Engineering Commons^™

Application Of Multiple Data Augmentation Techniques To Improve Training With Synthetic Sar Data In Common Cnn, Stephanie M.V. Saich

Browse all Theses and Dissertations

To address the issues of limited target data in the Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) problem set, synthetic data is often used to aid in filling the gap. This paper covers an in depth look at the use of colorization, dynamic range adjustment, and target extraction as data augmentation techniques to improve the accuracy of deep learning networks trained on synthetic SAR data. The use of multiple different data augmentations combine to dramatically improve the accuracy of a common Convolutional Neural Network (CNN) over the use of standard synthetic data. A comparison of increasing fraction of measured …

Pneumonia Detection With Limited And Imbalanced Data Using Energy-Based Out-Of-Distribution Technique, Jasbin Karki

Browse all Theses and Dissertations

The automated detection of pneumonia through chest X-ray presents a critical challenge in medical diagnostics, particularly due to the restrictions of limited and imbalanced chest X-ray data for training AI models. Traditional methods that depend on softmax confidence scores can be overconfident even when generating erroneous outputs especially when they are processing completely new inputs, leading to unreliable diagnostic results. This research addresses challenges in AI models which aim to develop a robust pneumonia detection system using an Energy-Based Out-of-Distribution (OOD) technique that can work effectively even with limited and imbalanced data. The study focused on creating a more reliable …

Machine-Learning-Assisted Design Of Deep Eutectic Solvents Based On Uncovered Hydrogen Bond Patterns, Usman Lame Abbas, Yuxuan Zhang, Joseph Tapia, Md Selim, Jin Chen, Jian Shi, Qing Shao

Markey Cancer Center Faculty Publications

Non-ionic deep eutectic solvents (DESs) are non-ionic designer solvents with various applications in catalysis, extraction, carbon capture, and pharmaceuticals. However, discovering new DES candidates is challenging due to a lack of efficient tools that accurately predict DES formation. The search for DES relies heavily on intuition or trial-and-error processes, leading to low success rates or missed opportuni- ties. Recognizing that hydrogen bonds (HBs) play a central role in DES formation, we aim to identify HB features that distinguish DES from non-DES systems and use them to develop machine learning (ML) models to discover new DES systems. We first analyze the …

Machine Learning-Based Disease Classification In Tomato (Solanum Lycopersicum) Plants = Klasifikasi Penyakit Berbasis Pembelajaran Mesin Pada Tanaman Tomat (Solanum Lycopersicum), Md Towfiqur Rahman, Sudipto Dhar Dipto, Israt Jahan June, Abdul Momin, Muhammad Rashed Al Mamun

Department of Agricultural and Biological Systems Engineering: Faculty Publications

English abstract

In Bangladesh, tomato cultivation faces significant challenges due to its susceptibility to various microorganisms, parasites, and bacterial infections. Typically, the early symptoms of these diseases first appear in roots and leaves, complicating timely detection. This study addresses the challenge of timely and accurate detection of diseases in tomato plants, crucial for effective plant protection management. Conventional manual inspection methods are time-consuming and subjective, resulting in delays in implementing necessary protection measures. Therefore, an image processing technique and machine learning algorithms were used for rapid and robust detection of diseases in tomato plant leaves, aiming to streamline the detection …

Data-Driven Approaches For Achieving Carbon Neutrality: Predictive Models For Reducing Co2 Emissions And Enhancing Industrial Sustainability, Farzana Islam

Graduate Theses, Dissertations, and Problem Reports (ETD)

In response to the escalating challenges posed by climate change and industrial inefficiency, this thesis presents a comprehensive investigation aimed at advancing the predictive modeling of global CO2 emissions and enhancing operational efficiency in steel manufacturing through Electric Arc Furnace (EAF) temperature optimization. Leveraging a rich dataset sourced from the World Development Indicators database alongside a meticulously curated dataset specific to EAF operations, our study applies an innovative blend of econometric and machine learning techniques, including Pooled Ordinary Least Squares (Pooled OLS), Random Effects (RE), Fixed Effects (FE), and Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) models. The …

