Data Science Commons^™

Towards Fair Patient-Trial Matching Via Patient-Criterion Level Fairness Constraint, Chia-Yuan Chang, Jiayi Yuan, Sirui Ding, Qiaoyu Tan, Kai Zhang, Xiaoqian Jiang, Xia Hu, Na Zou

Faculty, Staff and Student Publications

Clinical trials are indispensable in developing new treatments, but they face obstacles in patient recruitment and retention, hindering the enrollment of necessary participants. To tackle these challenges, deep learning frameworks have been created to match patients to trials. These frameworks calculate the similarity between patients and clinical trial eligibility criteria, considering the discrepancy between inclusion and exclusion criteria. Recent studies have shown that these frameworks outperform earlier approaches. However, deep learning models may raise fairness issues in patient-trial matching when certain sensitive groups of individuals are underrepresented in clinical trials, leading to incomplete or inaccurate data and potential harm. To …

Large Language Models For Healthcare Data Augmentation: An Example On Patient-Trial Matching, Jiayi Yuan, Ruixiang Tang, Xiaoqian Jiang, Xia Hu

Faculty, Staff and Student Publications

The process of matching patients with suitable clinical trials is essential for advancing medical research and providing optimal care. However, current approaches face challenges such as data standardization, ethical considerations, and a lack of interoperability between Electronic Health Records (EHRs) and clinical trial criteria. In this paper, we explore the potential of large language models (LLMs) to address these challenges by leveraging their advanced natural language generation capabilities to improve compatibility between EHRs and clinical trial descriptions. We propose an innovative privacy-aware data augmentation approach for LLM-based patient-trial matching (LLM-PTM), which balances the benefits of LLMs while ensuring the security …

Local Contrastive Learning For Medical Image Recognition, Syed A Rizvi, Ruixiang Tang, Xiaoqian Jiang, Xiaotian Ma, Xia Hu

Faculty, Staff and Student Publications

The proliferation of Deep Learning (DL)-based methods for radiographic image analysis has created a great demand for expert-labeled radiology data. Recent self-supervised frameworks have alleviated the need for expert labeling by obtaining supervision from associated radiology reports. These frameworks, however, struggle to distinguish the subtle differences between different pathologies in medical images. Additionally, many of them do not provide interpretation between image regions and text, making it difficult for radiologists to assess model predictions. In this work, we propose Local Region Contrastive Learning (LRCLR), a flexible fine-tuning framework that adds layers for significant image region selection as well as cross-modality …

Maximizing Productivity And Quality In Senior Thesis Writing With Artificial Intelligence And Natural Language Processing Driven Tools, Lauren Leadbetter

CMC Senior Theses

This project is a Python program designed to generate a senior thesis on a user-
inputted topic using natural language processing techniques. The program takes in a
topic from the user and then uses OpenAI API to deploy text models for text genera-
tion and evaluation, such as GPT-3 and Davinci-003. The resulting output is in .tex
format and includes a first-draft outline and paper, followed by self-generated assessment, with scoring, revisions, and feedback comments instructing manual revisions.

This submission is a sample using one available model of the project, meant to
demonstrate it’s functionality and limitations. Further model versions …

A Study On Global Reef Deterioration: Exploring Coral Bleaching, Emily Fernandez

CMC Senior Theses

This thesis is a study on coral bleaching and coral mortality, studying the relationship between variables such as depth, exposure, distance to shore, and temperature for percent bleaching. All of the analyses were made using two different data sets, that contain information about bleaching events in specific regions, and dates, and provide information factors such as depth, temperature, and exposure. Models were created for different relationships of variables for eco-regions, recent data, and countries. I attempted to find relationships between variables such as depth, temperature, exposure, and distance to shore, and how they affect coral bleaching. Unfortunately, I did not …

The Basil Technique: Bias Adaptive Statistical Inference Learning Agents For Learning From Human Feedback, Jonathan Indigo Watson

Theses and Dissertations--Computer Science

We introduce a novel approach for learning behaviors using human-provided feedback that is subject to systematic bias. Our method, known as BASIL, models the feedback signal as a combination of a heuristic evaluation of an action's utility and a probabilistically-drawn bias value, characterized by unknown parameters. We present both the general framework for our technique and specific algorithms for biases drawn from a normal distribution. We evaluate our approach across various environments and tasks, comparing it to interactive and non-interactive machine learning methods, including deep learning techniques, using human trainers and a synthetic oracle with feedback distorted to varying degrees. …

Should Academia Thrive For Research Citation In Policy? A Case Study On Five Universities In Illinois., Minhaz Suleman Ibrahim Patel

