Matching Items (262)
Filtering by
- All Subjects: Machine Learning
- Creators: Barrett, The Honors College
- Member of: Theses and Dissertations
Description
Cancer rates in our nearest relatives are largely unknown. Comparison of human cancer rates with other primates should help us to understand the nature of our susceptibilities to cancer. Data from deceased primates was gathered from 3 institutions, the Duke Lemur Center, San Diego Zoo, and Jungle Friends primate sanctuary. This data contained over 400 unique individuals across 45 species with information on cancer incidence and mortality. Cancer incidence ranged from 0-71% and cancer mortality ranged from 0-67%. We used weighted phylogenetic regressions to test for an association between life history variables (specifically body mass and lifespan) and cancer incidence as well as mortality. Cancer incidence did not correlate with both body mass and lifespan (p>.05) however, cancer mortality did (p<.05). However, it is uncertain if the variables can be used as reliable predictors of cancer, because the data come from different organizations. This analysis presents cancer incidence rates and cancer mortality rates in species where it was previously unknown, and in some primate species, is surprisingly high. Microcebus murinus(grey mouse lemur) appear to be particularly vulnerable to cancer, mostly lymphomas. Further studies will be required to determine the causes of these vulnerabilities.
ContributorsWalker, William Charles (Author) / Maley, Carlo (Thesis director) / Boddy, Amy (Committee member) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2017-05
Description
Open source image analytics and data mining software are widely available but can be overly-complicated and non-intuitive for medical physicians and researchers to use. The ASU-Mayo Clinic Imaging Informatics Lab has developed an in-house pipeline to process medical images, extract imaging features, and develop multi-parametric models to assist disease staging and diagnosis. The tools have been extensively used in a number of medical studies including brain tumor, breast cancer, liver cancer, Alzheimer's disease, and migraine. Recognizing the need from users in the medical field for a simplified interface and streamlined functionalities, this project aims to democratize this pipeline so that it is more readily available to health practitioners and third party developers.
ContributorsBaer, Lisa Zhou (Author) / Wu, Teresa (Thesis director) / Wang, Yalin (Committee member) / Computer Science and Engineering Program (Contributor) / W. P. Carey School of Business (Contributor) / Barrett, The Honors College (Contributor)
Created2016-12
Description
Literature on the undocumented population in the United States is rich, and is growing in the area of the 1.5 generation (which refers to undocumented individuals, typically under age 30, who have grown up in the U.S.), but is scant regarding the health of this population, how they alleviate illnesses and what resources they have to do so. While Deferred Action for Childhood Arrivals (DACA) provides temporary benefits to undocumented youth, a DACA health gap persists. Even for those who are awarded DACA, when compared to their citizen counterparts, resources are still unequal. The 1.5 generation faces unique health challenges and even with policy progress, circumstances tied to their documentation status leave them reverting back to limited resources. In this study, ten members of this generation were interviewed. Findings show that they suffer from minor physical health challenges, but significant mental and emotional health challenges without the means to access adequate healthcare comparable to their citizen counterparts.
ContributorsDay, Elinor Gabriela (Author) / Estrada, Emir (Thesis director) / Perez, Marisol (Committee member) / School of Human Evolution and Social Change (Contributor) / Barrett, The Honors College (Contributor)
Created2016-12
Description
Prior research has confirmed that supervised learning is an effective alternative to computationally costly numerical analysis. Motivated by NASA's use of abort scenario matrices to aid in mission operations and planning, this paper applies supervised learning to trajectory optimization in an effort to assess the accuracy of a less time-consuming method of producing the magnitude of delta-v vectors required to abort from various points along a Near Rectilinear Halo Orbit. Although the utility of the study is limited, the accuracy of the delta-v predictions made by a Gaussian regression model is fairly accurate after a relatively swift computation time, paving the way for more concentrated studies of this nature in the future.
