Data Science Commons^™

Physics-Guided Deep Learning For Solar Wind Modeling At L1 Point, Robert M. Johnson

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Neural networks are adept at finding patterns that are too long and too small for humans to find in data. Usually, this power is used to generate predictions with greater accuracy than most alternative models. However, we can also use this power to understand more about the data we train these networks on. We do this by changing the data that the networks train on and the data they are tested on. This allows us to both control the maximum length of a pattern and to compare data between different groups, in our case, different solar cycles. This thesis is …

Missing Value Imputation For Single Omics And Multi-Omics Data, Meng Song

Dissertations

The integration analyses of multi-omics data have the advantages of extending our understanding of biological system across multiple omics layers, unraveling the functional mechanism of complex disease development, and refining the discovery of novel drug targets. However, multi-omics studies often face challenges such as data heterogeneity, missing values problem, interpretability, and imbalance classes. Among these challenges, the missing values problem is a critical issue for large cohort studies as not all samples will get a complete measurement for all the omics layers. To address the problem of missing values in multi-omics data, I focused on the imputation of completely missing …

Deep Learning For Multi-Structured Javanese Gamelan Note Generator, Arik Kurniawati, Eko Mulyanto Yuniarno, Yoyon Kusnendar Suprapto

Knowledge Engineering and Data Science

Javanese gamelan, a traditional Indonesian musical style, has several song structures called gendhing. Gendhing (songs) are written in conventional notation and require gamelan musicians to recognize patterns in the structure of each song. Usually, previous research on gendhing focuses on artistic and ethnomusicological perspectives, but this study is to explore the correlation between gendhing as traditional music in Indonesia and deep learning technology that replaces the task of gamelan composers. This research proposes CNN-LSTM to generate notation of ricikan struktural instruments as an accompaniment to Javanese gamelan music compositions based on balungan notation, rhythm, song structure, and gatra …

Federated Generalized Linear Mixed Models For Collaborative Genome-Wide Association Studies

Faculty, Staff and Student Publications

Federated association testing is a powerful approach to conduct large-scale association studies where sites share intermediate statistics through a central server. There are, however, several standing challenges. Confounding factors like population stratification should be carefully modeled across sites. In addition, it is crucial to consider disease etiology using flexible models to prevent biases. Privacy protections for participants pose another significant challenge. Here, we propose distributed Mixed Effects Genome-wide Association study (dMEGA), a method that enables federated generalized linear mixed model-based association testing across multiple sites without explicitly sharing genotype and phenotype data. dMEGA employs a reference projection to …

Surface-Doped Zinc Gallate Colloidal Nanoparticles Exhibit Ph-Dependent Radioluminescence With Enhancement In Acidic Media, Amar S Kumbhar, Jianguo Wen, Zhiping Luo

Faculty, Staff and Student Publications

As abnormal acidic pH symbolizes dysfunctions of cells, it is highly desirable to develop pH-sensitive luminescent materials for diagnosing disease and imaging-guided therapy using high-energy radiation. Herein, we explored near-infrared-emitting Cr-doped zinc gallate ZnGa2O4 nanoparticles (NPs) in colloidal solutions with different pH levels under X-ray excitation. Ultrasmall NPs were synthesized via a facile hydrothermal method by controlling the addition of ammonium hydroxide precursor and reaction time, and structural characterization revealed Cr dopants on the surface of NPs. The synthesized NPs exhibited different photoluminescence and radioluminescence mechanisms, confirming the surface distribution of activators. It was observed that the colloidal NPs emit …

The Screening Of The Protective Antigens Of Aeromonas Hydrophila Using The Reverse Vaccinology Approach: Potential Candidates For Subunit Vaccine Development, Xiaoheng Zhao, Hanliang Cheng, Xiangning Chen, Jianhe Xu, Zhujin Ding

Faculty, Staff and Student Publications

The threat of bacterial septicemia caused by Aeromonas hydrophila infection to aquaculture growth can be prevented through vaccination, but differences among A. hydrophila strains may affect the effectiveness of non-conserved subunit vaccines or non-inactivated A. hydrophila vaccines, making the identification and development of conserved antigens crucial. In this study, a bioinformatics analysis of 4268 protein sequences encoded by the A. hydrophila J-1 strain whole genome was performed based on reverse vaccinology. The specific analysis included signal peptide prediction, transmembrane helical structure prediction, subcellular localization prediction, and antigenicity and adhesion evaluation, as well as interspecific and intraspecific homology comparison, thereby screening …

The Impact Framework And Implementation For Accessible In Silico Clinical Phenotyping In The Digital Era, Kurt Miller, Liwei Wang, Kirk E Roberts, Steven D Bedrick, William R Hersh, Hongfang Liu

Faculty, Staff and Student Publications

Clinical phenotyping is often a foundational requirement for obtaining datasets necessary for the development of digital health applications. Traditionally done via manual abstraction, this task is often a bottleneck in development due to time and cost requirements, therefore raising significant interest in accomplishing this task via in-silico means. Nevertheless, current in-silico phenotyping development tends to be focused on a single phenotyping task resulting in a dearth of reusable tools supporting cross-task generalizable in-silico phenotyping. In addition, in-silico phenotyping remains largely inaccessible for a substantial portion of potentially interested users. Here, we highlight the barriers to the usage of in-silico phenotyping …

