Complex-structured data applications

We design statistical models and computational tools that turn high‑dimensional, complex‑structured data into clinically meaningful insight.

What we do?

1. Create novel, interpretable, and useful features from complex data.
   We generate and derive feature sets that capture signal in digital data, spatial‑omic data, geospatial data, and imaging data, providing surrogate representations of complex data that power downstream discovery and decision‑making.

2. Predict clinical phenotypes.
   We integrate biomarkers derived from complex‑structured data to build statistical and machine learning models to predict clinical characteristics that matter to patients and providers.

3. Explain the biology.
   We interpret model findings to understand their biological relevance and implications, closing the loop from data to mechanism to impact.

Where it matters?

Our methods are developed and evaluated in real clinical and public‑health contexts—including oncology (cancer) and neurodegenerative disease (with a focus on Dementia and Alzheimer’s disease)—to ensure the science translates to better decisions to improve patient and public health.