College of Medicine
Brunson Lab
Led by Jason Brunson · Medicine (Pulmonary)
What we study
The Brunson Lab applies topology and geometric data analysis to medical and statistical problems. Recent student projects use persistent homology on lung CT scans with machine learning, and improve open-source R packages for topological and geometric data analysis.
What working here is like
- Computational
- No organisms (data only)
- Coding required
- Projects
- 3
- Papers
- 5
Current projects
Project ripserr: Enhancing R for Efficient Topological Data Analysis
The lab improves an R package that computes shapes in data using a fast C++ library. The student will update the code, add new ways to input data, and test the speed and compatibility for future projects like analyzing lung scans.
From the Spring 2025 UF Undergraduate Research Symposium · abstract book, p. 601
Tidy Geometric Data Analysis in R: Extensions and Applications
The lab extends {ordr}, a published R package that brings geometric data analysis tools like PCA and factor analysis together. The work checks it against other implementations and adds a guide that helps non-statisticians use these methods.
From the Spring 2025 UF Undergraduate Research Symposium · abstract book, p. 251
Show 1 more projectsShow fewer
Topological Feature Extraction for Machine Learning Classification and Texture Analysis of CT Scans: A Preliminary Study Toward an Application in Pulmonary Sarc
The team uses a math technique called persistent homology to pull shape features from lung CT scans and trains a computer model to tell different cancer types apart. The student processed the images, computed the features, and ran the classification tests.
From the Spring 2025 UF Undergraduate Research Symposium · abstract book, p. 158
Summaries are written by AI in plain language from public symposium abstracts and may be imperfect.
Recent papers, explained
- Multi-site cytokine levels are predictive of primary graft dysfunction following lung transplantation
2026 · Transplant International
Identifies blood markers that predict lung transplant complications.
- Identifying predictors of treatment response in Hoarding Disorder using machine learning
2026 · Behaviour Research and Therapy
Show 3 more papersShow fewer
- Use of Machine Learning and 71-Plex Immune Mediator Analysis to Identify Nasal Mucus Biomarkers Associated With Olfactory Loss in Patients with CRSwNP
2026 · American Journal of Rhinology and Allergy
Uses machine learning to find nasal mucus clues for smell loss.
- Understanding Racial Disparities in Lung Transplant Outcomes: A Comparative Study of White and Hispanic Patients
2025 · Clinical Transplantation
Compares lung transplant survival between Hispanic/Latino and White patients using US transplant data.
- Exploring the Impact of Biotic and Abiotic Surfaces on Protein Binding Modulation and Bacteria Attachment: Integrating Biological and Mathematical Approaches
2025 · ACS Nano
Tests how proteins and bacteria stick to dental and titanium surfaces using saliva and high-throughput methods.
Plain-language summaries are AI-written and reviewed by us; check the paper for details.
How to reach out
Jason Brunson hasn't set up this page yet. You can also look them up in the UF directory (search “Jason Brunson”).