On Confidence And Sense Of Belonging In Cybersecurity Students: Analysis & Prediction, Sadaf Amna Sarwari

Graduate Theses, Dissertations, and Problem Reports (ETD)

In recent years, there has been a rapid expansion of cybersecurity programs across higher education institutions in response to the widening skills gap in the cybersecurity job market. This study adopts quantitative and qualitative approaches to identify factors influencing West Virginia University (WVU)’s LANE Department of Computer Science and Electrical Engineering (LCSEE) students’ confidence and sense of belonging in the cybersecurity field. The results are based on data collected from surveys administered to LCSEE students in April 2022 and April 2023. The responses were analyzed using descriptive & inferential statistics and logistic regression techniques. Additionally, the 2023 data was utilized …

Molecular Understanding And Design Of Deep Eutectic Solvents And Proteins Using Computer Simulations And Machine Learning, Usman Lame Abbas

Theses and Dissertations--Chemical and Materials Engineering

Hydrophobic deep eutectic solvents (DESs) have emerged as excellent extractants. A major challenge is the lack of an efficient tool to discover DES candidates. Currently, the search relies heavily on the researchers’ intuition or a trial-and-error process, which leads to a low success rate or bypassing of promising candidates. DES performance depends on the heterogeneous hydrogen bond environment formed by multiple hydrogen bond donors and acceptors. Understanding this heterogeneous hydrogen bond environment can help develop principles for designing high performance DESs for extraction and other separation applications. This work investigates the structure and dynamics of hydrogen bonds in hydrophobic DESs …

Predictive Modeling Of Healthcare Traffic Using Machine Learning: A Comparative Study, Shadman Mahmood Khan Pathan, Sakan Binte Imran, M. M. Shabab Iqbal, Muhammad Enayetur Rahman, Md Nurul Absar Siddiky, Muhammad Rezaur Rahman, Md. Rafid Hassan, Nondon Lal Dey, Md. Sobuj Hossain

Electrical & Computer Engineering Faculty Publications

Effective healthcare traffic management is critical for ensuring prompt medical services, particularly in emergencies where delays can have life-threatening consequences. This study conducts a comparative analysis of three popular machine learning models—Linear Regression, Decision Trees, and Random Forests—for predicting healthcare-related traffic volumes. Utilizing a comprehensive dataset from a metropolitan interstate traffic system, the models were evaluated based on key performance metrics, including Mean Squared Error (MSE), R² Score, and execution time. The findings demonstrate that the Random Forest model outperforms the others, offering superior predictive accuracy and efficiency. These insights are valuable for optimizing traffic management in healthcare, ultimately contributing …

Explore Security And Machine Learning Applications In Next Generation Wireless Networks, Haolin Tang

Theses and Dissertations

Next-generation (NextG) or Beyond-Fifth-Generation (B5G) wireless networks have become a prominent focus in academic and industry circles. This is driven by the increasing demand for cutting-edge applications such as mobile health, self-driving cars, the metaverse, digital twins, virtual reality, and more. These diverse applications typically require high communication network performance, including spectrum utilization, data speed, and latency. New technologies are emerging to meet the communication requirements of various applications. Intelligent Reflecting Surface (IRS) and Artificial Intelligence (AI) are two representatives that have been demonstrated as promising and powerful technologies in NextG communications. While new technologies significantly enhance communication performance, they …

Machine Learning Assisted Optimization For Calculation And Automated Tuning Of Antennas, Lauren Linkous

Theses and Dissertations

The Antenna Calculation and Autotuning Tool (AntennaCAT) software suite represents a significant advancement in the field of antenna design by automating the entire design, CAD, simulation, and optimization process compatible with several EM simulation software suites. It is the first comprehensive implementation of machine learning in this context. In particular, this work includes the capability to create and export structured datasets from the aforementioned EM software for iterative improvement and includes an expandable selection of optimizers.