ContributorsSmallwood, Sarah Lynn (Author) / Peet, Matthew (Thesis director) / Liu, Huan (Committee member) / Mechanical and Aerospace Engineering Program (Contributor) / School of Earth and Space Exploration (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
Cancer is a disease that has no bias based on race, gender, sexuality, socioeconomic status, or religious beliefs. Millions upon millions of people are affected every day by this disease in many different ways. In order to show support and raise funds for these people to help with treatment costs, housing, and much more American Cancer Society created and event called Relay for Life. Relay for Life is an event that many people may describe as a walk-a-thon fundraiser, but to those who have had a personal experience with cancer they understand that Relay is much more. Relay for Life is more than a fundraiser; it is an event that brings hope, love, and care into a community. Many people across the country show up to a Relay event to hear the success stories of those who are in remission, show support for their family and friends who are still fighting, and simply volunteer in order to further remember those that they lost to cancer.
The impacts that Relay for Life supplies go beyond monetary value and branch into the world of emotional and mental value. The stories that you hear from cancer patients, caregivers, survivors, friends, and family all show the appreciation for this event even in the smallest of communities. Looking at the Relay for Life website you can see the thousands of submissions detailing exactly why that individual participates in this event. You can read stories of sorrow, drive, friendships that have formed, and hope that has sprouted because of Relay for Life. An event such as this that celebrates the fight and works to give the world more birthdays truly empowers its participants to make a difference and make a connection with each other.
In this project, I set out to reveal the importance of Relay for Life that can be seen and heard through everyone who participates across the nation. It is important to take both personal experience and monetary value into account when looking at how Relay has had a positive impact on the lives of those affected by cancer, but when looking at the broad picture it becomes obvious how this event means more than money.
The impacts that Relay for Life supplies go beyond monetary value and branch into the world of emotional and mental value. The stories that you hear from cancer patients, caregivers, survivors, friends, and family all show the appreciation for this event even in the smallest of communities. Looking at the Relay for Life website you can see the thousands of submissions detailing exactly why that individual participates in this event. You can read stories of sorrow, drive, friendships that have formed, and hope that has sprouted because of Relay for Life. An event such as this that celebrates the fight and works to give the world more birthdays truly empowers its participants to make a difference and make a connection with each other.
In this project, I set out to reveal the importance of Relay for Life that can be seen and heard through everyone who participates across the nation. It is important to take both personal experience and monetary value into account when looking at how Relay has had a positive impact on the lives of those affected by cancer, but when looking at the broad picture it becomes obvious how this event means more than money.
ContributorsTrisko, Rebecca Lynn (Author) / Roen, Duane (Thesis director) / Wales, Anna (Committee member) / Division of Teacher Preparation (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
This thesis dives into the world of machine learning by attempting to create an application that will accurately predict whether or not a sneaker will resell at a profit. To begin this study, I first researched different machine learning algorithms to determine which would be best for this project. After ultimately deciding on using an artificial neural network, I then moved on to collecting data, using StockX and Twitter. StockX is a platform where individuals can post and resell shoes, while also providing statistics and analytics about each pair of shoes. I used StockX to retrieve data about the actual shoe, which involved retrieving data for the network feature variables: gender, brand, and retail price. Additionally, I also retrieved the data for the average deadstock price for each shoe, which describes what the mean price of new, unworn shoes are selling for on StockX. This data was used with the retail price data to determine whether or not a shoe has been, on average, selling for a profit. I used Twitter’s API to retrieve links to different shoes on StockX along with retrieving the number of favorites and retweets each of those links had. These metrics were used to account for ‘hype’ of the shoe, with shoes traditionally being more profitable the larger the hype surrounding them. After preprocessing the data, I trained the model using a randomized 80% of the data. On average, the model had about a 65-70% accuracy range when tested with the remaining 20% of the data. Once the model was optimized, I saved it and uploaded it to a web application that took in user input for the five feature variables, tested the datapoint using the model, and outputted the confidence in whether or not the shoe would generate a profit.