CURE Proceedings

Academics and policymakers are seen as operating separately, which limits the potential impact of research on society. The influence of university research on policy documents is frequently underestimated, given that cutting-edge research is being conducted at universities. Therefore, it is crucial to unveil the role of academic research in fostering evidence-driven policymaking across various public service domains. In this study, we conducted an in-depth exploratory data analysis and statistical summarization to comprehensively understand the level of academic research present in policy documents. We chose five public universities from the state of Illinois and collected research and policy citation data for …

The Shortfalls Of Vulnerability Indexes For Public Health Decision-Making In The Face Of Emergent Crises: The Case Of Covid-19 Vaccine Uptake In Virginia, Lydia Cleveland Sa, Erika Frydenlund

VMASC Publications

Equitable and effective vaccine uptake is a key issue in addressing COVID-19. To achieve this, we must comprehensively characterize the context-specific socio-behavioral and structural determinants of vaccine uptake. However, to quickly focus public health interventions, state agencies and planners often rely on already existing indexes of "vulnerability." Many such "vulnerability indexes" exist and become benchmarks for targeting interventions in wide ranging scenarios, but they vary considerably in the factors and themes that they cover. Some are even uncritical of the use of the word "vulnerable," which should take on different meanings in different contexts. The objective of this study is …

Fitting Time Series Models To Fisheries Data To Ascertain Age, Kathleen S. Kirch, Norou Diawara, Cynthia M. Jones

OES Faculty Publications

The ability of government agencies to assign accurate ages of fish is important to fisheries management. Accurate ageing allows for most reliable age-based models to be used to support sustainability and maximize economic benefit. Assigning age relies on validating putative annual marks by evaluating accretional material laid down in patterns in fish ear bones, typically by marginal increment analysis. These patterns often take the shape of a sawtooth wave with an abrupt drop in accretion yearly to form an annual band and are typically validated qualitatively. Researchers have shown key interest in modeling marginal increments to verify the marks do, …

Evaluation Of Edison's Data Science Competency Framework Through A Comparative Literature Analysis, Karl R. B. Schmitt, Linda Clark, Katherine M. Kinnaird, Ruth E. H. Wertz, Björn Sandstede

Statistical and Data Sciences: Faculty Publications

During the emergence of Data Science as a distinct discipline, discussions of what exactly constitutes Data Science have been a source of contention, with no clear resolution. These disagreements have been exacerbated by the lack of a clear single disciplinary 'parent.' Many early efforts at defining curricula and courses exist, with the EDISON Project's Data Science Framework (EDISON-DSF) from the European Union being the most complete. The EDISON-DSF includes both a Data Science Body of Knowledge (DS-BoK) and Competency Framework (CF-DS). This paper takes a critical look at how EDISON's CF-DS compares to recent work and other published curricular or …

Deep Learning-Based Technique For The Perception Of The Cervical Cancer, Aya Haraz, Hossam El-Din Moustafa, Abeer Twakol Khaleel, Ahmed H. Eltanboly

Mansoura Engineering Journal

In third-world countries, cervical cancer is the most prevalent and leading cause of death. It is affected by a variety of factors, including smoking, poor nutritional status, immunological inadequacy, and prolonged use of contraception. The Pap smear test, which is intended to prevent cervical cancer, finds preneoplastic changes in cervical epithelial cells. This study framework classified cervical cancer cells from Pap smears into five specified cell types using machine learning-based classification algorithms. The SIPaKMeD database is used in this investigation. This public dataset, which was manually cropped from 966 cluster cell images taken from Pap smear slides, has 4045 isolated …

Blockchain And Puf-Based Secure Key Establishment Protocol For Cross-Domain Digital Twins In Industrial Internet Of Things Architecture, Khalid Mahmood, Salman Shamshad, Muhammad Asad Saleem, Rupak Kharel, Ashok Kumar Das, Sachin Shetty, Joel J. P. C. Rodrigues

VMASC Publications

Introduction:: The Industrial Internet of Things (IIoT) is a technology that connects devices to collect data and conduct in-depth analysis to provide value-added services to industries. The integration of the physical and digital domains is crucial for unlocking the full potential of the IIoT, and digital twins can facilitate this integration by providing a virtual representation of real-world entities.

