Computer Sciences Commons^™

Machine Learning And Scalable Informatics Methods To Predict Disease Status From Multimodal Biomedical Data, Hossein Mohammadian Foroushani

McKelvey School of Engineering Graduate Student Theses & Dissertations

Biological understanding of complex diseases such as stroke and obesity is critical for the advancement of medicine. Further knowledge discovery can provide effective biomarkers to improve disease diagnosis and prognosis, identify driver mutations, predict individual genetic susceptibility for early prevention and effective disease management, and facilitate development of personalized drugs. Stroke is the second leading cause of death and long-term disability in the world. Thus, stroke management is a time-sensitive emergency. The initial hours after stroke onset map the trajectory of subsequent neurologic complications. Cerebral edema develops hours to days after acute ischemic stroke and may result in midline shift …

Smart Sensing And Clinical Predictions With Wearables: From Physiological Signals To Mental Health, Ruixuan Dai

McKelvey School of Engineering Graduate Student Theses & Dissertations

Wearable devices such as smartwatches and wristbands are gaining adoption. Recent advances in technology in wearables enable remote health monitoring. However, there are challenges in exploiting wearables in healthcare applications. First, sensor readings from wearables are vulnerable to motion and noise artifacts. A robust pipeline is needed to extract reliable measurements from noisy signals. Second, while wearables support an increasing number of sensing modalities, there is a significant need to generate more clinically meaningful measurements with wearables. Finally, to incorporate wearables into clinical practice, we need to establish the link between wearable measurements and clinical outcomes, thus supporting clinical decisions. …

Development Of The Assessment Of Clinical Prediction Model Transportability (Apt) Checklist, Sean Chonghwan Yu

McKelvey School of Engineering Graduate Student Theses & Dissertations

Clinical Prediction Models (CPM) have long been used for Clinical Decision Support (CDS) initially based on simple clinical scoring systems, and increasingly based on complex machine learning models relying on large-scale Electronic Health Record (EHR) data. External implementation – or the application of CPMs on sites where it was not originally developed – is valuable as it reduces the need for redundant de novo CPM development, enables CPM usage by low resource organizations, facilitates external validation studies, and encourages collaborative development of CPMs. Further, adoption of externally developed CPMs has been facilitated by ongoing interoperability efforts in standards, policy, and …

Dynamic Continuous Distributed Constraint Optimization Problems, Khoi Hoang

McKelvey School of Engineering Graduate Student Theses & Dissertations

The Distributed Constraint Optimization Problem (DCOP) formulation is a powerful tool to model multi-agent coordination problems that are distributed by nature. The formulation is suitable for problems where the environment does not change over time and where agents seek their value assignment from a discrete domain. However, in many real-world applications, agents often interact in a more dynamic environment and their variables usually require a more complex domain. Thus, the DCOP formulation lacks the capabilities to model the problems in such dynamic and complex environments. To address these limitations, researchers have proposed Dynamic DCOPs (D-DCOPs) to model how DCOPs dynamically …

Human-Centered Machine Learning: Algorithm Design And Human Behavior, Wei Tang

McKelvey School of Engineering Graduate Student Theses & Dissertations

Machine learning is increasingly engaged in a large number of important daily decisions and has great potential to reshape various sectors of our modern society. To fully realize this potential, it is important to understand the role that humans play in the design of machine learning algorithms and investigate the impacts of the algorithm on humans.

Towards the understanding of such interactions between humans and algorithms, this dissertation takes a human-centric perspective and focuses on investigating the interplay between human behavior and algorithm design. Accounting for the roles of humans in algorithm design creates unique challenges. For example, humans might …

Design And Analysis Of Strategic Behavior In Networks, Sixie Yu

McKelvey School of Engineering Graduate Student Theses & Dissertations

Networks permeate every aspect of our social and professional life.A networked system with strategic individuals can represent a variety of real-world scenarios with socioeconomic origins. In such a system, the individuals' utilities are interdependent---one individual's decision influences the decisions of others and vice versa. In order to gain insights into the system, the highly complicated interactions necessitate some level of abstraction. To capture the otherwise complex interactions, I use a game theoretic model called Networked Public Goods (NPG) game. I develop a computational framework based on NPGs to understand strategic individuals' behavior in networked systems. The framework consists of three …