Machine Learning-Assisted Multiscale Simulation And Design Optimization Of Composite Jackets Under Low-Velocity Impacts, Masoud Mohammadi

Graduate Theses, Dissertations, and Problem Reports (ETD)

This research presents the development of a data-driven machine learning-assisted approach to simulate and optimize the design of composite materials subjected to low-velocity impacts. It focuses on a specific type of hybrid composite comprised of fiberglass and Kevlar fabrics stitched with Kevlar threads. The study begins by creating a multiscale finite element simulation for the composite under a low-velocity impact, which operates across three scales: microscale, mesoscale, and macroscale. This simulation accepts various inputs, such as different layer configurations and orientations, and provides impact outputs including maximum load and energy absorption capacity and displacement at failure.

Python and MATLAB scripts …

Improvements In Biomedical Image Analysis With Computational Intelligence And Data Fusion Techniques, Akanksha Maurya

Doctoral Dissertations

"An estimated 2 million new cases of basal cell carcinoma (BCC) are diagnosed each year in the United States, making it one of the most common skin cancers. Earlier detection of these cancers enables less invasive biopsies. Clinical detection consists of a preliminary visual observation of these skin lesions by an experienced dermatologist making it a specialized task highly dependent on their time, availability, and resources. Hence, there is a need for automating this process that can assist healthcare staff. In recent years, deep learning (DL) has been used extensively and successfully to diagnose different cancers in dermoscopic images. Telangiectasia …

Mission-Driven Inverse-Design Of Conventional And Unconventional Airplane Configurations Using Machine Learning, Rohan Srinivas Sharma

Doctoral Dissertations

"A key challenge in airplane design has been the long-desired ability to quickly create an airplane design with specific performance properties. To tackle this problem, researchers have traditionally leveraged design space exploration methodologies to iteratively explore a design space via surrogate models. On the other hand, inverse design is a process wherein performance objectives are fixed while a suitable design for satisfying these requirements is calculated. In recent years, artificial neural networks (ANNs) have been explored as viable surrogate models because traditional surrogate models struggle with the large number of output to input mapping needed for inverse design problems. In …

Accelerating Cavity Fault Prediction Using Deep Learning At Jefferson Laboratory, Md M. Rahman, A. Carpenter, K. Iftekharuddin, C. Tennant

Electrical & Computer Engineering Faculty Publications

Accelerating cavities are an integral part of the continuous electron beam accelerator facility (CEBAF) at Jefferson Laboratory. When any of the over 400 cavities in CEBAF experiences a fault, it disrupts beam delivery to experimental user halls. In this study, we propose the use of a deep learning model to predict slowly developing cavity faults. By utilizing pre-fault signals, we train a long short-term memory-convolutional neural network binary classifier to distinguish between radio-frequency (RF) signals during normal operation and RF signals indicative of impending faults. We optimize the model by adjusting the fault confidence threshold and implementing a multiple consecutive …

An Enhanced Deep Autoencoder For Flight Delay Prediction, Desmond B. Bisandu, Dan Andrei Soviani-Sitoiu, Irene Moulitsas

Journal of Aviation/Aerospace Education & Research

Accurate and timely flight delay prediction cannot be overemphasized because of the ever-increasing demand for air travel and its importance in deploying intelligent transportation systems. Nonetheless, there has not been a universal solution to the problem, as more intelligent flight decision systems are required for the aviation industry's future growth. Existing flight delay classification and prediction approaches are mainly shallow traffic models and do not satisfy many applications in the real world. Our motivation to rethink the deep architecture model for predicting flight delays emanates from the problem. In this research, we proposed a technique that modified stacked autoencoder architecture …

Leveraging Artificial Intelligence To Improve Data Configuration & Accuracy In Modern Flight Management Systems, Sreeram Chittayil

International Journal of Aviation, Aeronautics, and Aerospace

AI (Artificial intelligence) can automate the process of generating optimized flight routes using real-time data, such as AIRAC (Aeronautical Information Regulation and Control), significantly reducing the time needed for flight management tasks. While AIRAC data typically takes up to 28 days to refresh, AI could condense this process to just minutes, enhancing operational efficiency and ensuring pilots have timely and accurate flight information. The research includes practical experiments, prototype code, and visual case studies to demonstrate AI's role in optimizing FMS functions while addressing issues related to data input errors and human intervention. Key findings from trials show the algorithm's …

