Physical Sciences and Mathematics Commons^™

Computational Single-Cell Analysis Of Confocal Fluorescence Images With Dapi-Generated Masks, Munirah Alduhailan

Wayne State University Theses

Lipolysis is a metabolic pathway in which free fatty acids are mobilized from stored triglycerides. The rate-limiting enzyme in this process is adipose triglyceride lipase, which is regulated by α/β-hydrolase domain-containing protein 5 (ABHD5) via both natural and synthetic pathways. With advanced artificial neural networks, image processing methods can extract quantitative results from fluorescence images. The segmentation of complex biological images, in which regions of the image are labeled as distinct masks, is the first step in image analysis. Ilastik, a machine-learning software, performs image segmentation with a user-trained neural network and custom key feature labels. The software’s results are …

Defending Against Adversarial Attacks On Medical Imaging Ai Systems, Xin Li

Wayne State University Theses

Although deep learning systems trained on medical images have shown state-of-the-art performance in many clinical prediction tasks, recent studies demonstrate that these systems can be fooled by carefully crafted adversarial images. It has raised concerns on the practical deployment of deep learning based medical image classification systems. Although an array of defense techniques have been developed and proved to be effective in computer vision, defending against adversarial attacks on medical images remains largely an uncharted territory due to their unique challenges: crafted adversarial noises added to a highly standardized medical image can make it a hard sample for model to …

Novel Design And Facile Synthesis Of Functional Nanostructures For Energy And Environmental Application, Zhuojun Wang

Wayne State University Theses

Environmental pollution and energy crisis are two of the major obstacles in the development of a more sustainable society. It is an interesting but technically challenging task to design novel methods for the facile and scalable synthesis of multifunctional materials for applications in both energy storage and environmental remediation. In this thesis, I tried to design and synthesize novel nanostructured materials for energy and environmental applications, achieved highly promising results. I developed a facile electrochemical approach to synthesize the zinc oxide with the zinc element for photocatalytic reactions. It demonstrated dramatically improved performances in degradation of organic pollutant in water, …

Process Data Analytics Using Deep Learning Techniques, Majid Moradi Aliabadi

Wayne State University Theses

In chemical manufacturing plants, numerous types of data are accessible, which could be process operational data (historical or real-time), process design and product quality data, economic and environmental (including process safety, waste emission and health impact) data. Effective knowledge extraction from raw data has always been a very challenging task, especially the data needed for a type of study is huge. Other characteristics of process data such as noise, dynamics, and highly correlated process parameters make this more challenging.

In this study, we introduce an attention-based RNN for multi-step-ahead prediction that can have applications in model predictive control, fault diagnosis, …

Representation Learning With Autoencoders For Electronic Health Records, Najibesadat Sadatijafarkalaei

Wayne State University Theses

Increasing volume of Electronic Health Records (EHR) in recent years provides great opportunities for data scientists to collaborate on different aspects of healthcare research by applying advanced analytics to these EHR clinical data. A key requirement however

is obtaining meaningful insights from high dimensional, sparse and complex clinical data. Data science approaches typically address this challenge by performing feature learning in order to build more reliable and informative feature representations from clinical data followed by supervised learning. In this research, we propose a predictive modeling approach based on deep feature representations and word embedding techniques. Our method uses different deep …

Characterization, Modeling, And Thermal Management Of High-Performance Lithium Batteries, Minjun Bae

Wayne State University Theses

Lithium-ion (Li-ion) batteries, as one of the most advanced commercial rechargeable batteries, play a crucial role in modern society as they are extensively used in portable electronic devices. Nevertheless, the limited electrochemical performance and poor thermal management systems of Li-ion batteries have hindered the expansion of their future applications. In search of alternative electrode materials to develop a battery with higher electrochemical performance, lithium (Li) metal has attracted much attention as an ideal alternative anode material due to its high specific capacity and lowest redox potential. However, needle-like Li dendritic growth causes severe safety concerns and thus prohibits practical applications …

Methods For Classification Of Consumer Review Into The Ones Written Before Or After The Product Purchase, Md Mehedi Hasan

