Research Experience
Research interests
Throughout my experience as a research assistant and graduate student, I have developed a strong interest in several areas. However, I recognize that there is still so much more to explore!
Exploring the relationships between demographic, socioeconomic, environmental, and behavioral factors and health outcomes including colorectal cancer and maternal/child health
Uncovering health disparities and identifying at-risk populations (such as rural vs. urban disparities)
Integrating multimodal data in precision health studies to challenge the ‘one size fits all’ approach
Utilizing big data, such as electronic health records and claims databases
Implementing spatial modeling and spatial machine learning with applications in multi-dimensional health determinants modeling
Pharmacovigilance
Undergraduate Research Assistant 2020-2023
Arkansas Bioscience Institute ✭ Arkansas State University ✭ Dr. Sudeepa Bhattacharyya Lab
Description:
Collaborated on various research projects in a dry lab setting
Worked extensively with diverse human data sources including medical records, electronic health records, survey data, and insurance claims databases
Proficiently analyzed data using a variety of tools, with a primary focus on R and Python. Additionally, utilized SAS and SQL as needed
Conducted statistical modeling and generated detailed reports to communicate findings effectively
Graduate Research Assistant 2023-Current
Arkansas Bioscience Institute ✭ Arkansas State University ✭ Dr. Sudeepa Bhattacharyya Lab ✭ Funded by NSF EPSCoR DART
Description:
Continued research work in a dry lab environment, specializing in data analysis and interpretation
Managed and analyzed complex human data, including medical records, electronic health records, survey data, insurance claims datasets, and spatial data
Proficiently utilized programming languages such as R and Python, alongside SAS, SQL, and ArcGIS Pro for comprehensive data analysis and manipulation
Frequently utilized statistical and machine learning techniques such as regression, random forest, neural networks, and spatial modeling methods to derive meaningful insights from data.
Assumed a mentoring role, guiding and supporting new students in the lab, particularly those without prior statistical or programming experience
Summer Research Participant Summer 2024
U.S. Food and Drug Administration (FDA) ✭ National Center for Toxicological Research (NCTR)
Division of Bioinformatics and Biostatistics ✭ Dr. Wen Zou Lab
Project: Identification of Prescription Opioid-Associated Cardiovascular Adverse Events Through Comprehensive Analysis of FAERS Data
Description:
Mined the FDA Adverse Event Reporting System (FAERS) database to extract drug and adverse event reports.
Utilized the FDA Adverse Event Reporting System (FAERS) to uncover and compare cardiovascular risk profiles for FDA approved prescription opioids using data mining techniques as well as network analysis and hierarchical clustering.
Presented findings to a large, multidisciplinary audience on two occasions (see presentations tab)
Skills
Programming in Python, R, SAS
Experience working with these techniques:
📍
📍
Experience working with these data sources:
📍
📍
📍
Presenting results to a diverse audience
Regression (predictive, classification, and feature selection/regularization)
Survival models including Cox, Kaplan-Meier, and Fine-Gray competing risk models
Insurance claims databases
Birth/death certificates
Electronic health records (EHR)
📍
📍
📍
📍
📍
Predictive and classification modeling
Random Forest (and SHAP models)
Spatial statistics and modeling, including Moran’s I, geographically weighted regression (GWR), and geographically weighted random forest (GW-RF)
Cancer surveillance data including cancer registries and SEER
Spontaneous adverse event and drug error reporting databases, such as the FDA Adverse Event Reporting System (FAERS)
NIH All Of Us database
📍
📍
📍
📍
Working in an interdisciplinary team
Spatial modeling and map making
Traditional statistical techniques such as univariate and bivariate tests, descriptive analysis, multivariate analysis including PCA, …
Data mining and safety signal detection
Online databases like the SDOH database from AHRQ (https://www.ahrq.gov/sdoh/data-analytics/sdoh-data.html) and the EPA’s EJScreen tool (https://www.epa.gov/ejscreen)
Spatial data including POI, polygon, and raster
Select projects
Masters Thesis
A Spatial Machine Learning Framework for Analyzing CRC Incidence and Mortality Across US Counties
Uncover relationships between multidimensional factors and CRC across US counties. This work is ongoing.
Primary skills:
Programming in Python and R, handling and cleaning large datasets, high performance computing (HPC), creating maps and other spatial visualizations, quantifying spatial dependence, feature selection and engineering, global and spatial machine learning modeling, spatial hyperparameter tuning, …
Undergraduate Thesis
Individual Level Social Determinants of Health and Multimorbidity Severity Prediction
In this study, we compare the performance of individual level vs. neighborhood-aggregated level SDOH in disease modeling.
Primary skills used:
Programming in R, exploratory analysis, feature selection and engineering, regression modeling, hyperparameter optimization, model performance comparison, network analysis of categorical variables, clustering, …
Internship Project
Identification of Prescription Opioid-Associated Cardiovascular Adverse Events Through Comprehensive Analysis of FAERS Data
In this study, we mined the FDA Adverse Event Reporting System (FAERS) database to extract drug and adverse event reports. Using this data, we aimed to analyze cardiovascular risk profiles for prescription opioids and uncover associations between opioids in the context of cardiovascular events.
Primary skills used:
Programming in Python and R, data mining, signal detection via disproportionality analysis, network and hierarchical clustering analysis, …