Published · Scientific Reports
ISPAT
Informed spatially aware patterns for multiplexed immunofluorescence data.
Research Associate · Penn Medicine
Bayesian methods for complex biomedical data and clinical trials.
I develop Bayesian statistical methods, machine-learning models, and bioinformatic pipelines for clinical trials, biomedical phenotyping, molecular cohorts, and high-dimensional spatial data.
Current focus
Bayesian trial design & biomedical phenotyping
What I do
Methodology grounded in consequential biomedical questions.
I am a Research Associate in the Department of Biostatistics, Epidemiology and Informatics at Penn Medicine and the University of Pennsylvania. I work with Michael Harhay, PhD, and colleagues in the Center for Clinical Trials Innovation and the Palliative and Advanced Illness Research (PAIR) Center.
My current work spans ARDS, pneumonia, and sepsis phenotyping; molecular cohort studies; and Bayesian early-phase trial design. I bring to these problems a foundation in Bayesian variable selection, graphical models, and spatial methods for high-dimensional biomedical data.
Previously, I was a Postdoctoral Research Fellow at the University of Michigan, where I worked with Arvind Rao, PhD. I earned my PhD in Biostatistics from the University of Louisville in 2022 under Jeremy Gaskins, PhD, and the late Riten Mitra, PhD.
Flagship research software
From uploaded spatial data to a reproducible, citation-ready analysis.
The Spatial Methods Workbench turns a growing statistical methods program into a guided analysis platform. Twelve executable paths span 2D and 3D networks, spatial AI and machine learning, high-dimensional image regression, prediction, causal analysis, and tumor shape—each with compatibility checks and a complete provenance bundle.
01Upload & mapCoordinates · regions · outcomes · features
02Compatibility gateOnly valid pipelines become available
03Run analysisExploratory through advanced Bayesian engines
04Export provenanceResults · settings · objects · citations
Funding acknowledgement
Support for the Spatial Methods Workbench.
Research reported in this software was supported by the National Cancer Institute of the National Institutes of Health under Award Number R37CA214955-01A1. The content is solely the responsibility of the author and does not necessarily represent the official views of the National Institutes of Health.
Research themes
Four connected lines of inquiry
01
Graphical models
High-dimensional network estimation with horseshoe and spike-and-slab priors.
02Spatial statistics
Cell–cell interaction networks in 2D and 3D multiplexed tissue imaging.
03Multi-modal integration
Fusing histopathology, transcriptomics, and radiomics into coherent models.
04Causal & functional data
Mediation, sensitivity analysis, and interpretable functional biomarkers.
Selected work
From statistical idea to research tool
Preprint · 3D imaging
ISPAT 3D
Spatially varying conditional volumetric network estimation for three-dimensional tumor imaging.
Read preprintOpen-source methods
Seven research repositories
Reproducible implementations spanning spatial graphs, causal inference, functional depth, and imaging genomics.
Explore GitHubData modalities
One statistical toolkit, diverse biomedical questions.
- Critical-care phenotyping
- Early-phase clinical trials
- Molecular cohorts & spatial omics
- Tissue imaging & radiomics
Let’s connect