Non-Invasive Arterial Blood Pressure Measurement And Spo2 Estimation Using Ppg Signal: A Deep Learning Framework, Dulin Wang, Wentao Li, Sean I Savitz, Xiaoqian Jiang, Shayan Shams

Faculty, Staff and Student Publications

BACKGROUND: Monitoring blood pressure and peripheral capillary oxygen saturation plays a crucial role in healthcare management for patients with chronic diseases, especially hypertension and vascular disease. However, current blood pressure measurement methods have intrinsic limitations; for instance, arterial blood pressure is measured by inserting a catheter in the artery causing discomfort and infection.

METHOD: Photoplethysmogram (PPG) signals can be collected via non-invasive devices, and therefore have stimulated researchers' interest in exploring blood pressure estimation using machine learning and PPG signals as a non-invasive alternative. In this paper, we propose a Transformer-based deep learning architecture that utilizes PPG signals to conduct …

Age-Related Changes In Circadian Regulation Of The Human Plasma Lipidome, Shadab A. Rahman, Rose M. Gathungu, Vasant R. Marur, Melissa St. Hilaire, Karine Scheuermaier, Marina Belenky, Jackson S. Struble, Charles A. Czeisler, Steven W. Lockley, Elizabeth B. Klerman, Jeanne F. Duffy, Bruce S. Kristal

Computer and Data Science Faculty Publications

Aging alters the amplitude and phase of centrally regulated circadian rhythms. Here we evaluate whether peripheral circadian rhythmicity in the plasma lipidome is altered by aging through retrospective lipidomics analysis on plasma samples collected in 24 healthy individuals (9 females; mean ± SD age: 40.9 ± 18.2 years) including 12 younger (4 females, 23.5 ± 3.9 years) and 12 middle-aged older, (5 females, 58.3 ± 4.2 years) individuals every 3 h throughout a 27-h constant routine (CR) protocol, which allows separating evoked changes from endogenously generated oscillations in physiology. Cosinor regression shows circadian rhythmicity in 25% of lipids in both …

Topological Data Analysis Of Convolutional Neural Networks Using Depthwise Separable Convolutions, Eliot Courtois

Dissertations

In this dissertation, we present our contribution to a growing body of work combining the fields of Topological Data Analysis (TDA) and machine learning. The object of our analysis is the Convolutional Neural Network, or CNN, a predictive model with a large number of parameters organized using a grid-like geometry. This geometry is engineered to resemble patches of pixels in an image, and thus CNNs are a conventional choice for an image-classifying model.

CNNs belong to a larger class of neural network models, which, starting at a random initialization state, undergo a gradual fitting (or training) process, often a …

Hyperspectral Point Cloud Projection For The Semantic Segmentation Of Multimodal Hyperspectral And Lidar Data With Point Convolution-Based Deep Fusion Neural Networks, Kevin T. Decker, Brett J. Borghetti

Faculty Publications

The fusion of dissimilar data modalities in neural networks presents a significant challenge, particularly in the case of multimodal hyperspectral and lidar data. Hyperspectral data, typically represented as images with potentially hundreds of bands, provide a wealth of spectral information, while lidar data, commonly represented as point clouds with millions of unordered points in 3D space, offer structural information. The complementary nature of these data types presents a unique challenge due to their fundamentally different representations requiring distinct processing methods. In this work, we introduce an alternative hyperspectral data representation in the form of a hyperspectral point cloud (HSPC), which …

The Impact Of Covid-19 On The Financial Performance Of Largest Teaching Hospitals, Tiffany Champagne-Langabeer, James R Langabeer

Faculty, Staff and Student Publications

The COVID-19 pandemic disrupted hospital operations. Anecdotal evidence suggests financial performance likewise suffered, yet little empirical research supports this claim. This study aimed to explore the impact of the pandemic on the financial performance of the most prominent academic hospitals in the United States. Data from the 115 largest major teaching hospitals in the United States were extracted from the American Hospital Directory for three years (2019-2021). We hypothesized that the year and region would moderate the relationship between a hospital's return on assets (financial performance) and specific operational variables. We found evidence through descriptive statistics and multivariate moderated regressions …

Future Trends And Directions For Secure Infrastructure Architecture In The Education Sector: A Systematic Review Of Recent Evidence, Isaac Atta Senior Ampofo, Isaac Atta Junior Ampofo

Journal of Research Initiatives

The most efficient approach to giving large numbers of students’ access to computational resources is through a data center. A contemporary method for building the data center's computer infrastructure is the software-defined model, which enables user tasks to be processed in a reasonable amount of time and at a reasonable cost. The researcher examines potential directions and trends for a secured infrastructure design in this article. Additionally, interoperable, highly reusable modules that can include the newest trends in the education industry are made possible by cloud-based educational software. The Reference Architecture for University Education System Using AWS Services is presented …

