Treasures @ UT Dallas
Not a member yet
7697 research outputs found
Sort by
Electrochemical impedance spectroscopy wearable systems for reporting biomarker modulation in sweat
No full textThe commercial wearable device market today majorly consists of activity trackers and
smartwatches: that enable the monitoring of user states such as walking, sleeping, and exercising
using sensors relying on physically measurable quantities. These devices are the ones that make a
huge impact on the lives of people suffering from chronic illnesses and their quality of life.
Integrating a sweat-based electrochemical biosensor with a wearable device opens new avenues in
health management and decision support systems for healthcare providers as they can provide a
physiologically relevant and clinically acceptable output. Integrating a glucose-sensing sweat
biosensor adds more value in the lives of diabetics, who require support in terms of balancing
quality of life using good diet and exercise routines. This work is a methodology of understanding
the aspects of making such a wearable platform, starting from understanding the needs of the
wearable device user population. The current market technology is thoroughly studied to pick
relevant aspects and an electronic front end is designed within the bounds of good design practice
to enable good accuracy, ease of use, and 1-week battery life. This in turn is utilized to collect
human subject data to get an understanding of the performance of the sensors in varying environmental conditions and user states. Finally, mathematical modeling approaches are used to
build correlations between the outcome to be presented to the user against change in the recorded
data features as per the human subject experimentation
2D Materials: Theoretical Study of Magnetic and Contact Properties
No full textThe integrated circuit is without a doubt one of the most influential inventions in all
of human history. While every technological revolution has had massive impacts across
human societies, modern electronic circuits have increased the rate of change by orders of
magnitude and this process shows no signs of stopping. As society has become accustomed
to the rapid pace of technological development, the expectations for further improvements
are more and more demanding. The silicon transistor was the ideal vehicle for such a rapid
development, as transistors typically become more powerful and less costly to make when
their size is decreased. With the added bonus of being able to cram more transistors into the
same chip, the electronics revolution started, and a snowball effect of increasingly complexity
and performance was unleashed onto the market, leading to the highly interconnected society
we live in today.
However, the benefits of decreasing transistor dimensions cannot last forever. There are
certain extremely fundamental limits, at the nanometer scale, to how far one can go in
making smaller and smaller devices. At some point, transistors begin to suffer from all sorts
of performance-degrading issues, such as short-channel effects, increased leakage, fabrication
difficulties, etc. Even more fundamental issues arise once the device dimensions go down to
only a few nanometers, where quantum effects can seriously degrade traditional silicon-based
transistors.
It is with these scaling limitations in mind that researchers started looking very seriously
at a relatively new class of materials: two-dimensional (2D) materials. 2D materials are
atomically thin materials, consisting of a single layer not bound covalently in the out-of-plane
direction. The 2D nature of these materials is of course in stark contrast with more “normal”
materials, such as silicon or iron, which have covalent bonds in three dimensions. It turns out
that due to the special structure of 2D materials, the physical properties are also extremely
interesting, and worth investigating seriously. At present, various classes of 2D materials
have been found, and many 2D materials have corresponding stacked layered versions with
their own special properties. Add in, for example, the fact that one can dope these materials
of make heterostructures out of several different kinds, then one can start to appreciate the
vast parameter space that can be explored in the search for interesting applications.
In this work, I focus on the applications of 2D materials in logic and memory devices. More
specifically, I discuss the studies done by myself and my collaborators on the magnetic
properties of layered WSe2 and PtSe2, and the calculation of the contact resistance between
a metal and a 2D semiconductor.
In the first part of the thesis, I share our investigation on the nature and stability of magnetic
phases of doped intercalated WSe2 and PtSe2. We showed that, depending on the dopant, the
stable magnetic phase at low temperature can be drastically different in both stability and
type (ferro- or antiferromagnetic). We further showed that the presence of W or Pt vacancies
in the lattice can be used to control the thermodynamic stability of the intercalated structures.
Finally, we investigated the effect of the Pt vacancies on the magnetism in intercalated PtSe2.
We showed that even though the spin polarization around the Pt atoms is very small, the Pt
electronic cloud mediates longer magnetic interactions. Therefore, the presence or absence of
Pt vacancies has a strong impact on the magnetic phases in the intercalated PtSe2.