Model-Based Deep Learning For Computational Imaging, Xiaojian Xu

McKelvey School of Engineering Graduate Student Theses & Dissertations

This dissertation addresses model-based deep learning for computational imaging. The motivation of our work is driven by the increasing interests in the combination of imaging model, which provides data-consistency guarantees to the observed measurements, and deep learning, which provides advanced prior modeling driven by data. Following this idea, we develop multiple algorithms by integrating the classical model-based optimization and modern deep learning to enable efficient and reliable imaging. We demonstrate the performance of our algorithms by validating their performance on various imaging applications and providing rigorous theoretical analysis.

The dissertation evaluates and extends three general frameworks, plug-and-play priors (PnP), regularized …

Geometric Algorithms For Modeling Plant Roots From Images, Dan Zeng

McKelvey School of Engineering Graduate Student Theses & Dissertations

Roots, considered as the ”hidden half of the plant”, are essential to a plant’s health and pro- ductivity. Understanding root architecture has the potential to enhance efforts towards im- proving crop yield. In this dissertation we develop geometric approaches to non-destructively characterize the full architecture of the root system from 3D imaging while making com- putational advances in topological optimization. First, we develop a global optimization algorithm to remove topological noise, with applications in both root imaging and com- puter graphics. Second, we use our topology simplification algorithm, other methods from computer graphics, and customized algorithms to develop a high-throughput …

Tfa Inference: Using Mathematical Modeling Of Gene Expression Data To Infer The Activity Of Transcription Factors, Cynthia Ma

McKelvey School of Engineering Graduate Student Theses & Dissertations

Transcription factors (TFs) are a set of proteins that play a key role in the information processing system that enables a cell to respond to changes in internal and external state. By binding near a gene in a cell’s DNA, a TF can influence that gene’s expression level, triggering the appropriate increase or decrease in production levels of proteins that are needed to handle stressors like a change in nutrient availability or damage to the cell’s internal structures. Transcription factor activity (TFA) is a measure of how much effect a TF has on its target genes in a given sample …

Scheduling For High Throughput And Small Latency In Parallel And Distributed Systems, Zhe Wang

McKelvey School of Engineering Graduate Student Theses & Dissertations

Parallel and distributed systems are pervasive, such as web services, clouds, and cyber-physical systems. We often desire high throughput and small latency in the parallel and distributed system. However, since the system is distributed and the input is online, scheduling for high throughput while keeping the latency small is often challenging. In this dissertation, we developed scheduling algorithms, policies, and mechanisms to approach high throughput with small latency in various parallel and distributed applications. First, we developed AMCilk runtime system for running multi-programmed parallel jobs on many-processor machines. When running parallel jobs, the allocation of processors to the parallel jobs …

Integrating Physical Models And Deep Priors For Computational Imaging, Yu Sun

McKelvey School of Engineering Graduate Student Theses & Dissertations

This dissertation addresses integrating physical models and learning priors for computational imaging. The motivation of our work is driven by the recent discussion of learning-based methods that solve the imaging inverse problem by directly learning a measurement-to-image mapping from the existing data: they achieve superior performance over the traditional model-based methods but lack the physical model to impose sufficient interpretation and guarantee of the final image. We adopt the classic statistical inference as the underlying formulation and integrate learning models as implicit image priors, such that our framework is able to simultaneously leverage physical models and learning priors. Additionally, the …

Dataset Evaluation For Data Trading Using Expected Loss And Homomorphic Encryption, Minsung Joo

Senior Honors Papers / Undergraduate Theses

Supervised machine learning suffers from the ``garbage-in garbage-out" phenomenon where the performance of a model is limited by the quality of the data. While a myriad of data is collected every second, there is no general rigorous method of evaluating the quality of a given dataset. This hinders fair pricing of data in scenarios where a buyer may look to buy data for use with machine learning. In this work, I propose using the expected loss corresponding to a dataset as a measure of its quality, relying on Bayesian methods for uncertainty quantification. Furthermore, I present a secure multi-party computation …

