Engineering Commons^™

Predicting Patient Outcomes With Machine Learning For Diverse Health Data, Dingwen Li

McKelvey School of Engineering Theses & Dissertations

As digitized clinical and health data become ubiquitous, machine learning techniques have shown promise in predicting various clinical outcomes. In this thesis research, we exploit three types of data including (1) data collected through wearables outside hospitals, (2) electronic health records (EHR) data of inpatient in general hospital wards, (3) intraoperative data collected during surgery. This thesis work investigates machine learning approaches for the diverse clinical and health data with distinctive characteristics and challenges in the context of real-world clinical applications. Specifically, this thesis makes the following contributions to the state of the art of clinical machine learning.

Extracting informative …

Design Of Plastic Contaminant Eliminator In Seed Cotton, Joshua H. Tandio

Theses and Dissertations

Plastic contamination in cotton is a problem in cotton industry and researchers have worked on this problem with different approaches. This thesis documents the design of mechanical and electronic real-time systems for detecting and removing plastic contaminants. The mechanical system was designed to expose plastic embedded inside the seed cotton to the sensor and to separate plastic contaminated cotton from the process stream. The detection system consisted of an embedded computer interfaced with a USB camera and Neural Network (NN) software running in it. Two NN models were tested, a transfer learning model and a built-from-scratch original model. The original …

A Deep Recurrent Neural Network With Iterative Optimization For Inverse Image Processing Applications, Masaki Ikuta

Theses and Dissertations

Many algorithms and methods have been proposed for inverse image processing applications, such as super-resolution, image de-noising, and image reconstruction, particularly with the recent surge of interest in machine learning and deep learning methods.

As for Computed Tomography (CT) image reconstruction, the most recently proposed methods are limited to image domain processing, where deep learning is used to learn the mapping between a true image data set and a noisy image data set in the image domain. While deep learning-based methods can produce higher quality images than conventional model-based algorithms, these methods have a limitation. Deep learning-based methods used in …

Machine Learning For Unmanned Aerial System (Uas) Networking, Jian Wang

Doctoral Dissertations and Master's Theses

Fueled by the advancement of 5G new radio (5G NR), rapid development has occurred in many fields. Compared with the conventional approaches, beamforming and network slicing enable 5G NR to have ten times decrease in latency, connection density, and experienced throughput than 4G long term evolution (4G LTE). These advantages pave the way for the evolution of Cyber-physical Systems (CPS) on a large scale. The reduction of consumption, the advancement of control engineering, and the simplification of Unmanned Aircraft System (UAS) enable the UAS networking deployment on a large scale to become feasible. The UAS networking can finish multiple complex …

Analysis Of Deep Learning Methods For Wired Ethernet Physical Layer Security Of Operational Technology, Lucas Torlay

All Theses

The cybersecurity of power systems is jeopardized by the threat of spoofing and man-in-the-middle style attacks due to a lack of physical layer device authentication techniques for operational technology (OT) communication networks. OT networks cannot support the active probing cybersecurity methods that are popular in information technology (IT) networks. Furthermore, both active and passive scanning techniques are susceptible to medium access control (MAC) address spoofing when operating at Layer 2 of the Open Systems Interconnection (OSI) model. This thesis aims to analyze the role of deep learning in passively authenticating Ethernet devices by their communication signals. This method operates at …

Evaluation Of Robust Deep Learning Pipelines Targeting Low Swap Edge Deployment, David Carter Cornett

Masters Theses

The deep learning technique of convolutional neural networks (CNNs) has greatly advanced the state-of-the-art for computer vision tasks such as image classification and object detection. These solutions rely on large systems leveraging wattage-hungry GPUs to provide the computational power to achieve such performance. However, the size, weight and power (SWaP) requirements of these conventional GPU-based deep learning systems are not suitable when a solution requires deployment to so called "Edge" environments such as autonomous vehicles, unmanned aerial vehicles (UAVs) and smart security cameras.

