Radiomics And Ai: Considerations For Clinical Applications & Precision Medicine

Developing artificial intelligence (AI) schemes to assist the clinician towards enabling precision medicine requires embedded information captured by different data modalities, in an intuitive and generalizable fashion. The research in my group focuses on developing novel computational imaging features (termed “radiomic” features) together with histology or molecular data for disease characterization and treatment response evaluation in vivo. Towards this, we have designed unique tools that can capture biologically relevant and clinically intuitive measurements from routinely acquired imaging (MRI, CT, PET) or digitized images of tissue specimens. In addition to developing approaches to ensure these models generalize tio new unseen data, we have also evaluated their repeatability across imaging parameters and reproducibility across institution- or scanner-specific variations. Problems being addressed by us include: (a) predicting response to treatment to identify optimal therapeutic pathways, as well as (b) evaluating therapeutic response to guide follow-up procedures; in the context of colorectal cancers and digestive diseases.