Using Feature Selection Enhancement To Evaluate Attack Detection In The Internet Of Things Environment, Khawlah Harahsheh, Rami Al-Naimat, Chung-Hao Chen

Electrical & Computer Engineering Faculty Publications

The rapid evolution of technology has given rise to a connected world where billions of devices interact seamlessly, forming what is known as the Internet of Things (IoT). While the IoT offers incredible convenience and efficiency, it presents a significant challenge to cybersecurity and is characterized by various power, capacity, and computational process limitations. Machine learning techniques, particularly those encompassing supervised classification techniques, offer a systematic approach to training models using labeled datasets. These techniques enable intrusion detection systems (IDSs) to discern patterns indicative of potential attacks amidst the vast amounts of IoT data. Our investigation delves into various aspects …

Ensemble Learning With Sleep Mode Management To Enhance Anomaly Detection In Iot Environment, Khawlah Harahsheh, Rami Al-Naimat, Malek Alzaqebah, Salam Shreem, Esraa Aldreabi, Chung-Hao Chen

Electrical & Computer Engineering Faculty Publications

The rapid proliferation of Internet of Things (IoT) devices has underscored the critical need for energy-efficient cybersecurity measures. This presents the dual challenge of maintaining robust security while minimizing power consumption. Thus, this paper proposes enhancing the machine learning performance through Ensemble Techniques with Sleep Mode Management (ELSM) approach for IoT Intrusion Detection Systems (IDS). The main challenge lies in the high-power consumption attributed to continuous monitoring in traditional IDS setups. ELSM addresses this challenge by introducing a sophisticated sleep-awake mechanism, activating the IDS system only during anomaly detection events, effectively minimizing energy expenditure during periods of normal network operation. …

Predictions Of Lattice Parameters In Niti High-Entropy Shape-Memory Alloys Using Different Machine Learning Models, Tu-Ngoc Lam, Jiajun Jiang, Min-Cheng Hsu, Shr-Ruei Tsai, Mao-Yuan Luo, Shuo-Ting Hsu, Wen-Jay Lee, Chung-Hao Chen, E-Wen Huang

Electrical & Computer Engineering Faculty Publications

This work applied three machine learning (ML) models—linear regression (LR), random forest (RF), and support vector regression (SVR)—to predict the lattice parameters of the monoclinic B19′ phase in two distinct training datasets: previously published ZrO₂-based shape-memory ceramics (SMCs) and NiTi-based high-entropy shape-memory alloys (HESMAs). Our findings showed that LR provided the most accurate predictions for a_c, a_m, b_m, and c_m in NiTi-based HESMAs, while RF excelled in computing βm for both datasets. SVR disclosed the largest deviation between the predicted and actual values of lattice parameters for both training datasets. A combination approach …

An Enhanced Real-Time Intrusion Detection Framework Using Federated Transfer Learning In Large-Scale Iot Networks, Khawlah Harahsheh, Malek Alzaqebah, Chung-Hao Chen

Electrical & Computer Engineering Faculty Publications

The exponential growth of Internet of Things (IoT) devices has introduced critical security challenges, particularly in scalability, privacy, and resource constraints. Traditional centralized intrusion detection systems (IDS) struggle to address these issues effectively. To overcome these limitations, this study proposes a novel Federated Transfer Learning (FTL)-based intrusion detection framework tailored for large-scale IoT networks. By integrating Federated Learning (FL) with Transfer Learning (TL), the framework enhances detection capabilities while ensuring data privacy and reducing communication overhead. The hybrid model incorporates convolutional neural networks (CNNs), bidirectional gated recurrent units (BiGRUs), attention mechanisms, and ensemble learning. To address the class imbalance, Synthetic …

Advancing Chronic Kidney Disease Prediction Through Machine Learning And Deep Learning With Feature Analysis, Shiddarth Dey Tusar, S. M. Ahad Ali Chowdhury, Md. Jalal Uddin Chowdhury, Rana M. Pir, H. M. Nur A. Alam, Muhammad Rezaur Rahman, Md. Nural Absar Siddiky, Muhammad Enayetur Rahman