Objectives:: By combining digital twins with the IIoT, industries can simulate, predict, and control physical behaviors, enabling them to achieve broader value and support industry 4.0 and 5.0. Constituents of cooperative IIoT domains tend to interact and collaborate during their complicated …

Lessons Learned From Interdisciplinary Efforts To Combat Covid-19 Misinformation: Development Of Agile Integrative Methods From Behavioral Science, Data Science, And Implementation Science, Sahiti Myneni, Paula Cuccaro, Sarah Montgomery, Vivek Pakanati, Jinni Tang, Tavleen Singh, Olivia Dominguez, Trevor Cohen, Belinda Reininger, Lara S Savas, Maria E Fernandez

Faculty, Staff and Student Publications

BACKGROUND: Despite increasing awareness about and advances in addressing social media misinformation, the free flow of false COVID-19 information has continued, affecting individuals' preventive behaviors, including masking, testing, and vaccine uptake.

OBJECTIVE: In this paper, we describe our multidisciplinary efforts with a specific focus on methods to (1) gather community needs, (2) develop interventions, and (3) conduct large-scale agile and rapid community assessments to examine and combat COVID-19 misinformation.

METHODS: We used the Intervention Mapping framework to perform community needs assessment and develop theory-informed interventions. To supplement these rapid and responsive efforts through large-scale online social listening, we developed a …

A Query Engine For Self-Controlled Case Series, With An Application To Covid-19 Ehr Data, Xiaojin Li, Yan Huang, Licong Cui, Guo-Qiang Zhang

Faculty, Staff and Student Publications

Self-controlled case series (SCCS) is a statistical method in epidemiological study design that uses individuals as their own controls, with comparisons made within the same individuals at different time points of observation. SCCS has been applied in settings where it is difficult to identify comparison or control groups. To provide computational support for SCCS, we introduce a query engine called Self-Controlled Case Query (SCCQ) and use it to extract cohorts of self-controlled case series from a large-scale COVID-19 Electronic Health Records (EHR) dataset. Visual summary of the queried population through the R-Shiny visualization framework offers SCCQ's query result dashboard to …

Contextual Variation Of Clinical Notes Induced By Ehr Migration, Kurt Miller, Sungrim Moon, Sunyang Fu, Hongfang Liu

Faculty, Staff and Student Publications

The structure and semantics of clinical notes vary considerably across different Electronic Health Record (EHR) systems, sites, and institutions. Such heterogeneity hampers the portability of natural language processing (NLP) models in extracting information from the text for clinical research or practice. In this study, we evaluate the contextual variation of clinical notes by measuring the semantic and syntactic similarity of the notes of two sets of physicians comprising four medical specialties across EHR migrations at two Mayo Clinic sites. We find significant semantic and syntactic variation imposed by the context of the EHR system and between medical specialties whereas only …

Digital Solutions Observed In Clinical Trials: A Formative Feasibility Scoping Review, Taylor M Harrison, Sungrim Moon, Liwei Wang, Sunyang Fu, Hongfang Liu

Faculty, Staff and Student Publications

Growing digital access accelerates digital transformation of clinical trials where digital solutions (DSs) are increasingly and widely leveraged for improving trial efficiency, effectiveness, and accessibility. Many factors impact DS success including technology barriers, privacy concerns, or user engagement activities. It is unclear how those factors are considered or reported in the literature. Here, we perform a formative feasibility scoping review to identify gaps impacting DS quality and reproducibility in trials. Articles containing digital terms published in English from 2009 to 2022 were collected (n=4,167). 130 articles published between 2016 and 2022 were randomly selected for full-text review. Eligible articles (n=100) …

Gene Expression In Mice With Endothelium-Specific Telomerase Knockout, Zhanguo Gao, Yongmei Yu, Yulin Dai, Zhongming Zhao, Kristin Eckel-Mahan, Mikhail G Kolonin

Faculty, Staff and Student Publications

No abstract provided.

Multidrug Resistance In The Standardized Treatment Of Colon Cancer Harboring A Rare Fibrosarcoma B-Type (Braf) Pn581i Mutation: A Case Report, Xiaoyan Wang, Chenyi Zhao, Yang Gong, Ying Wang, Feng Guo

Faculty, Staff and Student Publications

BRAF non-V600 mutations are a distinct molecular subset of colorectal cancer (CRC) that has little to no clinical similarity to the BRAF V600 mutations. It is generally considered that the BRAF non-V600 mutations correlate with better survival of CRC patients. In this report, we present an unusual case of that a midlife female patient who was initially diagnosed with stage IIIC colon cancer, and multiple metastases were found 25 months after radical surgery. Next-generation sequencing (NGS) revealed the BRAF p.N581I (c.1742A>T) mutation. She received chemotherapy, targeted therapy, and immunotherapy. However, the disease progressed rapidly with rare metastasis of the …

Revisiting The Type Species Of The Genus Homidia (Collembola, Entomobryidae), Guo-Qiang Zhang, Yu-Xin Zhao, Feng Zhang

Faculty, Staff and Student Publications

Homidiacingula Börner, 1906, the type species of the genus Homidia Börner, 1906, is widespread from India to Southeast Asia, but its detailed morphological characteristics have not yet been described. We examined the morphology of specimens of H.cingula from Indonesia and southwestern China and confirmed their conspecific status by comparing their DNA barcoding sequences. We also compared the morphology of H.cingula with other two closely related species, confirming the valid species status of H.subcingula Denis, 1948. Our study provides new taxonomic and molecular data for the genus Homidia.