In the second part of this thesis, transport properties at a metal-2D semiconductor contact
are the main topic. More specifically, I, along with my collaborators, have created a flexible
model that can be used to efficiently simulate metal-2D semiconductor contacts and extract
key parameters, such as the contact resistance. We studied the effects of device parameters,
such as backgate bias, but also simulation parameters, such as the size of the simulation
domain used to solve the Poisson equation. Crucially, we found that the contact resistance
can be underestimated by over an order of magnitude when the Poisson domain is too small.
In the final chapter, I provide an overview of the main achievements of the thesis and discuss
potential avenues for future research
Efficient Machine Learning Algorithms for One- and Two-dimensional Biomedical Signals
No full textData analysis plays a crucial role in healthcare when it comes to diagnosing and detecting illnesses and medical conditions. Thanks to the advancements in data computing and machine
learning, healthcare professionals can leverage this technology to their advantage. There is
a vast amount of biomedical data available, ranging from patient records to medical imaging
and genomic sequencing, as well as clinical trial results. By analyzing this data with the
help of machine learning, healthcare professionals can gain valuable insights that could lead
to more accurate diagnoses, better treatment options, and improved health outcomes for
patients. It’s worth noting that any discovery made through data analysis has the potential
to enhance the quality of life for individuals.
This dissertation presents a successful method for detecting life-threatening ventricular arrhythmias, namely ventricular tachycardia, ventricular fibrillation, and ventricular flutter,
through the use of machine learning algorithms. The method leverages various statistical
features and is capable of detecting these arrhythmias over different ECG signal durations.
Our method can efficiently differentiate ventricular tachycardia/fibrillation/flutter (VTFL)
against normal sinus rhythm (NSR) with an accuracy, recall and positive predictive value of
98.21, 95.57 and 98.61 percents respectively. The discriminatory power of the same algorithm
between VTFL and non-VTFL as characterized by accuracy, recall and positive predictive
value are 98.12, 93.29, and 97.2 percents respectively.
This dissertation also suggests a novel way to identify ST segment depression through a
single ECG lead. The proposed method involves transforming one-dimensional ECG signals
into two-dimensional images to efficiently detect ST segment depression, a key indicator
of myocardial ischemia. The generalized algorithm (subject independent) developed, when
analyzed using a convolutional neural network with 8, 16, and 32 filters in its consecutive
convolutional layers yielded for ECG segments of 40 beats, a sensitivity, specificity and an
accuracy of 91.18, 98.79 and 95.51 percents respectively. A personalized algorithm (subject
dependent) built on the same CNN architecture as the generalized algorithm for detecting
ST segment depression yielded a sensitivity, specificity and precision of 97.04, 99.72 and
98.90 percents respectively.
In addition to one-dimensional ECG signals, this dissertation explores utilizing ultrasound
images in combination with Generative Adversarial Networks (GANs) for data augmentation
to enhance breast cancer detection. We achieved an accuracy of 90.2 percent, which is
higher that any results reported for a single model. By employing a GAN-based approach
to augment data, the detection process can be considerably improved, resulting in greater
performance when identifying the disease from ultrasound breast images. This advancement
is valuable for timely diagnosis and treatment, and has the ability to positively impact
patient outcomes
Essays In Revenue and Capacity Management
No full textThis dissertation consists of three main chapters that focus on developing near-optimal
and easy-to-implement techniques for novel applications such as post-acute care, cloud cost
management, and data monetization.
In Chapter 2, we study an infinite-horizon, stochastic optimization problem with a set of long-
term capacity investment decisions and a sequence of real-time order acceptance/rejection
decisions, which is motivated by an application at a post-acute care provider. The goal is to
maximize the long-run average expected profit per period. The decision to accept or reject
a referral has to be instantaneous; if accepted, the service episode starts immediately. We
develop a simple policy to the optimization problem, derive a worst-case guarantee on its
optimality gap, and demonstrate that this gap vanishes in a meaningful asymptotic regime.
In Chapter 3, we study an infinite-horizon, stochastic optimization problem from the viewpoint
of a firm that employs cloud resources to process incoming orders over time. We derive a
lower bound on the optimal cost by considering a set of decoupled problems, one for each
order. The solutions to these problems are then used to construct a feasible policy for the
original problem and derive an upper bound on that policy’s optimality gap. Importantly,
we show that our policy is asymptotically optimal: when the demand-rates of the orders are
scaled by a factor θ > 0, the policy’s optimality gap scales proportional to 1/√θ.