Electrical & Computer Engineering Faculty Publications

Chronic Kidney Diesease (CKD) is a significant health issue, ranking as the fourth leading cause of mortality worldwide. The traditional diagnosis and treatment process, reliant on medical experts, is time-consuming. Therefore, thereis an urgent need for more efficient diagnostic methods to improve patient outcomes and reduce mortality rates. In this study, we employ Machine Learning (ML) and Deep Learning (DL) techniques to predict CKD based on important features. Feature analysis was performed using a correlation matrix and the LASSO algo-rithm to identify the most relevant features for model training. We evaluated several ML and DL classifiers, including Logistic Regression (LR), …

Comparative Analysis Of Machine Learning Models For Predicting Healthcare Traffic: Insights For Optimized Emergency Response, Shadman Mahmood Khan Pathan, Sakan Binte Imran, M. M. Shabab Iqbal, Muhammad Enayetur Rahman, Md. Nurul Absar Siddiky, Muhammad Rezaur Rahman, Md Rafid Hasan, Nondon Lal Dey, Md Sobuj Hossain

Electrical & Computer Engineering Faculty Publications

Efficient management of healthcare traffic is crucial for ensuring timely access to medical services, particularly in emergency situations where delays can have severe consequences. This study presents a comparative analysis of three widely used machine learning models—Linear Regression, Decision Trees, and Random Forests—aimed at predicting healthcare-related traffic volumes. A large dataset from a metropolitan traffic system was used to train and evaluate the models based on key performance indicators, including Mean Squared Error (MSE), R² Score, and computational efficiency. The results reveal that the Random Forest model offers the best performance, achieving higher predictive accuracy and faster execution times compared …

Ai And Ml-Based Risk Assessment Of Chemicals: Predicting Carcinogenic Risk From Chemical-Induced Genomic Instability, Ajay Vikram Singh, Preeti Bhardwaj, Peter Laux, Prachi Pradeep, Madleen Busse, Andreas Luch, Akihiko Hirose, Christopher J. Osgood, Michael W. Stacey

Biological Sciences Faculty Publications

Chemical risk assessment plays a pivotal role in safeguarding public health and environmental safety by evaluating the potential hazards and risks associated with chemical exposures. In recent years, the convergence of artificial intelligence (AI), machine learning (ML), and omics technologies has revolutionized the field of chemical risk assessment, offering new insights into toxicity mechanisms, predictive modeling, and risk management strategies. This perspective review explores the synergistic potential of AI/ML and omics in deciphering clastogen-induced genomic instability for carcinogenic risk prediction. We provide an overview of key findings, challenges, and opportunities in integrating AI/ML and omics technologies for chemical risk assessment, …

A Survey On Few-Shot Class-Incremental Learning, Songsong Tian, Lusi Li, Weijun Li, Hang Ran, Xin Ning, Prayag Tiwari

Computer Science Faculty Publications

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup can easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental …

Triphlapan: Predicting Hla Molecules Binding Peptides Based On Triple Coding Matrix And Transfer Learning, Meng Wang, Chuqi Lei, Jianxin Wang, Yaohang Li, Min Li

Computer Science Faculty Publications

Human leukocyte antigen (HLA) recognizes foreign threats and triggers immune responses by presenting peptides to T cells. Computationally modeling the binding patterns between peptide and HLA is very important for the development of tumor vaccines. However, it is still a big challenge to accurately predict HLA molecules binding peptides. In this paper, we develop a new model TripHLApan for predicting HLA molecules binding peptides by integrating triple coding matrix, BiGRU + Attention models, and transfer learning strategy. We have found the main interaction site regions between HLA molecules and peptides, as well as the correlation between HLA encoding and binding …

Autonomous Strike Uavs In Support Of Homeland Security Missions: Challenges And Preliminary Solutions, Meshari Aljohani, Ravi Mukkamala, Stephan Olariu