From a technical perspective, I used Python for the whole project, while also using HTML/CSS for the front-end of the application. As for key packages, I used Keras, an open source neural network library to build the model; data preprocessing was done using sklearn’s various subpackages. All charts and graphs were done using data visualization libraries matplotlib and seaborn. These charts provided insight as to what the final dataset looked like. They showed how the brand distribution is relatively close to what it should be, while the gender distribution was heavily skewed. Future work on this project would involve expanding the dataset, automating the entirety of the data retrieval process, and finally deploying the project on the cloud for users everywhere to use the application.
From a technical perspective, I used Python for the whole project, while also using HTML/CSS for the front-end of the application. As for key packages, I used Keras, an open source neural network library to build the model; data preprocessing was done using sklearn’s various subpackages. All charts and graphs were done using data visualization libraries matplotlib and seaborn. These charts provided insight as to what the final dataset looked like. They showed how the brand distribution is relatively close to what it should be, while the gender distribution was heavily skewed. Future work on this project would involve expanding the dataset, automating the entirety of the data retrieval process, and finally deploying the project on the cloud for users everywhere to use the application.
ContributorsShah, Shail (Author) / Meuth, Ryan (Thesis director) / Nakamura, Mutsumi (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
The human hairless gene (HR) encodes a 130 kDa transcription factor that is primarily expressed in the brain and skin. In the promoter and 5'-untranslated regions (5'-UTR) of HR, there are three putative consensus p53 responsive elements (p53RE). p53 is a tumor suppressor protein that regulates cell proliferation, apoptosis, and other cell functions. The p53 protein, a known tumor suppressor, acts as a transcription factor and binds to DNA p53REs to activate or repress transcription of the target gene. In general, the p53 binding sequence is 5'-RRRCWWGYYY-3' where W is A or T, and R and Y are purines or pyrimidines, respectively. However, even if the p53 binding sequence does not match the consensus sequence, p53 protein might still be able to bind to the response element. The intent of this investigation was to identify and characterize the p53REs in the promoter and 5'-UTR of HR. If the three p53REs (p53RE1, p53RE2, and p53RE3) are functional, then p53 can bind there and might regulate HR gene expression. The first aim for this thesis was to clone the putative p53REs into a luciferase reporter and to characterize the transcription of these p53REs in glioblastoma (U87 MG) and human embryonic kidney (HEK293) cell lines. Through the transactivation assay, it was discovered that p53REs 2 and 3 were functional in HEK293, but none of the response elements were functional in U87 MG. Since p53 displayed a different regulatory capacity of HR expression in HEK293 and U87 MG cells, the second aim was to verify whether the p53REs are mutated in GBM U87 MG cells by genomic DNA sequencing.
ContributorsMaatough, Anas (Author) / Neisewander, Janet (Thesis director) / Hsieh, Jui-Cheng (Committee member) / Goldstein, Elliott (Committee member) / School of Life Sciences (Contributor) / School of Historical, Philosophical and Religious Studies (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
Laboratory animals represent an invaluable, yet controversial, resource in the field of biomedical research. Animal research has been behind many influential discoveries in the field of emerging therapeutics. They provide the link between the theory of the lab bench and the functional application of medicine to influence human health. The use of animals in research is a consideration which must be heavily weighed, and the implementation must be carried out at a very high standard in order to retain research integrity and responsibility. We are in the process of conducting an experiment using laboratory mice to demonstrate cancer treatment using vaccinia (VACV) mutants as a possible oncolytic therapy for certain strains of melanoma. VACV is a double-stranded DNA poxvirus with a large and easily altered genome. This virus contains many genes dedicated to immune evasion, but has shown sensitivity to cell death by necroptosis in mouse studies (5). We have identified the absence of the kinase RIP3 which is vital in the necroptosis pathway as a potential target for oncolytic therapy using VACV mutants in specific strains of melanoma. Multiple groups of SCID Beige mice were inoculated with different melanoma cell lines and observed for tumor growth. Upon reaching 1 cm3 in volume, tumors were injected with either VACV- Δ83N, VACV- Δ54N, or PBS, and observed for regression. It was hypothesized that melanoma tumors that are RIP3-/- such as the MDA5 cell line will show regression, but melanoma tumors that are RIP3-positive and capable of necroptosis, such as the 2427 cell line, will resist viral replication and continue to proliferate. Our results so far tentatively support this hypothesis, but the data collection is ongoing. Strict and specific protocols with regard to the ethical and responsible use of mice have been implemented and upheld throughout the experiment. Animals are closely monitored, and if their quality of life becomes too poor to justify their continued use in the experiment, they are humanely euthanized, even at the expense of valuable data. The importance of commitment to a high ethical standard is pervasive throughout our work. Animals represent an invaluable contribution to research, and it is important to maintain high standards and transparency with regard to their use. Education and engagement in critical discussions about the use and care of animals in the laboratory contribute to the overall merit and legitimacy of biomedical research in the public and professional eye as a whole, and give legitimacy to the continued use of animals as models to advance science and health.