In Chapter 4, we study a finite-horizon, stochastic optimization problem faced by a data
seller. The data seller monetizes its data set based on the current quality of data. The
quality of data decays over time, and the data seller has to dynamically update its data set
to improve quality and price it accordingly. The goal is to maximize total profit. We derive
structural insights into the optimal policy of the problem and advise the data seller on how
to update and price its data set optimally under different practical scenarios with different
types of data sets
Adventures in Imagination: Child Cognition in British Children’s Literature, 1865-1911
No full textWriting with an interdisciplinary approach, this dissertation analyzes the correlation of literature, psychology, and history as revealed in long nineteenth century classic British children’s literature (1865-1911). This dissertation analyzes Lewis Carroll’s Alice’s Adventures in Wonderland (1865) and Through the Looking Glass (1871), Robert Louis Stevenson’s Treasure Island (1888), Rudyard Kipling’s The Jungle Books (1894), Frances Hodgson Burnett’s The Secret Garden (1911), Edith Nesbit’s Psammead series (Five Children and It (1902), The Phoenix and the Carpet (1904), and The Story of the Amulet (1906)), and J.M. Barrie’s Peter and Wendy (1911). Throughout the chapters, I explore the various ways in which these forerunners of children’s literature depict child cognition in their stories. Broken into three parts, “The New Child and Cognition,” “The Natural Element of Child Cognition,” and “Imagination and Child Cognition,” I argue that authors of classic British children’s literature help form the early conceptions of child psychology by representing facets of child cognition within their stories. Further, I suggest that these stories were historically unique in presenting stories that were not only for children but also about children; with this, a new narrative style with a focus on the
child reader, specifically, became prevalent. While, of course, these authors are not prophetic in their psychological theories, they prove keen observers of children and convey the cognitive complexities of the developing child’s mind. I suggest that a modern reading of these stories not only reveals an incredible depth of understanding child cognition, but it also enables modern readers to examine the confluence of psychology, history, and literature within children’s literature that have implications in provide deeper understanding for the cognitive functions of contemporary children
Is There Globalization Backlash in Asia?
No full textThis project seeks to address the question of whether globalization backlash exists in Asia
and, if so, in which countries. This project aims to shed light on the policies implemented
by governments to counteract the potential negative effects of globalization in the region.
Over the past few years, there has been increasing resistance to globalization. Most studies
on globalization backlash have focused on Western countries, emphasizing concepts such as
embedded liberalism, notably the compensation thesis. In contrast, little attention is paid
to the Asia region. This project explores globalization backlash by examining Asia’s developed and developing countries. By utilizing a mixed-method– quantitative and qualitative
analysis. The results suggest that globalization backlash is observed mainly in developed
countries, and in response, these developed nations adopt policies such as the unemployment
insurance policy to safeguard domestic workers from the challenges posed by global trade
competition and preempt resistance against globalization. The study underscores the highly
conditional nature of embedded liberalism in Asia, offering valuable insight for policymakers
and scholars
Integrated GaN Power Conversion: Topology, Reliability and Implementation
No full textThe explosive growth of power electronics has resulted in high power demand and more stringent
requirements on power conversion systems. When bridging an increasing high input voltage and
a decreasing low output voltage, high step-down ratio power converters demand high power
density, high reliability, and high integration level. Compared to classic silicon power devices,
new arising gallium nitride (GaN) technology presents the superior figure of merits, and it is
regarded as a more promising power device candidate to overcome these challenges. However,
because of the high step-down conversion ratio and unique characteristics of GaN power devices,
GaN-based dc-dc power conversions face new challenges. Thus, a series of integrated GaN power
conversion topologies, schemes and implementations have been explored to address power
density, reliability and integration challenges.
Firstly, a GaN-based double step-down (DSD) power topology is presented for direct 48V/1V
power conversion. In order to realize closed-loop regulation of the DSD power converter, an
adaptive ON- and OFF-time (AO2T) control with elastic ON-time modulation is developed for
both steady state regulation and transient response enhancement. To reinforce the dual-phase operation reliability of the DSD converter, a master-phase mirroring technique enables adaptive
master-slave phase operation, accomplishing automatic phase current balancing.
