Engineering Commons^™

Machine Learning Model For Detecting Masked Hypertension In Young Adults, Brendyn Miller, Samuel Coeyman, Annemarie Wentzel, Carina M.C. Mels, William J. Richardson

Chemical Engineering Faculty Publications and Presentations

Introduction Cardiovascular disease (CVD) remains the leading global cause of mortality, with hypertension (HT) being a significant contributor, responsible for 56% of CVD-related deaths. Masked hypertension (MHT), a condition where patients exhibit normotensive blood pressure (BP) in clinical settings but elevated BP in out-of-clinic measurements, poses an elevated risk for cardiovascular complications and often goes undiagnosed. Current diagnostic methods, such as ambulatory BP monitoring (ABPM) and home BP monitoring (HBPM), have limitations in feasibility and accessibility. Methods This study aimed to address these challenges by leveraging machine learning (ML) models to predict MHT based on clinical data from a single …

A Novel Hybrid Intrusion Detection Model: A New Metaheuristic Approach For Feature Selection Based On Ai Techniques For Cyber Threat Detection, Maryam Mahdi Alhusseini, Alireza Rouhi

Iraqi Journal for Computer Science and Mathematics

The rapid increase in internet usage, digital transformation, and the rise of interconnected devices have greatly expanded the attack surface, introducing new and evolving cybersecurity challenges. Conventional security solutions frequently have difficulty adjusting to complex threats and the vast dimensionality of network traffic data, particularly in the case of imbalanced datasets. To tackle these challenges, this research introduces a Hybrid Intrusion Detection System (HyIDS-EVO) that combines the Energy Valley Optimizer (EVO) for feature selection and dimensionality reduction with machine learning classifiers, which include Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbors (KNN). The system’s effectiveness …

Integrated Optimization And Data-Driven Modeling For Seawater Intrusion Mitigation And Prediction, Assaad Hassan Kassem

Thesis/ Dissertation Defenses

Seawater intrusion (SWI) threatens the reliability of coastal groundwater especially in hyper-arid settings, where climatic stress and pumping accelerate salinization. This dissertation advances two complementary approaches to managing SWI: Part A optimizes mitigation measures, hydraulic (pumping/injection) and physical barriers (cutoff walls, subsurface dams) on benchmark models; Part B predicts SWI in the hyper-arid Fujairah (UAE) coastal aquifer using total dissolved solids (TDS) as a proxy, through machine learning-based models, spatially and dynamically. A bibliometric synthesis first maps the evolution of SWI models and mitigation strategies, identifying gaps that motivate the subsequent methodological developments. Part A employs the classical Henry problem …

Osc-Net: A Multi-Fidelity Machine Learning Model For Organic Solar Cells, Haizhou Yang, Adam Wold, Junlin Ou, Jeromy J. Rech, Wei You, Yi Wang

Faculty Publications

Organic solar cells (OSCs) have emerged as a promising renewable energy technology, offering advantages such as lightweight design, semitransparency, flexibility, and cost-effectiveness. Power conversion efficiency (PCE) is a key device performance parameter for OSCs, defined as the ratio of the electrical power output generated by the device to the incident solar power input. Despite significant advances, the development of high-performance OSCs remains a labor-intensive process, heavily dependent on expert experience, involving extensive synthesis, characterization, and iterative optimization. Data-driven methods offer a promising alternative for accelerating material discovery, but their effectiveness is often limited by the scarcity of high-quality experimental data. …

Optimization Of Rheological Models Using Machine Learning-Based Surrogates, Sourav Saha

LSU Master's Theses

This study addresses the challenge of accurately modeling flows of complex, multiphase media through fluid machines, where highly viscous, compressible, and shear-thinning behavior of the medium complicates the construction of predictive models. Such media include slurries, biomass, foods, and other materials of high value and societal importance. A high-fidelity CFD model of a screw-type fluid machine with an outlet flow restriction is developed. The flowing medium is treated as a single-phase, compressible, non-Newtonian fluid. The model incorporates a non-linear pressure-dependent density model and a strain-rate-dependent viscosity law (Cross power law). A parametric study is conducted by varying key rheological parameters …

