Matching Items (232)
Filtering by
- All Subjects: Machine Learning
- Creators: Barrett, The Honors College
- Member of: Theses and Dissertations
- Status: Published
Description
Cellular and molecular biologists often perform cellular assays to obtain a better understanding of how cells work. However, in order to obtain a measurable response by the end of an experiment, the cells must reach an ideal cell confluency. Prior to conducting the cellular assays, range-finding experiments need to be conducted to determine an initial plating density that will result in this ideal confluency, which can be costly. To help alleviate this common issue, a mathematical model was developed that describes the dynamics of the cell population used in these experiments. To develop the model, images of cells from different three-day experiments were analyzed in Photoshop®, giving a measure of cell count and confluency (the percentage of surface area covered by cells). The cell count data were then fitted into an exponential growth model and were correlated to the cell confluency to obtain a relationship between the two. The resulting mathematical model was then evaluated with data from an independent experiment. Overall, the exponential growth model provided a reasonable and robust prediction of the cell confluency, though improvements to the model can be made with a larger dataset. The approach used to develop this model can be adapted to generate similar models of different cell-lines, which will reduce the number of preliminary range-finding experiments. Reducing the number of these preliminary experiments can save valuable time and experimental resources needed to conduct studies using cellular assays.
ContributorsGuerrero, Victor Dominick (Co-author) / Guerrero, Victor (Co-author) / Watanabe, Karen (Thesis director) / Jurutka, Peter (Committee member) / School of Mathematical and Natural Sciences (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Leveraging Machine Learning and Wireless Sensing for Robot Localization - Location Variance Analysis
Description
Object localization is used to determine the location of a device, an important aspect of applications ranging from autonomous driving to augmented reality. Commonly-used localization techniques include global positioning systems (GPS), simultaneous localization and mapping (SLAM), and positional tracking, but all of these methodologies have drawbacks, especially in high traffic indoor or urban environments. Using recent improvements in the field of machine learning, this project proposes a new method of localization using networks with several wireless transceivers and implemented without heavy computational loads or high costs. This project aims to build a proof-of-concept prototype and demonstrate that the proposed technique is feasible and accurate.
Modern communication networks heavily depend upon an estimate of the communication channel, which represents the distortions that a transmitted signal takes as it moves towards a receiver. A channel can become quite complicated due to signal reflections, delays, and other undesirable effects and, as a result, varies significantly with each different location. This localization system seeks to take advantage of this distinctness by feeding channel information into a machine learning algorithm, which will be trained to associate channels with their respective locations. A device in need of localization would then only need to calculate a channel estimate and pose it to this algorithm to obtain its location.
As an additional step, the effect of location noise is investigated in this report. Once the localization system described above demonstrates promising results, the team demonstrates that the system is robust to noise on its location labels. In doing so, the team demonstrates that this system could be implemented in a continued learning environment, in which some user agents report their estimated (noisy) location over a wireless communication network, such that the model can be implemented in an environment without extensive data collection prior to release.
Modern communication networks heavily depend upon an estimate of the communication channel, which represents the distortions that a transmitted signal takes as it moves towards a receiver. A channel can become quite complicated due to signal reflections, delays, and other undesirable effects and, as a result, varies significantly with each different location. This localization system seeks to take advantage of this distinctness by feeding channel information into a machine learning algorithm, which will be trained to associate channels with their respective locations. A device in need of localization would then only need to calculate a channel estimate and pose it to this algorithm to obtain its location.
As an additional step, the effect of location noise is investigated in this report. Once the localization system described above demonstrates promising results, the team demonstrates that the system is robust to noise on its location labels. In doing so, the team demonstrates that this system could be implemented in a continued learning environment, in which some user agents report their estimated (noisy) location over a wireless communication network, such that the model can be implemented in an environment without extensive data collection prior to release.
ContributorsChang, Roger (Co-author) / Kann, Trevor (Co-author) / Alkhateeb, Ahmed (Thesis director) / Bliss, Daniel (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
At present, the vast majority of human subjects with neurological disease are still diagnosed through in-person assessments and qualitative analysis of patient data. In this paper, we propose to use Topological Data Analysis (TDA) together with machine learning tools to automate the process of Parkinson’s disease classification and severity assessment. An automated, stable, and accurate method to evaluate Parkinson’s would be significant in streamlining diagnoses of patients and providing families more time for corrective measures. We propose a methodology which incorporates TDA into analyzing Parkinson’s disease postural shifts data through the representation of persistence images. Studying the topology of a system has proven to be invariant to small changes in data and has been shown to perform well in discrimination tasks. The contributions of the paper are twofold. We propose a method to 1) classify healthy patients from those afflicted by disease and 2) diagnose the severity of disease. We explore the use of the proposed method in an application involving a Parkinson’s disease dataset comprised of healthy-elderly, healthy-young and Parkinson’s disease patients.