Wayne State University Theses

Online consumer reviews provide a wealth of information about products and services that, if properly identified and extracted, could be of immense value to businesses. While classification of reviews according to sentiment polarity has been extensively studied in previous work, many more focused types of review analysis remain open problems. In this work, we introduce a novel text classification problem of separating post-purchase from pre-purchase consumer review fragments that can facilitate identification of immediate actionable insights based on the feedback from the customers, who actually purchased and own a product. To address this problem, we propose the features, which are …

Molecular Exchange Monte Carlo. A Generalized Method For Identity Exchanges In Grand Canonical Monte Carlo Simulations, Mohammad Soroush Barhaghi

Wayne State University Theses

A generalized identity exchange algorithm is presented for Monte Carlo simulations in the grand canonical ensemble. The algorithm, referred to as Molecular Exchange Monte Carlo (MEMC), may be applied to multicomponent systems of arbitrary molecular topology, and provides significant enhancements in the sampling of phase space over a wide range of compositions and temperatures. Three different approaches are presented for the insertion of large molecules, and the pros and cons of each method are discussed. The performance of the algorithms is highlighted through grand canonical Monte Carlo histogram-reweighting simulations performed on several systems, including 2,2,4-trimethylpentane+neopentane, butane+perfluorobutane, methane+n-alkanes, and water+impurity. Relative …

Utilizing Log Files For Treatment Planning And Delivery Qa In Radiotherapy, Carl W. Stanhope

Wayne State University Theses

Purpose: Monte Carlo-based log file quality assurance (LF-MC QA) is investigated as an alternative method to phantom-based patient-specific quality assurance in radiotherapy (e.g. ArcCHECK QA (AC QA)).

Methods: First, the shortcomings of AC QA were investigated. The sensitivity dependence of ArcCHECK diodes on dose rate (in-field) and energy (primarily out-of-field) was quantified. LF-MC QA was then analyzed on the phantom geometry. Planned (‘Plan’) and LF-reconstructed CS and MC doses were compared with each other and AC measurement via statistical (mean ± StdDev(σ)) and gamma analyses to isolate dosimetric uncertainties and quantify the relative accuracies of AC QA and LF-MC QA. …

An Empirical Study On Deterministic Collusive Attack Using Inter Component Communication In Android Applications, Tanzeer Hossain

Wayne State University Theses

Security threats using intent based inter component communication (ICC) channels in Android are under constant scrutiny of software engineering researchers. Though prior research provides empirical evidence on the existence of collusive communication channels in popular android apps, little is known about developers’willful involvement and motivation to exploit these channels.To shed light on this matter, in this paper we devised a novel methodology to deterministically identify developers’ involvement in establishing collusive inter app communication channels. We incorporate static analysis and relational database technology to discover sensitive collusive channels and domain knowledge of the Android SDK to build a model to identify …

Uncertainty In Combustion Reaction Rates And Its Effects On Combustion Simulations, Joshua Piehl

Wayne State University Theses

This work investigates the uncertainties in reaction rates of an n-dodecane model on turbulent spray combustion simulations. Six major reactions were found to significantly impact the ignition delay of the mechanism in a 0-D batch reactor model. These reactions’ rates were independently modified and placed into individual mechanisms. These newly developed mechanisms were simulated in a 3-D turbulent spray simulation and a 0-D batch reactor at a pressure of 60 bar and temperatures from 900 to 1100 K. The combustion characteristics (e.g. ignition delays, flame lift-off length, liquid and vapor penetration) of the modified mechanisms were compared to those produced …

The Impact Of Machine Learning Algorithms On Benchmarking Process In Healthcare Service Delivery, Egbe-Etu Emmanuel Etu

Wayne State University Theses

Currently, organizations have adopted and implemented a variety of innovative quality management philosophies, approaches, and techniques to stay competitive in an ever-changing global economy. Benchmarking is one of such techniques deployed by organizations to stay competitive. The motivation for this research stems from a real-world problem being faced by hospitals in the healthcare industry who have amassed a ton of data and want to embark on benchmarking project to assess the performance of the emergency departments due to challenges faced with poor management of operations which has led to high patient boarding rates, high patient wait-times, poor quality service, low …

The Influence Of The Physical Workspace On Creative Performance: Alternative Mediation Models, Reed James Bramble

Wayne State University Theses

The present study assessed how individuals respond to stimuli in their immediate work environment. Specifically, I conducted a laboratory experiment with a student sample to test the influence of natural stimuli in the workspace on individuals’ creative performance. Additionally, I tested a series of potential mediating variables that could have driven the effect: state positive affect, state openness to experience, and self-regulation of effort. While the results indicated that the hypotheses were broadly unsupported, supplemental analyses revealed that the intervention significantly predicted effort regulation in a high-performing subgroup. Study limitations and recommendations for future directions are discussed.