Computer Science Faculty Publications

Unmanned Aerial Vehicles (UAVs) are becoming crucial tools in modern homeland security applications, primarily because of their cost-effectiveness, risk reduction, and ability to perform a wider range of activities. This study focuses on the use of autonomous UAVs to conduct, as part of homeland security applications, strike missions against high-value terrorist targets. Owing to developments in ledger technology, smart contracts, and machine learning, activities formerly carried out by professionals or remotely flown UAVs are now feasible. Our study provides the first in-depth analysis of the challenges and preliminary solutions for the successful implementation of an autonomous UAV mission. Specifically, we …

The Impacts Of Impurities And Stress State On Polycrystalline Ice Deformation And Investigations Into The Densification Of Polar Firn, Ayobami Oladapo Ogunmolasuyi

Dartmouth College Ph.D Dissertations

Glacier ice cores are unparalleled natural archives of past climate, containing ancient air and impurities. Understanding glacier and ice sheet behavior and their impact on impurity records is crucial for interpreting ice core records. H₂SO₄ is particularly significant, as it provides the strongest evidence for the diffusion of chemical signals and decreases the strength of single crystals and polycrystals of ice. However, its effects on recrystallization, grain growth, and fabric development mechanisms, as well as the effects of grain boundaries on its post-depositional diffusion remain unclear.

This thesis investigates the effects of the microstructural location of H …

Designing A Use-Error Robust Machine Learning Model For Quantitative Analysis Of Diffuse Reflectance Spectra, Allison Scarbrough, Keke Chen, Bing Yu

Biomedical Engineering Faculty Research and Publications

Significance

Machine learning (ML)-enabled diffuse reflectance spectroscopy (DRS) is increasingly used as an alternative to the computation-intensive inverse Monte Carlo (MCI) simulation to predict tissue’s optical properties, including the absorption coefficient,μaand reduced scattering coefficient, μs′.

Aim

We aim to develop a use-error-robust ML algorithm for optical property prediction from DRS spectra.

Approach

We developed a wavelength-independent regressor (WIR) to predict optical properties from DRS data. For validation, we generated 1520 simulated DRS spectra with the forward Monte Carlo model, where μa=0.44 to 2.45cm−1, and μs′=6.53 to 9.58cm−1. We introduced common use-errors, such as wavelength miscalibrations and intensity fluctuations. Finally, we …

Distributed Intelligence: Exploring Federated Paradigms Across Computing And Networking On The Edge, Yang Deng

Dissertations

Computer Graphics (CG) revolves around virtual content creation using computational methods, spanning applications from games to visual effects. Typically, the creation of CG content is led by expert practitioners who guide computational algorithms towards satisfactory results. Thus, creating CG content often requires manual iterations encompassing algorithm design, parameter tuning, and aesthetic feedback. This work investigates how to leverage crowd-sourcing to streamline such creation processes, focusing on animation and simulation. In animation, a novel crowd-sourcing framework is proposed for combat animation, enabling users to analyze motion similarities, and retrieve matching motions using novel crowd-sourced motion features. Such features enable quantifying previously …

Enhancing Wettability Prediction In The Presence Of Organics For Hydrogen Geo-Storage Through Data-Driven Machine Learning Modeling Of Rock/H2/Brine Systems, Zeeshan Tariq, Muhammad Ali, Nurudeen Yekeen, Auby Baban, Bicheng Yan, Shuyu Sun, Hussein Hoteit

Research outputs 2022 to 2026

The success of geological H2 storage relies significantly on rock–H2–brine interactions and wettability. Experimentally assessing the H2 wettability of storage/caprocks as a function of thermos-physical conditions is arduous because of high H2 reactivity and embrittlement damages. Data-driven machine learning (ML) modeling predictions of rock–H2–brine wettability are less strenuous and more precise. They can be conducted at geo-storage conditions that are impossible or hazardous to attain in the laboratory. Thus, ML models were utilized in this research to accurately model the wettability behavior of a ternary system consisting of H2, rock minerals (quartz and mica), and brine at different operating geological …