Secondly, to improve the system reliability of automotive electronics, a low EMI noise high stepdown ratio GaN-based buck converter is designed for direct battery-to-load power conversion. It
employs an anti-aliasing multi-rate spread-spectrum modulation (MR-SSM) technique to suppress
EMI noise and an in-cycle adaptive zero-voltage switching (ZVS) technique to minimize switching
losses. Compared to the classic fixed-rate SSM (FR-SSM), the MR-SSM technique adaptively
spreads EMI spectra in a wider frequency range without aliasing spikes and, thus, reduces peak
EMI noise more effectively. To improve efficiency, an elastic dead-time (tdead) controller facilitates
in-cycle adaptive ZVS despite of a continuous switching frequency variation. For the enhancement
of GaN power devices driving reliability, a pulse-reinforced level shifting technique is proposed
to immune high switching node voltage dv/dt transition.
Thirdly, to enhance the GaN power device reliability of GaN-based power conversion system, an
on-chip self-calibrated full-profile dynamic on-resistance sensing strategy is proposed to monitor
the online healthy state of power devices. It achieves instant dynamic on-resistance sensing beyond
megahertz. Moreover, complicated high-speed current sensing circuits are avoided to reduce
implementation cost, and the random sensing errors are calibrated automatically for high sensing
accuracy. The online state-of-health condition of GaN-based power converter is thus monitored
comprehensively, precisely, and efficiently.
Finally, one monolithic integrated e-mode GaN asymmetrical half-bridge (AHB) power converter
is implemented for direct 48V/1V power conversion, which minimizes non-ideal parasitics,
enhances power conversion reliability, and reduces system complexity significantly. In the AHB converter, an auto-lock auto-break (A2
) level shifting technique is developed to address the
challenges of pull-up performance, device breakdown risk and dv/dt immunity at switching node
voltage. The self-bootstrapped hybrid (SBH) gate driving technique adaptively achieves rail-torail dynamic gate driving in normal operation and robust static gate driving during large transients.
The on-die temperature sensing facilitates hot spot monitoring and thermal management for high
reliability.
In this dissertation, all the proposed GaN dc-dc power converters have been fabricated and tested
to demonstrate the proposed system topologies, control schemes, circuit techniques. The
measurement results successfully validate the effectiveness of the designs. The high switching
frequency, low EMI noise, high reliability, and monolithic integration have been verified to enable
GaN dc-dc power conversions
High-resolution Time-lapse Seismic Velocity Estimation Using Improved Envelope Full Waveform Inversion
No full textAn inverse problem is a general framework used to convert observed measurements into information about a physical object or system. Seismic exploration serves as a typical example
of an inverse problem, utilizing seismic energy to probe beneath the Earth’s surface and often
aiding in the search for valuable deposits of oil, gas, or minerals. Man-made seismic waves
are generated at source stations, radiating outward as three-dimensional waves that travel
downward into the Earth. Upon encountering changes in the Earth’s geological layering,
they produce echoes that travel back up to the surface and are captured by geophones or
hydrophones. These recorded seismic waves can be employed to infer physical properties
such as P-wave velocity (Vp) and subsurface density. Specifically, Full Waveform Inversion
(FWI) is a high-resolution seismic inversion method based on fitting the entire content of
seismic waveforms with an optimization objective function to extract physical parameters of
the medium sampled by seismic waves. Furthermore, 4D seismic data (seismic data acquired
repeatedly at multiple times) can be used to invert the changes in seismic response during
the production phase. Likewise, time-lapse FWI using the entire 4D seismic waveform has
the potential to reveal high-resolution velocity changes related to reservoir production.
However, FWI is a highly nonlinear inverse problem and suffers from a well-known cycle-
skipping issue, and heavily depends on low frequencies, long offset data, a good initial velocity
model, and accurate source wavelet estimation. Envelope inversion (EI) was proposed to help
overcome cycle-skipping issue and recover the long-wavelength background model by fitting
the envelop of entire content of seismic waveforms, rather than the waveforms themselves.
Nevertheless, some issues like instability of the adjoint source and gradient term make it
challenging to use in practice for many seismology applications.