Towards Automated And Explainable Insider Threat Response In Electronic Health Records: A Role-Aware Machine Learning Framework, Luca Lippi Ornstil

Master's Theses

Healthcare remains a prime target for cyberattacks, with insider misuse and credential compromise posing major risks to Electronic Health Records (EHRs). This thesis introduces a role-aware, explainable anomaly detection and response framework integrated with OpenEMR to address post-authentication threats. Four models—Local Outlier Factor (LOF), Isolation Forest, Autoencoder, and Graph Neural Network (GNN)—detect behavioral deviations across temporal, device, and role-based features, with LOF serving as the primary runtime detector. A configurable policy engine maps anomaly severity to proportional actions, from email alerts to read-only restrictions or account suspension, all reversible and auditable. Evaluation on real EHR logs shows the system’s operational …

Possibilities Of Digitizing And Applying Artificial Intelligence To National Occupational Classification (Noc-2025) In Uzbekistan, Shohrux Nurali O‘G‘Li Narzullayev

Chemical Technology, Control and Management

This article examines the process of digitizing National Occupational Classification (NOC-2025) in Uzbekistan, developed on the basis of the International Standard Classification of Occupations (ISCO-08), and the possibilities of applying artificial intelligence technologies to it. Although this classification exists today in a national form, and its digitization and the introduction of artificial intelligence elements to it based on modern technologies remain a pressing issue. In order to digitize the classification, international systems such as the International Standard Classification of Occupations (ISCO-08, ILO), European Skills, Competences, Qualifications and Occupations (ESCO), Occupational Information Network (O*NET, USA) and National Occupational Classification (NOC, Canada) …

Visible Image-Based Machine Learning For Identifying Abiotic Stress In Sugar Beet Crops, Seyed Reza Haddadi, Masoumeh Hashemi, Richard C. Peralta, Masoud Soltani

Plants, Soils and Climate Student Research

Previous researches have proved that the synchronized use of inexpensive RGB images, image processing, and machine learning (ML) can accurately identify crop stress. Four Machine Learning Image Modules (MLIMs) were developed to enable the rapid and cost-effective identification of sugar beet stresses caused by water and/or nitrogen deficiencies. RGB images representing stressed and non-stressed crops were used in the analysis. To improve robustness, data augmentation was applied, generating six variations on each image and expanding the dataset from 150 to 900 images for training and testing. Each MLIM was trained and tested using 54 combinations derived from nine canopy and …

Improved Streamflow Forecasting Through Swe-Augmented Spatio-Temporal Graph Neural Networks, Akhila Akkala, Soukaina Filali Boubrahimi, Shah Muhammad Hamdi, Pouya Hosseinzadeh, Ayman Nassar

Computer Science Student Research

Streamflow forecasting in snowmelt-dominated basins is essential for water resource planning, flood mitigation, and ecological sustainability. This study presents a comparative evaluation of statistical, machine learning (Random Forest), and deep learning models (Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Spatio-Temporal Graph Neural Network (STGNN)) using 30 years of data from 20 monitoring stations across the Upper Colorado River Basin (UCRB). We assess the impact of integrating meteorological variables—particularly, the Snow Water Equivalent (SWE)—and spatial dependencies on predictive performance. Among all models, the Spatio-Temporal Graph Neural Network (STGNN) achieved the highest accuracy, with a Nash–Sutcliffe Efficiency (NSE) of 0.84 …

Foundations Of Artificial Intelligence In Healthcare Diagnostics: A Systematic Survey, Raghad Tariq Al-Hassani

Al-Esraa University College Journal for Engineering Sciences

Artificial Intelligence (AI) is becoming the cornerstone of the future of healthcare diagnostics, that has to ability to change the healthcare diagnostic landscape in terms of diagnostic accuracy, speed, and availability. This systematic review investigates the basic methods, tools, applications, and challenges involved in the integration of AI in diagnostic medicine. It emphasizes the using of machine learning models, deep learning networks (e.g., CNNs), NLP for clinical documentation, and smart computing infrastructures, such as edge device and IoMT. They are making possible real-time, data-driven decision making that is already at human-expert-level performance or, in some cases, even better (in the …

Towards Autonomous Energy Management: Machine Learning For Effective Auditing And Optimization, Sameh O. Abdellatif, Sherif Ashraf, Mira Mohsen