ContributorsBergamaschi, Julia (Author) / Kibler, Karen (Thesis director) / Jacobs, Bertram (Committee member) / School of Human Evolution and Social Change (Contributor) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
Medical records are increasingly being recorded in the form of electronic health records (EHRs), with a significant amount of patient data recorded as unstructured natural language text. Consequently, being able to extract and utilize clinical data present within these records is an important step in furthering clinical care. One important aspect within these records is the presence of prescription information. Existing techniques for extracting prescription information — which includes medication names, dosages, frequencies, reasons for taking, and mode of administration — from unstructured text have focused on the application of rule- and classifier-based methods. While state-of-the-art systems can be effective in extracting many types of information, they require significant effort to develop hand-crafted rules and conduct effective feature engineering. This paper presents the use of a bidirectional LSTM with CRF tagging model initialized with precomputed word embeddings for extracting prescription information from sentences without requiring significant feature engineering. The experimental results, run on the i2b2 2009 dataset, achieve an F1 macro measure of 0.8562, and scores above 0.9449 on four of the six categories, indicating significant potential for this model.
ContributorsRawal, Samarth Chetan (Author) / Baral, Chitta (Thesis director) / Anwar, Saadat (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
The prevalence of bots, or automated accounts, on social media is a well-known problem. Some of the ways bots harm social media users include, but are not limited to, spreading misinformation, influencing topic discussions, and dispersing harmful links. Bots have affected the field of disaster relief on social media as well. These bots cause problems such as preventing rescuers from determining credible calls for help, spreading fake news and other malicious content, and generating large amounts of content which burdens rescuers attempting to provide aid in the aftermath of disasters. To address these problems, this research seeks to detect bots participating in disaster event related discussions and increase the recall, or number of bots removed from the network, of Twitter bot detection methods. The removal of these bots will also prevent human users from accidentally interacting with these bot accounts and being manipulated by them. To accomplish this goal, an existing bot detection classification algorithm known as BoostOR was employed. BoostOR is an ensemble learning algorithm originally modeled to increase bot detection recall in a dataset and it has the possibility to solve the social media bot dilemma where there may be several different types of bots in the data. BoostOR was first introduced as an adjustment to existing ensemble classifiers to increase recall. However, after testing the BoostOR algorithm on unobserved datasets, results showed that BoostOR does not perform as expected. This study attempts to improve the BoostOR algorithm by comparing it with a baseline classification algorithm, AdaBoost, and then discussing the intentional differences between the two. Additionally, this study presents the main factors which contribute to the shortcomings of the BoostOR algorithm and proposes a solution to improve it. These recommendations should ensure that the BoostOR algorithm can be applied to new and unobserved datasets in the future.
ContributorsDavis, Matthew William (Author) / Liu, Huan (Thesis director) / Nazer, Tahora H. (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / Department of Information Systems (Contributor) / Barrett, The Honors College (Contributor)
Created2018-12