Selection And Validation Of Optimal Reference Genes For Rt-Qpcr Analyses In Aphidoletes Aphidimyza Rondani (Diptera: Cecidomyiidae), Xiu-Xian Shen, Guo-Qiang Zhang, Yu-Xin Zhao, Xiao-Xiao Zhu, Xiao-Fei Yu, Mao-Fa Yang, Feng Zhang

Faculty, Staff and Student Publications

Aphidoletes aphidimyza is a predator that is an important biological agent used to control agricultural and forestry aphids. Although many studies have investigated its biological and ecological characteristics, few molecular studies have been reported. The current study was performed to identify suitable reference genes to facilitate future gene expression and function analyses via quantitative reverse transcription PCR. Eight reference genes glyceraldehyde-3-phosphate dehydrogenase (GAPDH), RPS13, RPL8, RPS3, α-Tub, β-actin, RPL32, and elongation factor 1 alpha (EF1-α) were selected. Their expression levels were determined under four different experimental conditions (developmental stages, adult …

Efficient Federated Kinship Relationship Identification, Xinyue Wang, Leonard Dervishi, Wentao Li, Xiaoqian Jiang, Erman Ayday, Jaideep Vaidya

Faculty, Staff and Student Publications

Kinship relationship estimation plays a significant role in today's genome studies. Since genetic data are mostly stored and protected in different silos, retrieving the desirable kinship relationships across federated data warehouses is a non-trivial problem. The ability to identify and connect related individuals is important for both research and clinical applications. In this work, we propose a new privacy-preserving kinship relationship estimation framework: Incremental Update Kinship Identification (INK). The proposed framework includes three key components that allow us to control the balance between privacy and accuracy (of kinship estimation): an incremental process coupled with the use of auxiliary information and …

Split Learning For Distributed Collaborative Training Of Deep Learning Models In Health Informatics, Zhuohang Li, Chao Yan, Xinmeng Zhang, Gharib Gharibi, Zhijun Yin, Xiaoqian Jiang, Bradley A Malin

Faculty, Staff and Student Publications

Deep learning continues to rapidly evolve and is now demonstrating remarkable potential for numerous medical prediction tasks. However, realizing deep learning models that generalize across healthcare organizations is challenging. This is due, in part, to the inherent siloed nature of these organizations and patient privacy requirements. To address this problem, we illustrate how split learning can enable collaborative training of deep learning models across disparate and privately maintained health datasets, while keeping the original records and model parameters private. We introduce a new privacy-preserving distributed learning framework that offers a higher level of privacy compared to conventional federated learning. We …

Integrating Comorbidity Knowledge For Alzheimer's Disease Drug Repurposing Using Multi-Task Graph Neural Network, Ko-Hong Lin, Kang-Lin Hsieh, Xiaoqian Jiang, Yejin Kim

Faculty, Staff and Student Publications

Alzheimer's Disease (AD) is a multifactorial disease that shares common etiologies with its multiple comorbidities, especially vascular diseases. To predict repurposable drugs for AD utilizing the relatively well-investigated comorbidities' knowledge, we proposed a multi-task graph neural network (GNN)-based pipeline that incorporates the corresponding biomedical interactome of these diseases with their genetic markers and effective therapeutics. Our pipeline can accurately capture the interactions and disease classification in the network. Next, we predicted drugs that might interact with the AD module by the node embedding similarity. Our candidates are mostly BBB permeable, and literature evidence showed their potential for treating AD pathologies, …

Federated Learning Based Futuristic Biomedical Big-Data Analysis And Standardization, Afifa Salsabil Fathima, Syed Muzamil Basha, Syed Thouheed Ahmed, Sandeep Kumar Mathivanan, Sukumar Rajendran, Saurav Mallik, Zhongming Zhao