The Challenges Of Applying Computational Legal Analysis To Mhealth Security And Privacy Regulations, Brian Tung

McKelvey School of Engineering Graduate Student Theses & Dissertations

As our world has grown in complexity, so have our laws. By one measure, the United States Code has grown over 30x as long since 1935, and the 186,000-page Code of Federal Regulations has grown almost 10x in length since 1938. Our growing legal system is too complicated; it’s impossible for people to know all the laws that apply to them. However, people are still subject to the law, even if they are unfamiliar with it. Therein lies the need for computational legal analysis. Tools of computation (e.g., data visualization, algorithms, and artificial intelligence) have the potential to transform civic …

Photoacoustic Imaging, Feature Extraction, And Machine Learning Implementation For Ovarian And Colorectal Cancer Diagnosis, Eghbal Amidi

McKelvey School of Engineering Graduate Student Theses & Dissertations

Among all cancers related to women’s reproductive systems, ovarian cancer has the highest mortality rate. Pelvic examination, transvaginal ultrasound (TVUS), and blood testing for cancer antigen 125 (CA-125), are the conventional screening tools for ovarian cancer, but they offer very low specificity. Other tools, such as magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET), also have limitations in detecting small lesions. In the USA, considering men and women separately, colorectal cancer is the third most common cause of death related to cancer; for men and women combined, it is the second leading cause of cancer deaths. …

A Neuromorphic Machine Learning Framework Based On The Growth Transform Dynamical System, Ahana Gangopadhyay

McKelvey School of Engineering Graduate Student Theses & Dissertations

As computation increasingly moves from the cloud to the source of data collection, there is a growing demand for specialized machine learning algorithms that can perform learning and inference at the edge in energy and resource-constrained environments. In this regard, we can take inspiration from small biological systems like insect brains that exhibit high energy-efficiency within a small form-factor, and show superior cognitive performance using fewer, coarser neural operations (action potentials or spikes) than the high-precision floating-point operations used in deep learning platforms. Attempts at bridging this gap using neuromorphic hardware has produced silicon brains that are orders of magnitude …

Preference Elicitation In Constraint-Based Models: Models, Algorithms, And Applications, Atena M. Tabakhi

McKelvey School of Engineering Graduate Student Theses & Dissertations

Constraint-based models offer powerful approaches for describing and resolving many combinatorial optimization problems in a centralized or distributed environment. In such models, the goal is to find a value assignment to a set of variables given a set of preferences expressed by means of cost functions such that the sum over all costs is optimized. The importance of constraint-based models is outlined by the impact of their applications in a wide range of agent-based systems. Many real-life combinatorial problems can be naturally formalized using constraint-based models. Examples of such applications are supply-chain management, roster scheduling, meeting scheduling, combinatorial auctions, bioinformatics, …

Reasoning About Scene And Image Structure For Computer Vision, Zhihao Xia

McKelvey School of Engineering Graduate Student Theses & Dissertations

The wide availability of cheap consumer cameras has democratized photography for novices and experts alike, with more than a trillion photographs taken each year. While many of these cameras---especially those on mobile phones---have inexpensive optics and make imperfect measurements, the use of modern computational techniques can allow the recovery of high-quality photographs as well as of scene attributes.

In this dissertation, we explore algorithms to infer a wide variety of physical and visual properties of the world, including color, geometry, reflectance etc., from images taken by casual photographers in unconstrained settings. We specifically focus on neural network-based methods, while incorporating …

Bayesian Quadrature With Prior Information: Modeling And Policies, Henry Chai

McKelvey School of Engineering Graduate Student Theses & Dissertations

Quadrature is the problem of estimating intractable integrals. Such integrals regularly arise in engineering and the natural sciences, especially when Bayesian methods are applied; examples include model evidences, normalizing constants and marginal distributions. This dissertation explores Bayesian quadrature, a probabilistic, model-based quadrature method. Specifically, we study different ways in which Bayesian quadrature can be adapted to account for different kinds of prior information one may have about the task. We demonstrate that by taking into account prior knowledge, Bayesian quadrature can outperform commonly used numerical methods that are agnostic to prior knowledge, such as Monte Carlo based integration. We focus …