In this study, we firstly examine the EI methods with different envelope function exponent p-
values from an adjoint source analysis perspective, and reveal the adverse effects of the adjoint
source weighting term on the data residuals, which degrade or mute important inversion
gradient contributions, especially from reflected or scattered phases in the seismic data. We
develop the theory and implementation for an improved envelope inversion (IEI) method
by adding an upshift constant (zero-frequency DC bias term) to the seismic waveform data
before calculating the envelope functions, which can help solve the instability issue of the
adjoint source of EI with power value p = 1. We test the performance of all three methods
in detail using the highly realistic SEAM4D model and data, and show the advantages of
our new IEI method compared to conventional EI and FWI methods.
In addition, we implement IEI on marine seismic data and find that the amplitude mismatching between the modeled and observed data is a critical factor that prevents the successful
application of IEI on real data. To match amplitude better, we introduce an amplitude normalization strategy using the amplitude of the head waves as the normalization reference.
We demonstrate the successful application of IEI on a 2D marine seismic data example. We
estimate a reasonable velocity model with IEI and a high-resolution velocity model with IEI
+ FWI using data bandpassed to a maximum frequency content of 24 Hz. In addition, we
apply the traditional EI methods with different power values on the envelope and discuss
their limitations on the real data example. Furthermore, we implement inversions using different initial velocity models to show the ability of IEI to overcome the cycle-skipping issue
and recover low-wavenumber velocity components compared to FWI, including a low-velocity
reservoir zone.
Finally, time-lapse FWI can introduce inversion artifacts, which might mask the true time-
lapse signature within the reservoir zone, especially for the inversion on real data examples.
We implement sequential bootstrap time-lapse FWI on the time-lapse marine seismic data
from North Australia and observe noticeable inversion artifacts. To better understand these
time-lapse FWI artifacts, we implement time-lapse FWI using several seismic data with
different frequency bands and observe the distribution of the potential true time-lapse signature and inversion artifacts. On the inverted Vp change models using data with different
frequency bands, the distribution of inversion artifacts exhibits inconsistencies, while the
inverted Vp change at the reservoir remains consistent. In order to comprehend these observations, we analyze the origin of these time-lapse artifacts from a theoretical perspective
and find that a portion of these artifacts may arise from inversion null space and differences
in data residuals between baseline and monitoring inversions. Building on these insights,
we develop a novel time-lapse FWI method to suppress these inversion artifacts. We use
the energy of the inverted Vp change model utilizing data with a lower dominant frequency
band as a gradient-weighting term for time-lapse FWI using data with a higher dominant
frequency band. The test of the novel time-lapse FWI method on marine data demonstrates
its capability to effectively suppress inversion artifacts
Human Milk Rheology: a Path Toward Understanding the Physiology of Breastfeeding
No full textBreastfeeding is the recommended method of feeding infants and requires coordinated mechanical activities from both the mother and infant to succeed. Previous studies on breastfeeding biomechanics have been limited to infant-applied forces that are easily obtained. The
flow of milk within the human mammary ducts plays a critical role in breastfeeding but lacks
extensive study. The rheological profile of raw human milk provides a medium to facilitate
further study on the mechanics of breastfeeding with greater emphasis on the contribution
of the mammary gland in response to infant-applied forces.
The following work presents a new understanding in infant-applied pressures, an expanded
exploration of milk rheological behavior, and the response of milk to specific suckling patterns. The results of these experimental studies provide greater insight into breastfeeding
mechanics
Cooperative Collision Avoidance for Autonomous Vehicles Using Monte Carlo Tree Search
No full textAutonomous vehicles require an effective cooperative action planning strategy in an emergency situation. Most action planning approaches for autonomous vehicles do not scale well
with the number of vehicles. In this dissertation, we present COCOA (Cooperative Collision
Avoidance), an efficient cooperative action planning algorithm for autonomous vehicles in colliding situations. In COCOA, autonomous vehicles drive together in coalition formations for
information sharing and cooperation. When a coalition member detects a colliding situation
with a misbehaving vehicle, all coalition members explicitly cooperate to find conflict-free
action plans to avoid collisions with the misbehaving vehicle. COCOA employs a hierarchical decision-making approach where action planning is achieved at two levels: at the vehicle
level and at the coalition level. In emergency scenarios involving multiple coalitions, COCOA
employs a sequential and hierarchical decision-making approach. Leaders of the coalitions
in a coalition sequence cooperate to finalize action plans for their coalition members that
are free of inter-coalition conflicts. The COCOA algorithm is validated through extensive
realistic simulations in a multi-agent-based traffic simulation system