Using 210po/210pb Disequilibria To Characterize The Biogeochemistry And Quantify The Dynamics Of Sea Ice In The Arctic, Katherine Denise Krupp

Wayne State University Theses

Understanding the biogeochemical cycles of the Arctic Ocean is highly significant due to its many unique characteristics which impact the cycling of important trace elements, isotopes, and nutrients. Furthermore, with its seasonally-changing sea ice extent and substantial role in the global radiation balance, the Arctic is considered a sensitive indicator of global change. The study of 210Po and 210Pb disequilibria is used to determine 1) the residence time of ice-rafted sediments in sea ice, 2) ice floe dynamics such as ice accumulation rates, and 3) biogeochemical characterization of the Arctic due to the natural occurrence and particle reactivity of 210Po …

Differential Activation Of Dead Box Rna Helicases Rhlb And Rhle By Hfq/Srnas And Their Target Mrnas, Amit Kumar

Wayne State University Theses

Number of small RNA (sRNA) gene regulators have mounted in E. coli over the years whereas the number of validated protein partners has not changed considerably. Hfq has remained the only well studied global regulatory partner of sRNAs in E. coli. However, direct or indirect involvement of other protein partners has always been speculated. Study from Blasi lab has shown that CsdA, one of the five DEAD-box RNA helicases of E. coli, is required for the DsrA mediated upregulation of rpoS under cold stress condition. Previous study from our lab has identified two other DEAD-box RNA helicases, RhlB and RhlE, …

Design, Synthesis And Analysis Of Potential Photo-Activatable Cathepsin K Inhibitors, Khalin Evania Nisbett

Wayne State University Theses

Abstract

DESIGN, SYNTHESIS AND ANALYSIS OF POTENTIAL PHOTO-ACTIVATABLE CATHEPSIN K INHIBITORS

by

KHALIN NISBETT

May 2017

Advisor: Dr. Jeremy Kodanko

Major: Chemistry

Degree: Master of Science

Tightly regulated cysteine CA proteases play a major role in maintaining the homeostasis within cells. Subsequently, when these proteases are dysregulated and mislocalized they disrupt healthy cell dynamics and contribute to many life-threatening pathologies such arteriosclerosis, osteoporosis and cancer. As such many pharmaceutical companies and research teams are highly interested in these proteases as targets. One emergent strategy is the spatiotemporal control of biological processes. In relation to this, a series of spatiotemporally controlled …

Characterizing Mineral Crystallographic Preferred Orientations (Cpos) Along The Eastern North American Margin In The Southern Appalachians: Implications For Middle And Lower Crustal Seismic Anisotropy, Laura Robles

Wayne State University Theses

The Southern Appalachians Mountains have a complex and distinctive tectonic history, characterized by structural deformation extending from late Precambrian throughout the Paleozoic. The Southern Appalachians, known as the Blue Ridge of western North Carolina and eastern Tennessee, is mainly composed of schists and gneisses, granitic and ultramafic intrusives, as well as a sedimentary portion. The main tool for understanding the composition and structure of the deeper parts of the crust that cannot be directly sampled by seismology. Seismic anisotropy is the directional dependence of seismic velocity, and it is an invaluable tool for interpreting the development of sub-surficial dynamics. Seismic …

Adaptive Block And Batch Sizing Forbatched Stream Processing System, Quan Zhang

Wayne State University Theses

The need for real-time and large-scale data processing has led to the development of frameworks for distributed stream processing in clouds. To provide fast, scalable, and fault tolerant stream processing, recent Distributed Stream Processing Systems (DSPS) have proposed to treat streaming workloads as a series of batch jobs, instead of a series of records. Batch-based stream processing systems could process data at high rate, however, it also leads to large end-to-end latency. In this thesis we concentrate on minimizing the end-to-end latency of batched streaming system by leveraging adaptive batch size and execution parallelism tuning. We propose, DyBBS, a heuristic …

