We are at the forefront of utilizing artificial intelligence (AI) and advanced computational biology techniques to unravel intricate biological data, with a particular emphasis on omics and imaging data from cancer patients, especially those with prostate and breast cancers.
Our research is structured around three pivotal and inter-connected themes:
- Mechanistic Prediction of Outcome in Cancer Patients: We incorporate prior biological knowledge into the decision rules of machine learning algorithms to build mechanistic classifiers that are both robust and interpretable.
- Microenvironment Analysis: We leverage single-cell RNA sequencing (scRNA-seq) and spatial omics profiling to elucidate the dynamics of the tumor microenvironment at different disease stages. Our objective is to pinpoint the molecular and spatial factors that influence tumor progression and distant metastasis.
- Deep Learning in Histopathology: We design and implement deep learning models on whole slide images of patient specimens to extract a set of spatio-morphological features, which we utilize to forecast specific clinical and molecular phenotypes.
Our overarching goal, guided by these themes, is to identify the primary mediators of cancer metastatic progression. By harnessing state-of-the-art computational tools, we aim to shed light on the intricate interplay of tumor and metastatic microenvironments specific to prostate and breast cancers.