Electrical Engineering

This study presents a fully automated procedure for energy management and auditing, applicable to a diverse range of residential and commercial loads, leveraging machine learning techniques across three key phases: load classification, benchmarking, and smart monitoring. The model effectively categorizes energy loads based on consumption patterns, establishes performance benchmarks through historical data analysis, and employs real-time monitoring to identify inefficiencies and predict future energy usage. Evaluating the model through four distinct case studies demonstrates its capability to optimize energy consumption in a techno-economic manner, achieving significant energy savings of 34.73 MWh/year for essential loads in Egypt, 215.67 MWh/year for HVAC …

Nonlinear Design Scaling Of Electric Machines Based On Hybrid De And Meta-Modeling Application To Synchronous Motors With Combined Pm Stator And Reluctance Rotor Excitation, Oluwaseun A. Badewa, Dan M. Ionel

Electrical and Computer Engineering Graduate Research

This paper presents an innovative method for nonlinear scaling of electric machines by integrating machine learning (ML)-based meta-modeling with a differential evolution (DE) algorithm. The technique is applied to high-performance combined-excitation synchronous electric motors which exhibit highly nonlinear characteristics, making performance scaling challenging. The proposed approach employs an ML meta-model trained on data obtained from finite element analysis (FEA), utilizing an experimentally validated model for nonlinear scaling and performance prediction at different power ratings. The accuracy of the meta-model in capturing the nonlinear relationships between design parameters and motor performance is first assessed using metrics such as R-squared (R2) and …

Design Optimization And Scaling Of Coreless Afpm Machines Using Hybrid Fea-Based Differential Evolution And Machine Learning, Matin Vatani, David R. Stewart, Donovin D. Lewis, Dan M. Ionel

Electrical and Computer Engineering Graduate Research

This paper presents a machine learning (ML) based design framework for the fast and accurate optimization of coreless axial flux permanent magnet (AFPM) machines. Although the absence of magnetic cores eliminates material nonlinearity, the design process remains highly nonlinear due to the complex influence of geometric parameters. To overcome the computational challenges of finite element analysis (FEA)-based optimization, a series of multi-objective differential evolution (MODE) optimizations were conducted across various machine sizes at constant power output. The resulting design data was used to train an artificial neural network (ANN), enabling rapid prediction of machine performance without the need for repeated …

Fast, Slow, And Metacognitive Thinking In Ai, M. Bergamaschi Ganapini, M. Campbell, F. Fabiano, L. Horesh, J. Lenchner, A. Loreggia, N. Mattei, F. Rossi, Biplav Srivastava, K. B. Venable

Faculty Publications

Inspired by the ”thinking fast and slow” cognitive theory of human decision making, we propose a multi-agent cognitive architecture (SOFAI) that is based on ”fast”/”slow” solvers and a metacognitive module. We then present experimental results on the behavior of an instance of this architecture for AI systems that make decisions about navigating in a constrained environment. We show that combining the two decision modalities through a separate metacognitive function allows for higher decision quality with less resource consumption compared to employing only one of the two modalities. Analyzing how the system achieves this, we also provide evidence for the emergence …

Machine-Learning-Assisted Discovery Of Lattice Dynamics Signatures Of Sodium Superionic Conductors, Ogheneyoma Maxwell Aghoghovbia

Theses and Dissertations

Sodium superionic conductors are key to the development of all-solid-state sodium batteries. Discovery of new superionic conductors has traditionally relied on insights from material defect chemistry and the transition/hopping theory, while the role of lattice vibrations, i.e., phonons, remains underexplored. We identify key lattice dynamics signatures that govern ionic conductivity by analyzing the phonon mean squared displacement (MSD) of Na⁺ ions. By high-throughput screening of a dataset of 3903 Na-containing structures, we establish a strong positive correlation between phonon MSD and diffusion coefficients, providing a quantitative correlation between lattice dynamics and ion transport. To accelerate this discovery, we incorporate …

Soft Sensing Of Biological Oxygen Demand In Industrial Wastewater Using Machine Learning Models, Muhammad Hassnain, Sarada M.W. Lee, Muhammad Rizwan Azhar