Multi-Way Factorization Machine For Sentiment Analysis, Jingwei Zhang

Wayne State University Theses

Sentiment analysis is a process of learning the relationship between sentiment label

and text. The research value of sentiment analysis is two-fold: first, it has a wide range

of applications in many sectors and industries, e.g., the industry has flourished due to the

proliferation of commercial applications such as using sentiment analysis as an integrated

part of customer experience strategy. Second, it offers an array of new challenging problems

for research community such as word feature embedding and machine learning. Albeit earlier

methods such as Naïve Bayes (NB), Random Forest (RF), k-Nearest-Neighbours (kNN),

Support Vector Machine (SVM) and more recent …

On Detecting Faces And Classifying Facial Races With Partial Occlusions And Pose Variations, Tarik Alafif Alafif

Wayne State University Theses

In this dissertation, we present our contributions in face detection and facial race classification.

Face detection in unconstrained images is a traditional problem in computer vision community. Challenges still remain. In particular, the detection of partially occluded faces with pose variations has not been well addressed. In the first part of this dissertation, our contributions are three-fold. First, we introduce our four image datasets consisting of large-scale labeled face dataset, noisy large-scale labeled non-face dataset, CrowdFaces dataset, and CrowdNonFaces dataset intended to be used for face detection training. Second, we improve Viola-Jones (VJ) face detection results by first training a …

Integrative Pathway Analysis Pipeline For Mirna And Mrna Data, Diana Mabel Diaz Herrera

Wayne State University Theses

The identification of pathways that are involved in a particular phenotype helps us understand the underlying biological processes. Traditional pathway analysis techniques aim to infer the impact on individual pathways using only mRNA levels. However, recent studies showed that gene expression alone is unable to capture the whole picture of biological phenomena. At the same time, MicroRNAs (miRNAs) are newly discovered gene regulators that have shown to play an important role in diagnosis, and prognosis for different types of diseases. Current pathway analysis techniques do not take miRNAs into consideration. In this project, we investigate the effect of integrating miRNA …

Comparison Of The Scavenging Intensity, Remineralization And Residence Time Of 210po And 210pb At Key Interfaces (Biotic, Sediment-Water And Hydrothermal) Along The Geotraces East Pacific Zonal Transect, John Niedermiller

Wayne State University Theses

The contrasting geochemical behavior of two long-lived progeny of 222Rn, namely 210Po and 210Pb, provide valuable insights on the extent of recycling of biogenic particulate matter and their preferential removal from the water column. A total of 135 water samples were collected and analyzed from six vertical profiles for 210Po and 210Pb along the US GEOTRACES East Pacific Zonal Transect. This study shows decreasing atmospheric 210Pb flux from the measured 210Pb activity in aerosols indicating an overall decrease from east to west along the cruise transect. Differences in scavenging intensities, inventories, and residence times of 210Po and 210Pb at key …

Crustal Cooling In The Neutron Star Low-Mass X-Ray Binary Ks 1731−260, Rachael Lynn Merritt

Wayne State University Theses

Neutron stars in binary systems can undergo periods of accretion (outburst), where in- falling material heats the crust of the star out of thermal equilibrium with the core. When accretion stops (quiescence), we can directly observe the thermal relaxation of the crust. Crustal cooling of accretion-heated neutron stars provides insight into the stellar interior of neutron stars. The neutron star X-ray transient, KS 1731−260, was in outburst for 12.5 years before returning to quiescence in 2001. Here, we present a 150 ks Chandra observation of KS 1731−260 taken in August 2015, about 14.5 years into quiescence. We find that the …

Seismic Anisotropy As A Function Of Mineralogy And Rock Type In Chester Gneiss Dome, Southeast Vermont, Brittany Watling

Wayne State University Theses

Our knowledge of composition and structure in middle and lower crustal rocks is limited due to restricted accessibility, and thus comes mainly from studies of seismic velocities. But isotropic seismic velocities are not sufficient to distinguish between the many possible rock types in the middle and lower crust. Seismic anisotropy is the directional dependence of seismic velocity, and may provide further constraints on mineralogy and rock type. This study is focused on characterizing the seismic anisotropy of rocks from the Chester dome in southeast Vermont. We have simplified the Chester dome rock types into four groups based on rock type …

Facile Synthesis Of Tertiary Aliphatic Amine– Containing Cyclic Motif Via Neutral Aminyl Radical Cyclization, Heng Chen

Wayne State University Theses

Tertiary amine–containing cyclic motifs are prevalent in natural products and pharmaceutical lead targets. Inspired by these biological active molecules, one focus in our lab is the development of neutral aminyl radical chemistry in the synthesis of tertiary aliphatic amine–containing cyclic compounds. In this project, I investigated the mono- and tandem cyclization of aminyl radical with different types of olefins (terminal olefin, enone, and styrenyl). The effect of the hydrogen atom donor and the limitation of the reaction will be discussed.