The objective of this work is to benchmark FPGA-based alternatives to conventional GPU systems that have the …

Learning State-Dependent Sensor Measurement Models To Improve Robot Localization Accuracy, Troi André Williams

USF Tampa Graduate Theses and Dissertations

This dissertation proposes a novel method called state-dependent sensor measurement models (SDSMMs). Such models dynamically predict the state-dependent bias and uncertainty of sensor measurements, ultimately improving fundamental robot tasks such as localization. In our first investigation, we introduced the state-dependent sensor measurement model framework, described their properties, stated the input and output of these models, and described how to train them. We also explained how to integrate such models with an Extended Kalman Filter and a Particle Filter, two popular robot state estimation algorithms. We validated the proposed framework through a series of localization tasks. The results showed that our …

Numerical Modeling Of Advanced Propulsion Systems, Peetak P. Mitra

Doctoral Dissertations

Numerical modeling of advanced propulsion systems such as the Internal Combustion Engine (ICE) is of great interest to the community due to the magnitude of compute/algorithmic challenges. Fuel spray atomization, which determines the rate of fuel-air mixing, is a critical limiting process for the phenomena of combustion within ICEs. Fuel spray atomization has proven to be a formidable challenge for the state-of-the-art numerical models due to its highly transient, multi-scale, and multi-phase nature. Current models for primary atomization employ a high degree of empiricism in the form of model constants. This level of empiricism often reduces the art of predictive …

Deep Learning For High-Impedance Fault Detection And Classification, Khushwant Rai

Electronic Thesis and Dissertation Repository

High-Impedance Faults (HIFs) are a hazard to public safety but are difficult to detect because of their low current amplitude and diverse characteristics. Supervised machine learning techniques have shown great success in HIF detection; however, these approaches rely on resource-intensive signal processing techniques and fail in presence of non-HIF disturbances and even for scenarios not included in training data. This thesis leverages unsupervised learning and proposes a Convolutional Autoencoder framework for HIF Detection (CAE-HIFD). In CAE-HIFD, Convolutional Autoencoder learns only from HIF signals by employing cross-correlation; consequently, eliminating the need for diverse non-HIF scenarios in training. Furthermore, this thesis proposes …

Forecasting Pedestrian Trajectory Using Deep Learning, Arsal Syed

UNLV Theses, Dissertations, Professional Papers, and Capstones

In this dissertation we develop different methods for forecasting pedestrian trajectories. Complete understanding of pedestrian motion is essential for autonomous agents and social robots to make realistic and safe decisions. Current trajectory prediction methods rely on incorporating historic motion, scene features and social interaction to model pedestrian behaviors. Our focus is to accurately understand scene semantics to better forecast trajectories. In order to do so, we leverage semantic segmentation to encode static scene features such as walkable paths, entry/exits, static obstacles etc. We further evaluate the effectiveness of using semantic maps on different datasets and compare its performance with already …

Medical Image Segmentation Using Machine Learning, Masoud Khani

Theses and Dissertations

Image segmentation is the most crucial step in image processing and analysis. It can divide an image into meaningfully descriptive components or pathological structures. The result of the image division helps analyze images and classify objects. Therefore, getting the most accurate segmented image is essential, especially in medical images. Segmentation methods can be divided into three categories: manual, semiautomatic, and automatic. Manual is the most general and straightforward approach. Manual segmentation is not only time-consuming but also is imprecise. However, automatic image segmentation techniques, such as thresholding and edge detection, are not accurate in the presence of artifacts like noise …