Research outputs 2022 to 2026

Traditional methods for determining biological oxygen demand (BOD) from industrial water resource recovery facilities (WRRFs) are time-consuming and often impractical for real-time process control. This study explores the application of machine learning (ML) and artificial intelligence (AI) models for the prediction of final effluent BOD (F-BOD) based on physicochemical and operational parameters by leveraging nineteen years of historical laboratory and instrumentation data from the WRRF of an essential oil manufacturing plant. The predictions from these models are then used to simulate the process dynamics, assessing the optimal operational boundary conditions for all input parameters at which the target (F-BOD) falls …

Machine Learning Classification Of Eeg Responses To Pain-Related Vs Non-Pain-Related Stimulus In Preterm Infants, Lojain Hamwi, Hang Du, Sara Jasim, Xiaogang Wang, Vibhuti Shah, Carol Cheng, Lorenzo Fabrizi, Maria Fitzgerald, Judith Meek, Nicole Racine, Ian Stedman, Rebecca Pillai Riddell

Michigan Tech Publications

INTRODUCTION: Unmanaged pain in preterm infants can lead to long-term developmental consequences. Current pain assessment methods lack specificity, resulting in possible pain mismanagement in Neonatal Intensive Care Units (NICUs). This study explores the application of machine learning (ML) to differentiate between pain-related and non-pain-related cortical activity in preterm infants. OBJECTIVE: To evaluate the performance of ML models in distinguishing cortical EEG activity during a painful procedure in preterm infants across different postmenstrual ages (PMAs). METHODS: This observational study was conducted from June 2015 to May 2024 at Mount Sinai Hospital in Toronto, Canada, and University College London Hospital, United Kingdom. …

Iot-Enabled Machine Learning Framework For Prediction Of Eutrophication, Hocine Dai, Akli Abbas, Houssam Eddine-Othman Lachemat, Aicha Aid

Iraqi Journal for Computer Science and Mathematics

This study presents an innovative predictive monitoring framework that integrates the Internet of Things (IoT) with advanced machine learning (ML) techniques to model the relationship between oxidized nitrate (NOX)—employed as the sole predictor—and chlorophyll a (CHLA), a key proxy for algal biomass. By utilising a single optimally selected parameter, the approach significantly reduces sensor deployment complexity and instrumentation costs, while minimising data acquisition and computational requirements. Logarithmic and Yeo-Johnson transformations were applied to the predictor and target variables, respectively, to address distributional skewness and enhance variance homogeneity. An optimised Random Forest model demonstrated strong predictive performance, achieving a coefficient of …

Machine-Learning-Assisted Discovery Of Lattice Dynamics Signatures Of Sodium Superionic Conductor, Ogheneyoma Aghoghovbia, Riccardo Rurali, Mohammed Al-Fahdi, Joshua Ojih, De-En Jiang, Ming Hu

Faculty Publications

Sodium superionic conductors are key to the development of all-solid-state sodium batteries. Discovery of new superionic conductors has traditionally relied on insights from material defect chemistry and the transition/hopping theory, while the role of lattice vibrations, i.e., phonons, remains underexplored. We identify key lattice dynamics signatures that govern ionic conductivity by analyzing the phonon mean squared displacement (MSD) of Na⁺ ions. By high-throughput screening of a dataset of 3903 Na-containing structures, we establish a strong positive correlation between phonon MSD and diffusion coefficients, providing a quantitative correlation between lattice dynamics and ion transport. To accelerate this discovery, we …

Transparent Eeg Analysis: Leveraging Autoencoders, Bi-Lstms, And Shap For Improved Neurodegenerative Diseases Detection, Badr Mouazen, Ahmed Bendaouia, Omaima Bellakhdar, Khaoula Laghdaf, Aya Ennair, El Hassan Abdelwahed, Giovanni De Marco

Manufacturing & Industrial Engineering Faculty Publications

Highlights

• Novel hybrid architecture: Combined autoencoders with bidirectional LSTM networks for enhanced EEG signal classification, achieving 98% accuracy in distinguishing AD, FTD, and healthy controls.

• Explainable AI integration: Implemented SHAP (SHapley Additive exPlanations) framework to enhance model transparency and identify entropy as the most influential feature for neurodegenerative disease detection.

