for reliable quantum computation and accurately simulating complex quantum many- body dynamics. In this dissertation, we develop improved quantum control protocols for three distinct objectives, quantum error suppression, quantum optimal control, and analog quantum algorithms, achieving performance beyond standard approaches. First, we introduce new dynamical decoupling protocols, including both determin- istic and randomized constructions, that can substantially outperform conventional deterministic sequences. We then extend the randomized approach to dynamically corrected gates. Second, we propose a randomized quantum optimal control …