Hardware For Quantized Mixed-Precision Deep Neural Networks, Andres Rios

Open Access Theses & Dissertations

Recently, there has been a push to perform deep learning (DL) computations on the edge rather than the cloud due to latency, network connectivity, energy consumption, and privacy issues. However, state-of-the-art deep neural networks (DNNs) require vast amounts of computational power, data, and energyâ??resources that are limited on edge devices. This limitation has brought the need to design domain-specific architectures (DSAs) that implement DL-specific hardware optimizations. Traditionally DNNs have run on 32-bit floating-point numbers; however, a body of research has shown that DNNs are surprisingly robust and do not require all 32 bits. Instead, using quantization, networks can run on …

Adaptive Mobile Eeg Noise Cancellation Using 2d Convolutional Autoencoders For Bci Authentication, Tyree Lewis

USF Tampa Graduate Theses and Dissertations

Electroencephalography (EEG) signals can be used for many purposes and has the potential to be adapted to various systems. When EEG is recorded from users, these studies are performed primarily in an indoor environment, while the user is stationary. This is due to the levels of noise that are experienced when recording EEG data, to minimize errors in the data. This thesis aims to adapt tasks that are performed indoors to an external environment by removing both noise and artefacts in EEG, using a 2D Convolutional Autoencoder (CAE). The data is recorded from subjects is passed into the 2D CAE …

Signal Processing And Data Analysis For Real-Time Intermodal Freight Classification Through A Multimodal Sensor System., Enrique J. Sanchez Headley

Graduate Theses and Dissertations

Identifying freight patterns in transit is a common need among commercial and municipal entities. For example, the allocation of resources among Departments of Transportation is often predicated on an understanding of freight patterns along major highways. There exist multiple sensor systems to detect and count vehicles at areas of interest. Many of these sensors are limited in their ability to detect more specific features of vehicles in traffic or are unable to perform well in adverse weather conditions. Despite this limitation, to date there is little comparative analysis among Laser Imaging and Detection and Ranging (LIDAR) sensors for freight detection …

Improving Presentation Consistency Of Radiographic Images Using Deep Learning, Najib Akram Maheen Aboobacker

Master's Theses (2009 -)

In general radiography, inconsistency of brightness and contrast in initial presentation is a common complaint from radiologists. Inconsistencies, which may be a result of variations in patient positioning, dose, protocol selection andimplant could lead to additional workflow by technologists and radiologists to adjust the images. To tackle this challenge posed by conventional histogram-based display approach, first a Deep Learning based Brightness Contrast (DL BC) algorithm to improve the consistency in presentation by using a residual neural network trained to classify X-ray images based on a novel NxM grid of brightness and contrast combinations is proposed. More than 8,500 unique images …

Applying Deep Learning On Financial Sentiment Analysis, Cuiyuan Wang

Dissertations, Theses, and Capstone Projects

Portfolio Investment has always been appealing to investors and researchers. In the past, people tend to use historical trading information of the securities to predict the return or manage the portfolio. Nowadays, the literature has been proved that the market sentiment could predict asset prices. Specifically, it has been shown that the stock market movement is related to financial news and social media events. Thus, it becomes necessary to extract the sentiment of the financial news. We explicitly introduce the application of dictionary methods, traditional machine learning models and deep learning models on text classification. The experiment results show that …

Soarnet, Deep Learning Thermal Detection For Free Flight, Jake T. Tallman

Master's Theses

Thermals are regions of rising hot air formed on the ground through the warming of the surface by the sun. Thermals are commonly used by birds and glider pilots to extend flight duration, increase cross-country distance, and conserve energy. This kind of powerless flight using natural sources of lift is called soaring. Once a thermal is encountered, the pilot flies in circles to keep within the thermal, so gaining altitude before flying off to the next thermal and towards the destination. A single thermal can net a pilot thousands of feet of elevation gain, however estimating thermal locations is not …

Convolutional Neural Networks For Deflate Data Encoding Classification Of High Entropy File Fragments, Nehal Ameen