Continuous-Time And Complex Growth Transforms For Analog Computing And Optimization, Oindrila Chatterjee

McKelvey School of Engineering Graduate Student Theses & Dissertations

Analog computing is a promising and practical candidate for solving complex computational problems involving algebraic and differential equations. At the fundamental level, an analog computing framework can be viewed as a dynamical system that evolves following fundamental physical principles, like energy minimization, to solve a computing task. Additionally, conservation laws, such as conservation of charge, energy, or mass, provide a natural way to couple and constrain spatially separated variables. Taking a cue from these observations, in this dissertation, I have explored a novel dynamical system-based computing framework that exploits naturally occurring analog conservation constraints to solve a variety of optimization …

Machine Learning For Analog/Mixed-Signal Integrated Circuit Design Automation, Weidong Cao

McKelvey School of Engineering Graduate Student Theses & Dissertations

Analog/mixed-signal (AMS) integrated circuits (ICs) play an essential role in electronic systems by processing analog signals and performing data conversion to bridge the analog physical world and our digital information world.Their ubiquitousness powers diverse applications ranging from smart devices and autonomous cars to crucial infrastructures. Despite such critical importance, conventional design strategies of AMS circuits still follow an expensive and time-consuming manual process and are unable to meet the exponentially-growing productivity demands from industry and satisfy the rapidly-changing design specifications from many emerging applications. Design automation of AMS IC is thus the key to tackling these challenges and has been …

Machine Learning In Complex Scientific Domains: Hospitalization Records, Drug Interactions, Predictive Modeling And Fairness For Class Imbalanced Data, Arghya Datta

McKelvey School of Engineering Graduate Student Theses & Dissertations

Machine learning has demonstrated potential in analyzing large, complex datasets and has become ubiquitous across many fields of scientific research. As machine learning is actively deployed in many complex and critical domains, it is essential for machine learning to engage with domain expertise to aid in knowledge discovery as well as address challenges in predictive modeling in complex domains. Domain expertise represents an essential and elaborate collection of knowledge that is often under-utilized when applying machine learning in complex domains. In this dissertation, I have addressed existing challenges regarding knowledge discovery in complex domains via engagement with domain expertise, particularly …

Improving Additional Adversarial Robustness For Classification, Michael Guo

McKelvey School of Engineering Graduate Student Theses & Dissertations

Although neural networks have achieved remarkable success on classification, adversarial robustness is still a significant concern. There are now a series of approaches for designing adversarial examples and methods to defending against them. This paper consists of two projects. In our first work, we propose an approach by leveraging cognitive salience to enhance additional robustness on top of these methods. Specifically, for image classification, we split an image into the foreground (salient region) and background (the rest) and allow significantly larger adversarial perturbations in the background to produce stronger attacks. Furthermore, we show that adversarial training with dual-perturbation attacks yield …

A Collaborative Knowledge-Based Security Risk Assessments Solution Using Blockchains, Tara Thaer Salman

McKelvey School of Engineering Graduate Student Theses & Dissertations

Artificial intelligence and machine learning have recently gained wide adaptation in building intelligent yet simple and proactive security risk assessment solutions. Intrusion identification, malware detection, and threat intelligence are examples of security risk assessment applications that have been revolutionized with these breakthrough technologies. With the increased risk and severity of cyber-attacks and the distributed nature of modern threats and vulnerabilities, it becomes critical to pose a distributed intelligent assessment solution that evaluates security risks collaboratively. Blockchain, as a decade-old successful distributed ledger technology, has the potential to build such collaborative solutions. However, in order to be used for such solutions, …

