The overarching goal of our research program is to identify biomarkers for early detection, diagnosis, prognostication, and prediction of therapy response in cancer. To achieve this goal, we curate large-scale imaging and/or genetic data sets with clinical annotations, and apply established or develop novel statistical and machine learning methods. Our work mainly focuses on two areas:

1.    Radiomic analysis of medical image (PET, CT, MRI, etc).

2.    Genomic and transcriptomic analyses of molecular profiles.

Our ultimate goal is to identify clinically useful prognostic and predictive biomarkers to guide selection of optimal therapy and improve outcomes for cancer patients.