Feature Grouping Using Weighted L1 Norm For High-Dimensional Data, Karthik Kumar Padthe`

Wayne State University Theses

Building effective prediction models from high-dimensional data is an important problem in several domains such as in bioinformatics, healthcare analytics and general regression analysis. Extracting feature groups automatically from such data with several correlated features is necessary, in order to use regularizers such as the group lasso which can exploit this deciphered grouping structure to build effective prediction models. Elastic net, fused-lasso and Octagonal Shrinkage Clustering Algorithm for Regression (oscar) are some of the popular feature grouping methods proposed in the literature which recover both sparsity and feature groups from the data. However, their predictive ability is affected adversely when …

Targeted Delivery Of Nrf2 Sirna Using Modular Polymeric Micellar Nanodelivery System For Efficient Target Gene Knockdown In Hepatocellular Carcinoma, Shaimaa Mohamed Ibrahim Yousef

Wayne State University Theses

Tumor selective drug delivery as well as chemotherapy associated multi drug resistance (MDR) pose tremendous hurdles for effective cancer therapy. In this regard, designing multifunctional nanocarriers loaded with drug/gene payloads and engineered with tumor targeting ligands can serve as a modular platform for targeted drug/gene delivery. In this study we undertook the synthesis of a self-assembling block copolymer constructed using poly(styrene-co-maleic anhydride, partial iso-octyl ester) (SMAPIE) and branched polyethylenimine (PEI) as building blocks and evaluated its micelle forming ability, siRNA complexation and siRNA delivery potentials. In addition, we engineered galactosamine decorated nanomicelles using modular “click” chemistry based approaches for evaluating …

Modified Synthesis Of Ligands For Lanthanide Ions, Lauren Elizabeth Hopper

Wayne State University Theses

Ligands of relevance to lanthanide ions have been synthesized and modified with respect to pKa and thermal conditions (both microwave and non-microwave). Microwave-assistance has been used previously to accelerate the rates of reactions of ligand syntheses, primarily as a thermal component in agreement with the Arrhenius equation. The modified syntheses of intermediates for lanthanide complexes presented in this thesis enable rapid synthesis of common intermediates—DO2A-methyl ester, DO2A-ethyl ester, and DO2A t-butyl ester—that previously required days to synthesize. Additionally, this modified synthesis can be performed with or without microwave irradiation, suggesting that the thermal component of the rate acceleration is not …

Novel Design And Synthesis Of Transition Metal Hydroxides And Oxides For Energy Storage Device Applications, Peifeng Li

Wayne State University Theses

Supercapacitors (SCs) and Li-ion batteries (LIBs) are two types of important electrical energy storage devices with high power density and high energy density respectively. However, to satisfy the increasing demand of high-performance energy storage devices, the energy density of SCs and power/energy densities of LIBs have to be further improved. The exploration, research, and development of electrode materials with high-performance for applications in SCs and LIBs are still needed to meet the ever-increasing demand on energy and power densities. Herein, the amorphous Ni-Co-Mo ternary hydroxides nanoflakes for SCs and oxides nanoflakes for LIBs with ultrathin stature, abundant open spaces, and …

Development Of A Dreissena Bioassay To Assess The Toxicity Of Contaminants Across Two Life-History Stages, Caroline Joyce Addis

Wayne State University Theses

Dreissenid mussels (Dreissena polymorpha and D. bugensis) have rapidly become widespread and ubiquitous in North America since their introduction into the Great Lakes in the 1980s. The resulting environmental and economic impacts of their invasion have been extensive, negatively affecting biodiversity and costing millions of dollars in control efforts and damage to power generation and water treatment facilities. Although dreissenids are often associated with negative impacts, they may present a practical tool for toxicology studies. The typically sessile behavior of the benthic adults coupled with the planktonic nature of the veligers allow for a more complete evaluation of water quality …