Real-Time Virtualization And Coordination For Edge Computing, Haoran Li

McKelvey School of Engineering Graduate Student Theses & Dissertations

Recent years have witnessed the emergence of edge computing as an enabling platform for time-sensitive services. However, existing edge computing platforms face a multitude of challenges in meeting the latency requirements of time-sensitive applications. (1) Traditional real-time virtualization platforms require offline configuration of the scheduling parameters of virtual machines (VMs) based on their worst-case workloads. However, this static approach results in pessimistic resource allocation when the workloads in the VMs change dynamically. (2) Edge computing operators must deliver consistent tail latency performance for time-sensitive applications deployed on different edge sites. Traditionally, significant effort is required to test, tune, and configure …

Stochastic Goal Recognition Design, Christabel Wayllace

McKelvey School of Engineering Graduate Student Theses & Dissertations

Goal Recognition Design (GRD) is the problem of finding the least amount of environment modifications to force an acting agent to reveal its goal as early as possible. Figuring out an agent’s goal by observing its behavior is a problem studied in Psychology, Economics, and Artificial Intelligence, where it is known as goal recognition. Contrary to most common approaches where the focus is on finding faster algorithms to detect the goal, GRD takes an offline approach and focuses on environment design to facilitate goal recognition. This thesis investigates GRD problems when action outcomes are stochastic, which is the case of …

Data Driven Architectural Geometric And Photometric Modeling, Huayi Zeng

McKelvey School of Engineering Graduate Student Theses & Dissertations

This thesis presents novel algorithms of architectural modeling, a crucial computer vision task to understand architectures by parsing visual input such as image or sensor data into digital representations. Compelling modeling algorithms serve as the fundamental block for a wide range of applications, including augmented reality, digital mapping, virtual simulation. While numerous handcrafted approaches have been proposed, the problem remains challenging in various difficult cases, such as occlusions and imperfect input. This dissertation studies four novel data-driven methods to perform high quality modeling. I first propose two geometric architectural modeling algorithms to recover the geometric primitives from aerial and facade …

Towards Deploying Robust Machine Learning Systems, Liang Tong

McKelvey School of Engineering Graduate Student Theses & Dissertations

Machine learning (ML) has come to be widely used in a broad array of settings, including important security applications such as network intrusion, fraud, and malware detection, as well as other high-stakes settings, such as autonomous driving. A general approach is to extract a set of features, or numerical attributes, of entities in question, collect a training data set of labeled examples (for example, indicating which instances are malicious and which are benign), learn a model which labels previously unseen instances presented in terms of their extracted features, and then investigate alerts raised by instances predicted as malicious. Despite the …

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

McKelvey School of Engineering Graduate Student 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 …

Deep Learning For Task-Based Image Quality Assessment In Medical Imaging, Weimin Zhou

McKelvey School of Engineering Graduate Student Theses & Dissertations

It has been advocated to use objective measures of image quality (IQ) for assessing and optimizing medical imaging systems. Objective measures of IQ quantify the performance of an observer at a specific diagnostic task. Binary signal detection tasks and joint signal detection and localization (detection-localization) tasks are commonly considered in medical imaging. When optimizing imaging systems for binary signal detection tasks, the performance of the Bayesian Ideal Observer (IO) has been advocated for use as a figure-of-merit (FOM). The IO maximizes the observer performance that is summarized by the receiver operating characteristic (ROC) curve. When signal detection-localization tasks are considered, …

Machine Learning Morphisms: A Framework For Designing And Analyzing Machine Learning Work Ows, Applied To Separability, Error Bounds, And 30-Day Hospital Readmissions, Eric Zenon Cawi

McKelvey School of Engineering Graduate Student Theses & Dissertations

A machine learning workflow is the sequence of tasks necessary to implement a machine learning application, including data collection, preprocessing, feature engineering, exploratory analysis, and model training/selection. In this dissertation we propose the Machine Learning Morphism (MLM) as a mathematical framework to describe the tasks in a workflow. The MLM is a tuple consisting of: Input Space, Output Space, Learning Morphism, Parameter Prior, Empirical Risk Function. This contains the information necessary to learn the parameters of the learning morphism, which represents a workflow task. In chapter 1, we give a short review of typical tasks present in a workflow, as …