• Optimal temporal segmentation: Demonstrated that 5-s EEG windows with 50% overlap provide the best balance between classification accuracy and computational efficiency.

• Comprehensive feature extraction: Utilized Power Spectral Density (PSD) analysis across standard frequency bands (Delta, Theta, Alpha, Beta, Gamma) following autoencoder-based dimensionality reduction.

• Superior performance validation: Outperformed traditional machine learning …

  ---

  Go to article

Developing Workflows For Passive Acoustic Detection Of Bedload Transport, Quinn Morgan

Dissertations and Theses

Bedload transport is defined as the amount of sediment, including gravel and rocks, traveling down stream. Monitoring bedload transport is important for river safety, hydrological studies and conservation efforts. Existing methods of directly measuring bedload transport (or bedload flux) involve lowering a collection device into a river and measuring the sediment collected; which can be expensive and time consuming. Hydroacoustic sensors, such as hydrophones, have had success tracking bedload flux remotely. This works by measuring the relatively high frequency of sediment impacts to map onto total bedload transported. No perfected method for detection of sediment generated noise (SGN) currently exists. …

A Polar Turbulence Invariant Map With Applicability To Realisable Machine Learning Turbulence Models, James G. Wnek, Christopher Schrock, Eric M. Wolf, Mitch Wolff

Mechanical and Materials Engineering Faculty Publications

Invariant maps are a useful tool for turbulence modelling, and the rapid growth of machine learning-based turbulence modelling research has led to renewed interest in them. They allow different turbulent states to be visualised in an interpretable manner and provide a mathematical framework to analyse or enforce realisability. Current invariant maps, however, are limited in machine learning models by the need for costly coordinate transformations and eigendecomposition at each point in the flow field. This paper introduces a new polar invariant map based on an angle that parametrises the relationship of the principal anisotropic stresses, and a scalar that describes …

Efficient Inference Of Performance Models In Openmp Applications, Gaurav Punjabi

Computer Science and Engineering Master's Theses

Choosing the best OpenMP parameters such as thread count, scheduling type, and chunk size is essential for optimizing parallel program performance. One of the promising approaches is to infer a (pre-trained) performance model at runtime to determine the parameters to run OpenMP parallel regions. Such a performance prediction model can require programs’ code information such as intermediate representation (IR) and other at-runtime information (e.g., input sizes) to make a performance prediction. In such a scenario, extracting or querying the IR information at runtime can create a significant runtime overhead. This thesis proposes a compiler-asssited tuning framework that shifts IR extraction …

Physics-Based Machine Learning Framework For Predicting Structure-Property Relationships In Ded-Fabricated Low-Alloy Steels †, Atiqur Rahman, Md Hazrat Ali, Asad Waqar Malik, Muhammad Arif Mahmood, Frank Liou

Mechanical and Aerospace Engineering Faculty Research & Creative Works

The Directed Energy Deposition (DED) process has demonstrated high efficiency in manufacturing steel parts with complex geometries and superior capabilities. Understanding the complex interplays of alloy compositions, cooling rates, grain sizes, thermal histories, and mechanical properties remains a significant challenge during DED processing. Interpretable and data-driven modeling has proven effective in tackling this challenge, as machine learning (ML) algorithms continue to advance in capturing complex property structural relationships. However, accurately predicting the prime mechanical properties, including ultimate tensile strength (UTS), yield strength (YS), and hardness value (HV), remains a challenging task due to the complex and non-linear relationships among process …

Predicting Groundwater Withdrawals Using Machine Learning With Limited Metering Data: Assessment Of Training Data Requirements, Dawit Asfaw, Ryan G. Smith, Sayantan Majumdar, Katherine Grote, Bin Fang, B. B. Wilson, V. Lakshmi, J. J. Butler

Geosciences and Geological and Petroleum Engineering Faculty Research & Creative Works