ContributorsRahman, Farhan Nadir (Co-author) / Nawar, Afra (Co-author) / Turaga, Pavan (Thesis director) / Krishnamurthi, Narayanan (Committee member) / Electrical Engineering Program (Contributor) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
In this project, the use of deep neural networks for the process of selecting actions to execute within an environment to achieve a goal is explored. Scenarios like this are common in crafting based games such as Terraria or Minecraft. Goals in these environments have recursive sub-goal dependencies which form a dependency tree. An agent operating within these environments have access to low amounts of data about the environment before interacting with it, so it is crucial that this agent is able to effectively utilize a tree of dependencies and its environmental surroundings to make judgements about which sub-goals are most efficient to pursue at any point in time. A successful agent aims to minimizes cost when completing a given goal. A deep neural network in combination with Q-learning techniques was employed to act as the agent in this environment. This agent consistently performed better than agents using alternate models (models that used dependency tree heuristics or human-like approaches to make sub-goal oriented choices), with an average performance advantage of 33.86% (with a standard deviation of 14.69%) over the best alternate agent. This shows that machine learning techniques can be consistently employed to make goal-oriented choices within an environment with recursive sub-goal dependencies and low amounts of pre-known information.
ContributorsKoleber, Derek (Author) / Acuna, Ruben (Thesis director) / Bansal, Ajay (Committee member) / W.P. Carey School of Business (Contributor) / Software Engineering (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
This thesis dives into the world of artificial intelligence by exploring the functionality of a single layer artificial neural network through a simple housing price classification example while simultaneously considering its impact from a data management perspective on both the software and hardware level. To begin this study, the universally accepted model of an artificial neuron is broken down into its key components and then analyzed for functionality by relating back to its biological counterpart. The role of a neuron is then described in the context of a neural network, with equal emphasis placed on how it individually undergoes training and then for an entire network. Using the technique of supervised learning, the neural network is trained with three main factors for housing price classification, including its total number of rooms, bathrooms, and square footage. Once trained with most of the generated data set, it is tested for accuracy by introducing the remainder of the data-set and observing how closely its computed output for each set of inputs compares to the target value. From a programming perspective, the artificial neuron is implemented in C so that it would be more closely tied to the operating system and therefore make the collected profiler data more precise during the program's execution. The program is designed to break down each stage of the neuron's training process into distinct functions. In addition to utilizing more functional code, the struct data type is used as the underlying data structure for this project to not only represent the neuron but for implementing the neuron's training and test data. Once fully trained, the neuron's test results are then graphed to visually depict how well the neuron learned from its sample training set. Finally, the profiler data is analyzed to describe how the program operated from a data management perspective on the software and hardware level.
ContributorsRichards, Nicholas Giovanni (Author) / Miller, Phillip (Thesis director) / Meuth, Ryan (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
Vitellogenin (vg) is a precursor protein of egg yolk in honeybees, but it is also known to have immunological functions. The purpose of this experiment was to determine the effect of vg on the viral load of deformed wing virus (DWV) in worker honey bees (Apis mellifera). I hypothesized that a reduction in vg expression would lead to an increase in the viral load. I collected 180 worker bees and split them into four groups: half the bees were subjected to a vg gene knockdown by injections of double stranded vg RNA, and the rest were injected with green fluorescent protein (gfp) double stranded RNA. Half of each group was thereafter injected with DWV, and half given a sham injection. The rate of mortality in all four groups was higher than expected, leaving only 17 bees total. I dissected these bees' fat bodies and extracted their RNA to test for vg and DWV. PCR results showed that, out of the small group of remaining bees, the levels of vg were not statistically different. Furthermore, both groups of virus-injected bees showed similar viral loads. Because of the high mortality rate bees and the lack of differing levels of vg transcript between experimental and control groups, I could not draw conclusions from these results. The high mortality could be caused by several factors: temperature-induced stress, repeated stress from the two injections, and stress from viral infection. In addition, it is possible that the vg dsRNA batch I used was faulty. This thesis exemplifies that information cannot safely be extracted when loss of sampling units result in a small datasets that do not represent the original sampling population.
ContributorsCrable, Emma Lewis (Author) / Amdam, Gro (Thesis director) / Wang, Ying (Committee member) / Dahan, Romain (Committee member) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2017-12
Description
Triops (Branchiopoda: Notostraca) and Streptocephalus (Branchiopoda: Anostraca) are two crustaceans which cohabitate in ephemeral freshwater pools. They both lay desiccation resistant eggs that disperse passively to new hydrologically isolated environments. The extent of genetic distance among regions and populations is of perennial interest in animals that live in such isolated habitats. Populations in six natural ephemeral pool habitats located in two different regions of the Sonoran Desert and a transition area between the Sonoran and Chihuahuan Deserts were sampled. Sequences from Genbank were used for reference points in the determination of species as well as to further identify regional genetic distance within species. This study estimated the amount of within and between genetic distance of individuals from each region and population through the use of a neutral marker, cytochrome oxidase I (COI). We concluded that, although the method of passive dispersal may differ between the two genera, the differences do not results in different patterns of genetic distances between regions and populations. Furthermore, we only found the putative species, Triops longicaudatus "short", with enough distinct speciation. Although Triops longicaudatus "long" and Triops newberryi may be in the early stages of speciation, this study does not find enough support to conclude that they have separated.
