Modeling Imaging Phenotypes to predict Omic profiles in Cancer

The potential for radiomics to support oncology decision-making has grown substantially in recent years, as scanning techniques such a CT, MRI and PET have been found to offer unique information regarding the tumor phenotype and microenvironment that is distinct from that provided by path-omics data such as histopathology, genomics and proteomics. Such imaging and omics data can be correlated with one another to identify associations between them. Imaging-omic based biopsies bear potential to yield better pathological outcomes improving treatment for cancer patients. However, the field lacks a unified software platform wherein imaging and omic features can be brought together to conduct a variety of correlational analyses and build robust and artificially intelligent (AI) models that aid the prediction of omic profiles of tumors from their radiological images or vice-versa. This also includes prediction of patient outcomes for therapies based on their omic and imaging profiles. ImaGene is publicly available as both Graphical User Interface and Command-line interface- that can be utilized by researchers and clinicians globally to conduct imaging-omic studies for a variety of cancer types, and further validate and deploy it in clinic to aid effective monitoring, diagnosis, and treatment of cancer patients.

 

A clinical data scientist addressing gaps between clinical and research omics