The future of major aquifer systems supporting irrigated agriculture is threatened due to unsustainable groundwater pumping. Metering of pumping is key for implementing robust groundwater management, but metering is limited in most aquifers. Although machine learning methods have been used to estimate pumping over certain regions, these studies have not fully demonstrated the data quantity and input parameter requirements to accurately estimate regional groundwater pumping. This study determined the data quantity required and identified relevant features to develop Random Forests-based annual groundwater pumping estimates (2008–2020) over the Kansas High Plains aquifer. We predicted pumping at two spatial scales, i.e., point …

Low-Resource Ecoacoustic Audio Classification, Enis Berk Coban

Dissertations, Theses, and Capstone Projects

Ecoacoustic monitoring via machine learning enables scalable analysis but is often constrained by labeled data scarcity, particularly in remote regions like the Arctic. This thesis confronts low-resource ecoacoustic audio classification by developing and evaluating complementary machine learning methodologies. We introduce EDANSA, the first publicly available, expert- labeled Arctic dataset of its kind, curated via novel active learning, alongside a baseline CNN. We systematically evaluate transfer learning, showing general audio embeddings effectively bootstrap classifiers for challenging Arctic sounds, significantly outperforming direct label mapping. Optimizing label utility, we investigate standard data augmentation and introduce novel audio data valuation via Shapley values, revealing …

Multi-Fidelity Machine Learning Modeling For Aerodynamic Response Prediction Of Aerospace Vehicles, Ethan S. Jackman

Theses and Dissertations

Hypersonic vehicle design requires understanding complex aerodynamic phenomena across the full flight regime. This study presents a novel MF surrogate modeling methodology that enables the prediction the full field response across a vehicle’s surface. A Space-Filling Curve (SFC) is used to convert unstructured data into 1D vectors. The a Convolutional Autoencoder is used with transfer learning to reduce the dimensionality of the data. An Emulator-Embedded Neural Network (E2NN) combines multi-fidelity data for fast, accurate predictions. A benchmark analytical example and hypersonic application are used to evaluate the methodology. Using various numbers of samples and sampling strategies it is found that …

Retracted: Iot Flow Parameters Classification Based On Machine Learning Techniques, El-Sayed M. El-Kenawy, Marwa M. Eid, Ban Salman Shukur, Amel Ali Alhussan, Doaa Sami Khafaga

Iraqi Journal for Computer Science and Mathematics

In recent years, there has been a highly remarkable convergence of artificial intelligence (AI) and the Internet of Things (IoT), which has made rapid progress in smart city initiatives by developing smart devices for such cities. Since these devices are increasingly diversified, they require a resilient communication network to demonstrate high performance in managing consistent traffic flows. A machine learning model intended for identifying network parameters from diverse devices, in addition to proposing modifications meant for network performance enhancement, is developed in this study. In relation to packet data as a network traffic parameter, employing gateway devices can facilitate its …

Research On Autonomous Deviation Correction Of Tunnel Boring Machines And Parameters Based On Machine Learning, Zhang Jun, Li Maopeng

Journal of China & Foreign Highway

In order to solve the problem of realizing the autonomous deviation correction of tunnel boring machines (TBMs ), a TBM deviation correction control method that integrated the random forest (RF) algorithm with the genetic algorithm (GA) was proposed based on actual engineering data.The method combined a prediction model with an optimization model,using target deviation values as input to invert and output the required TBM deviation correction parameter values,thereby further improving the automation level of TBM deviation correction.By comparing it with the actual data,the feasibility of the model was verified.The results show that the RF algorithm-based prediction model achieves an R2 …

Ai-Driven Solutions For Electricity Fraud Detection: A Data-Centric Approach, Fatemeh Alimoradi, Zahra Alimoradi, S. Mohammadali Zanjani, Ghazanfar Shahgholian

NJF Intelligent Engineering Journal

Electricity fraud detection presents a significant challenge for power distribution companies, as non-technical losses resulting from fraudulent activities adversely affect revenue and operational efficiency. This study presents a machine learning-driven approach for accurately identifying fraudulent electricity consumption patterns. A comprehensive analysis of electricity usage data is conducted using multiple classification models, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), a Stacking Model integrating SVM, KNN, Random Forest, and Gradient Boosting, as well as a Weighted Model leveraging confidence-based prediction adjustments. Model performance is evaluated using key metrics, including accuracy, precision, recall, F1-score, AUC-ROC, and confusion matrices. The results indicate that …