The Effect Of Sustainability Information Disclosure On The Cost Of Equity Capital: An Empirical Analysis Based On Gartner Top 50 Supply Chain Rankings, Lingyu Li, Xianrong Zheng, Shuxi Wang

Information Technology & Decision Sciences Faculty Publications

While disclosing financial information has been widely proved to reduce the financing cost of a company, the impact of non-financial information, such as sustainability information, disclosing on the financing cost of the company is still in debate. The goal of this paper is to explore the impact of disclosing sustainability-related information on the cost of equity for firms. The paper first introduces the concept of sustainability information disclosure, and then exhibits its benefit through exploring its impact on reducing a firm’s financing cost. It uses the Gartner supply chain top 50 rankings to construct the experiment environment to test for …

Ideology Prediction From Scarce And Biased Supervision: Learn To Disregard The “What” And Focus On The “How”!, Chen Chen, Dylan Walker, Venkatesh Saligrama

Business Faculty Articles and Research

We propose a novel supervised learning approach for political ideology prediction (PIP) that is capable of predicting out-of-distribution inputs. This problem is motivated by the fact that manual data-labeling is expensive, while self-reported labels are often scarce and exhibit significant selection bias. We propose a novel statistical model that decomposes the document embeddings into a linear superposition of two vectors; a latent neutral context vector independent of ideology, and a latent position vector aligned with ideology. We train an end-to-end model that has intermediate contextual and positional vectors as outputs. At deployment time, our model predicts labels for input documents …

Exploring The Impact Of Students Demographic Attributes On Performance Prediction Through Binary Classification In The Kdp Model, Issah Iddrisu, Peter Appiahene, Obed Appiah, Inusah Fuseini

Knowledge Engineering and Data Science

During the course of this research, binary classification and the Knowledge Discovery Process (KDP) were used. The experimental and analytical capabilities of Rapid Miner's 9.10.010 instructional environment are supported by five different classifiers. Included in the analysis were 2334 entries, 17 characteristics, and one class variable containing the students' average score for the semester. There were twenty experiments carried out. During the studies, 10-fold cross-validation and ratio split validation, together with bootstrap sampling, were used. It was determined whether or not to use the Random Forest (RF), Rule Induction (RI), Naive Bayes (NB), Logistic Regression (LR), or Deep Learning (DL) …

Characterization And Estimation Of Musculoskeletal Pain Using Machine Learning, Boluwatife Faremi

Master's Theses

Traditional scales utilized for recording pain are known to be highly subjective and biased due to inaccuracies in recollecting actual pain intensities. As a result, machine learning (ML) models that are trained using these scores as ground truth are reported to have low performance for objective pain classification because of the huge disparity between what was felt in moments of pain and the scores recorded afterward.

In the present study, two devices were designed for gathering real-time, continuous in-session subjective pain scores and the recording of the autonomic nervous system (ANS) altered endodermal (EDA) activity. 24 participants were recruited to …

Fast Inference Of Genetic Recombination Rates In Biobank Scale Data

Faculty, Staff and Student Publications

Although rates of recombination events across the genome (genetic maps) are fundamental to genetic research, the majority of current studies only use one standard map. There is evidence suggesting population differences in genetic maps, and thus estimating population-specific maps, are of interest. Although the recent availability of biobank-scale data offers such opportunities, current methods are not efficient at leveraging very large sample sizes. The most accurate methods are still linkage disequilibrium (LD)-based methods that are only tractable for a few hundred samples. In this work, we propose a fast and memory-efficient method for estimating genetic maps from population genotyping data. …

Weakly Supervised Spatial Relation Extraction From Radiology Reports

Faculty, Staff and Student Publications

OBJECTIVE: Weak supervision holds significant promise to improve clinical natural language processing by leveraging domain resources and expertise instead of large manually annotated datasets alone. Here, our objective is to evaluate a weak supervision approach to extract spatial information from radiology reports.

MATERIALS AND METHODS: Our weak supervision approach is based on data programming that uses rules (or labeling functions) relying on domain-specific dictionaries and radiology language characteristics to generate weak labels. The labels correspond to different spatial relations that are critical to understanding radiology reports. These weak labels are then used to fine-tune a pretrained Bidirectional Encoder Representations from …

Multi-Task Learning With Dynamic Re-Weighting To Achieve Fairness In Healthcare Predictive Modeling, Xiaoqian Jiang, Kai Zhang

Faculty, Staff and Student Publications

The emphasis on fairness in predictive healthcare modeling has increased in popularity as an approach for overcoming biases in automated decision-making systems. The aim is to guarantee that sensitive characteristics like gender, race, and ethnicity do not influence prediction outputs. Numerous algorithmic strategies have been proposed to reduce bias in prediction results, mitigate prejudice toward minority groups and promote prediction fairness. The goal of these strategies is to ensure that model prediction performance does not exhibit significant disparity among sensitive groups. In this study, we propose a novel fairness-achieving scheme based on multitask learning, which fundamentally differs from conventional fairness-achieving …