Faculty, Staff and Student Publications

Medical data processing and analytics exert significant influence in furnishing dependable decision support for prospective biomedical applications. Given the sensitive nature of medical data, specialized techniques and frameworks tailored for application-centric processing are imperative. This article presents a conceptualization for the analysis and uniformitarian of datasets through the implementation of Federated Learning (FL). The realm of medical big data stems from diverse origins, necessitating the delineation of data provenance and attribute paradigms to facilitate feature extraction and dependency assessment. The architecture governing the data collection framework is intricately linked to remote data transmission, thereby engendering efficient customization oversight. The operational …

Warehouses In The Inland Empire: Displacing Land And Life, Katherine Gelsey

Pomona Senior Theses

The Inland Empire in Southern California embodies unique spatial and social configurations as a consequence of how settler colonialism has manifested locally in the region since the Spanish Mission Period. This work uses GIS software to estimate patterns of land conversion for residential, agricultural, and warehouse land from 2012 to 2022. Preliminary analysis suggests that thousands of people have been displaced by warehouse expansion over the ten-year period. In the twenty-first century, the Southern California logistics industry continues processes of land dispossession and racialized labor exploitation through displacing agricultural and residential land, exposing disproportionately low-income Black and Latine communities living …

Applications Of Transfer Learning From Malicious To Vulnerable Binaries, Sean Patrick Mcnulty

Graduate Student Theses, Dissertations, & Professional Papers

Malware detection and vulnerability detection are important cybersecurity tasks. Previous research has successfully applied a variety of machine learning methods to both. However, despite their potential synergies, previous research has yet to unite these two tasks. Given the recent success of transfer learning in many domains, such as language modeling and image recognition, this thesis investigated the use of transfer learning to improve vulnerability detection. Specifically, we pre-trained a series of models to detect malicious binaries and used the weights from those models to kickstart the detection of vulnerable binaries. In our study, we also investigated five different data representations …

The Health-Promoting Effects And The Mechanism Of Intermittent Fasting, Simin Liu, Min Zeng, Weixi Wan, Ming Huang, Xiang Li, Zixian Xie, Shang Wang, Yu Cai

Faculty, Staff and Student Publications

Intermittent fasting (IF) is an eating pattern in which individuals go extended periods with little or no energy intake after consuming regular food in intervening periods. IF has several health-promoting effects. It can effectively reduce weight, fasting insulin levels, and blood glucose levels. It can also increase the antitumor activity of medicines and cause improvement in the case of neurological diseases, such as memory deficit, to achieve enhanced metabolic function and prolonged longevity. Additionally, IF activates several biological pathways to induce autophagy, encourages cell renewal, prevents cancer cells from multiplying and spreading, and delays senescence. However, IF has specific adverse …

Text Classification Of Cancer Clinical Trial Eligibility Criteria, Yumeng Yang, Soumya Jayaraj, Ethan Ludmir, Kirk Roberts

Faculty, Staff and Student Publications

Automatic identification of clinical trials for which a patient is eligible is complicated by the fact that trial eligibility are stated in natural language. A potential solution to this problem is to employ text classification methods for common types of eligibility criteria. In this study, we focus on seven common exclusion criteria in cancer trials: prior malignancy, human immunodeficiency virus, hepatitis B, hepatitis C, psychiatric illness, drug/substance abuse, and autoimmune illness. Our dataset consists of 764 phase III cancer trials with these exclusions annotated at the trial level. We experiment with common transformer models as well as a new pre-trained …

A Novel Nih Research Grant Recommender Using Bert, Jie Zhu, Braja Gopal Patra, Hulin Wu, Ashraf Yaseen

Faculty, Staff and Student Publications

Research grants are important for researchers to sustain a good position in academia. There are many grant opportunities available from different funding agencies. However, finding relevant grant announcements is challenging and time-consuming for researchers. To resolve the problem, we proposed a grant announcements recommendation system for the National Institute of Health (NIH) grants using researchers' publications. We formulated the recommendation as a classification problem and proposed a recommender using state-of-the-art deep learning techniques: i.e. Bidirectional Encoder Representations from Transformers (BERT), to capture intrinsic, non-linear relationship between researchers' publications and grants announcements. Internal and external evaluations were conducted to assess the …

Sensitive Data Detection With High-Throughput Machine Learning Models In Electrical Health Records, Kai Zhang, Xiaoqian Jiang

Faculty, Staff and Student Publications

In the era of big data, there is an increasing need for healthcare providers, communities, and researchers to share data and collaborate to improve health outcomes, generate valuable insights, and advance research. The Health Insurance Portability and Accountability Act of 1996 (HIPAA) is a federal law designed to protect sensitive health information by defining regulations for protected health information (PHI). However, it does not provide efficient tools for detecting or removing PHI before data sharing. One of the challenges in this area of research is the heterogeneous nature of PHI fields in data across different parties. This variability makes rule-based …