University of New Orleans Theses and Dissertations

Data reconstruction is significantly improved in terms of speed and accuracy by reliable data encoding fragment classification. To date, work on this problem has been successful with file structures of low entropy that contain sparse data, such as large tables or logs. Classifying compressed, encrypted, and random data that exhibit high entropy is an inherently difficult problem that requires more advanced classification approaches. We explore the ability of convolutional neural networks and word embeddings to classify deflate data encoding of high entropy file fragments after establishing ground truth using controlled datasets. Our model is designed to either successfully classify file …

Inference And Learning In Spiking Neural Networks For Neuromorphic Systems, Amar Shrestha

Dissertations - ALL

Neuromorphic computing is a computing field that takes inspiration from the biological and physical characteristics of the neocortex system to motivate a new paradigm of highly parallel and distributed computing to take on the demands of the ever-increasing scale and computational complexity of machine intelligence esp. in energy-limited systems such as Edge devices, Internet-of-Things (IOT), and cyber physical systems (CPS). Spiking neural network (SNN) is often studied together with neuromorphic computing as the underlying computational model . Similar to the biological neural system, SNN is an inherently dynamic and stateful network. The state and output of SNN do not only …

Multi-Label/Multi-Class Deep Learning Classification Of Spatiotemporal Data, Natalie Sommer

Dissertations - ALL

Human senses allow for the detection of simultaneous changes in our environments. An unobstructed field of view allows us to notice concurrent variations in different parts of what we are looking at. For example, when playing a video game, a player, oftentimes, needs to be aware of what is happening in the entire scene. Likewise, our hearing makes us aware of various simultaneous sounds occurring around us. Human perception can be affected by the cognitive ability of the brain and acuity of the senses. This is not a factor with machines. As long as a system is given a signal …

Inference And Learning In Spiking Neural Networks For Neuromorphic Systems, Amar Shrestha

Dissertations - ALL

Neuromorphic computing is a computing field that takes inspiration from the biological and physical characteristics of the neocortex system to motivate a new paradigm of highly parallel and distributed computing to take on the demands of the ever-increasing scale and computational complexity of machine intelligence esp. in energy-limited systems such as Edge devices, Internet-of-Things (IOT), and cyber physical systems (CPS). Spiking neural network (SNN) is often studied together with neuromorphic computing as the underlying computational model . Similar to the biological neural system, SNN is an inherently dynamic and stateful network. The state and output of SNN do not only …

Multi-Label/Multi-Class Deep Learning Classification Of Spatiotemporal Data, Natalie Sommer

Dissertations - ALL

Human senses allow for the detection of simultaneous changes in our environments. An unobstructed field of view allows us to notice concurrent variations in different parts of what we are looking at. For example, when playing a video game, a player, oftentimes, needs to be aware of what is happening in the entire scene. Likewise, our hearing makes us aware of various simultaneous sounds occurring around us. Human perception can be affected by the cognitive ability of the brain and acuity of the senses. This is not a factor with machines. As long as a system is given a signal …

Assessment And Diagnosis Of Human Colorectal And Ovarian Cancer Using Optical Imaging And Computer-Aided Diagnosis, Yifeng Zeng

McKelvey School of Engineering Theses & Dissertations

Tissue optical scattering has recently emerged as an important diagnosis parameter associated with early tumor development and progression. To characterize the differences between benign and malignant colorectal tissues, we have created an automated optical scattering coefficient mapping algorithm using an optical coherence tomography (OCT) system. A novel feature called the angular spectrum index quantifies the scattering coefficient distribution. In addition to scattering, subsurface morphological changes are also associated with the development of colorectal cancer. We have observed a specific mucosa structure indicating normal human colorectal tissue, and have developed a real-time pattern recognition neural network to localize this specific structure …

Lecture Video Transformation Through An Intelligent Analysis And Post-Processing System, Xi Wang