ContributorsMurphy Jr., Patrick Joseph (Author) / Rutowski, Ronald (Thesis director) / Cartwright, Reed (Committee member) / Lessios, Nikos (Committee member) / School of Life Sciences (Contributor) / School of Human Evolution and Social Change (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
The Beauty Within is a ceramics show displaying human body anatomy, which seeks to bridge aspects of my biological sciences major in the School of Life Sciences with aspects of my studio art minor in the Herberger Institute for Design and the Arts. My goal in creating the show was to change the opinion of people on human body organs from unease to admiration by recreating these organs in an artistic light. By stylizing the construction of the pieces and bringing in the contemporary form of art \u2014 makeup art \u2014 I hoped to bring a new light to the pieces and highlight the beauty within the human body. By leaving the pieces partly unfinished I further hoped to draw attention to the natural beauty within the pieces regardless of the makeup that covers them. By holding the show in the human anatomy lab room on campus and having both animal and human organs on display I was able to create that sense of disgust toward the organs in the viewers. The beauty of my created pieces was then directly contrasted with the disgust felt about the real organs by displaying each of my pieces next to a real organ. The reactions of the viewers reflected a change in view from the actual organs to my re-created organs, and therefore the goal of the show was achieved.
ContributorsThomas, Brandon Lee (Author) / Weiser, Kurt (Thesis director) / Chung, Samuel (Committee member) / School of Art (Contributor) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2017-05
Description
This study focused on the connection between the EnvZ/OmpR two-component regulatory system and the iron homeostasis system in Escherichia coli, specifically how a mutant form of EnvZ11/OmpR is able to reduce the expression of fepA::lacZ, a reporter gene fusion in E. coli. FepA is one of several outer membrane siderophore receptors that allow extracellular siderophores bound to iron to enter the cells to power various biological processes. Previous studies have shown that in E. coli cells that expressed a mutant allele of envZ, called envZ11, which led to altered expression of various iron genes including down regulation of fepA::lacZ. The wild type EnvZ/OmpR system is not considered to regulate iron genes, but because these envz11 strains had downregulated fepA::lacZ, this study was undertaken to understand the connection and mechanisms of this downregulation. A large number of Lac+ revertants were obtained from the B32-2483 strain (envz11 and fepA::lacZ) and 7 Lac+ revertants that had reversion mutations not directly correcting the envZ11 allele were further characterized. With P1 phage transduction genetic mapping that involved moving a kanamycin resistance marker linked to fepA::lacZ, two Lac+ revertants were found to have their reversion mutations in the fepA promoter region, while the other five revertants had their mutations mapping outside the fepA region. These two revertants underwent DNA sequencing and found to carry two different single base pair mutations in two different locations of the fepA promoter region. Each one is in the Fur repressor binding region, but one also may have affected the Shine-Dalgarno region involved in translation initiation. All 7 reveratants underwent beta-galactosidase assays to measure fepA::lacZ expression. The two revertants that had mutations in the fepA promoter region had significantly increased fepA activity, with the revertant with the Shine-Dalgarno mutation having the most elevated fepA expression. The other 5 revertants that did not map in the fepA region had fepA expression elevated to the same level as that found in the wild type EnvZ/OmpR background. The data suggest that the negative effect of envZ11 can be overcome by multiple mechanisms, including directly correcting the envZ11 allele or changing the fepA promoter region.
ContributorsKalinkin, Victor Arkady (Co-author) / Misra, Rajeev (Co-author, Thesis director) / Mason, Hugh (Committee member) / Foy, Joseph (Committee member) / Biomedical Informatics Program (Contributor) / School of Life Sciences (Contributor) / W. P. Carey School of Business (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
Honey bees (Apis mellifera) are responsible for pollinating nearly 80\% of all pollinated plants, meaning humans depend on honey bees to pollinate many staple crops. The success or failure of a colony is vital to global food production. There are various complex factors that can contribute to a colony's failure, including pesticides. Neonicotoids are a popular pesticide that have been used in recent times. In this study we concern ourselves with pesticides and its impact on honey bee colonies. Previous investigations that we draw significant inspiration from include Khoury et Al's \emph{A Quantitative Model of Honey Bee Colony Population Dynamics}, Henry et Al's \emph{A Common Pesticide Decreases Foraging Success and Survival in Honey Bees}, and Brown's \emph{ Mathematical Models of Honey Bee Populations: Rapid Population Decline}. In this project we extend a mathematical model to investigate the impact of pesticides on a honey bee colony, with birth rates and death rates being dependent on pesticides, and we see how these death rates influence the growth of a colony. Our studies have found an equilibrium point that depends on pesticides. Trace amounts of pesticide are detrimental as they not only affect death rates, but birth rates as well.
ContributorsSalinas, Armando (Author) / Vaz, Paul (Thesis director) / Jones, Donald (Committee member) / School of Mathematical and Statistical Sciences (Contributor) / School of International Letters and Cultures (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05