Masters Theses

Lecture videos are good sources for people to learn new things. Students commonly use online videos to explore various domains. However, some recorded videos are posted on online platforms without being post-processed due to technology and resource limitations. In this work, we focus on the research of developing an intelligent system to automatically extract essential information, including the main instructor and screen, in a lecture video in several scenarios by using modern deep learning techniques. This thesis aims to combine the extracted essential information to render the videos and generate a new layout with a smaller file size than the …

Towards Secure Deep Neural Networks For Cyber-Physical Systems, Jiangnan Li

Doctoral Dissertations

In recent years, deep neural networks (DNNs) are increasingly investigated in the literature to be employed in cyber-physical systems (CPSs). DNNs own inherent advantages in complex pattern identifying and achieve state-of-the-art performances in many important CPS applications. However, DNN-based systems usually require large datasets for model training, which introduces new data management issues. Meanwhile, research in the computer vision domain demonstrated that the DNNs are highly vulnerable to adversarial examples. Therefore, the security risks of employing DNNs in CPSs applications are of concern.

In this dissertation, we study the security of employing DNNs in CPSs from both the data domain …

Heterogeneous Graph-Based User-Specific Review Helpfulness Prediction, Dongkai Chen

Dartmouth College Master’s Theses

With the popularity of e-commerce and review websites, it is becoming increasingly important to identify the helpfulness of reviews. However, existing works on predicting reviews’ helpfulness have three major issues: (i) the correlation between helpfulness and features from review text is not clear yet, although many standard features are proposed, (ii) the relations between users, reviews and products have not been considered, (iii) the effectiveness of the existing approaches have not been systematically compared. To address these challenges, we first analyze the correlation between standard features and review helpfulness that are widely used in other work. Based on this analysis, …

Machine Learning With Topological Data Analysis, Ephraim Robert Love

Doctoral Dissertations

Topological Data Analysis (TDA) is a relatively new focus in the fields of statistics and machine learning. Methods of exploiting the geometry of data, such as clustering, have proven theoretically and empirically invaluable. TDA provides a general framework within which to study topological invariants (shapes) of data, which are more robust to noise and can recover information on higher dimensional features than immediately apparent in the data. A common tool for conducting TDA is persistence homology, which measures the significance of these invariants. Persistence homology has prominent realizations in methods of data visualization, statistics and machine learning. Extending ML with …

Expanding Image Datasets For Deep Learning By Evaluating Independence Through Coefficient Correlation And Mean-Squared Error, Ayman Yousef

Biomedical Engineering Undergraduate Honors Theses

With deep learning being leveraged more regularly in the field of image classification, particularly in medical imaging, network optimizations have become a field in and of itself. With open source, comprehensive medical image datasets few and far, computational dataset expansion has become a useful tool for researchers without the ability to further manually collect data. However, with the rich amount of data that imaging modalities like multi-photon microscopy collect at a time, there is potential to expand datasets through proper utilization of this data that often time goes unused. Previous deep learning studies have shown that improper expansion can conflate …

Raw Depth Image Enhancement Using A Neural Network, Xuan Xie

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

The term image is often used to denote a data format that records information about a scene’s color. This dissertation object focuses on a similar format for recording distance information about a scene, “depth images”. Depth images have been used extensively in consumer-level applications, such as Apple’s Face ID, based on depth images for face recognition.

However, depth images suffer from low precision and high errors, and some post-processing techniques need to be utilized to improve their quality. Deep learning, or neural networks, are frameworks that use a series of hierarchically arranged nonlinear networks to process input data. Although each …

Deep Learning Based Sound Event Detection And Classification, Alireza Nasiri

Theses and Dissertations

Hearing sense has an important role in our daily lives. During the recent years, there has been many studies to transfer this capability to the computers. In this dissertation, we design and implement deep learning based algorithms to improve the ability of the computers in recognizing the different sound events.

In the first topic, we investigate sound event detection, which identifies the time boundaries of the sound events in addition to the type of the events. For sound event detection, we propose a new method, AudioMask, to benefit from the object-detection techniques in computer vision. In this method, we convert …