Advancing clinically meaningful AI algorithms to improve oncologic outcomes

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

The goals are to: 1) generate a "toolkit" of algorithms that directly link RT biology tumor response/failure to clinical areas of need to improve patient outcomes 2) define and create optimal methodology linking RT dose/volumes to AI to outcomes and 3) identify surrogates for response/failure as pertaining to radiation oncology ie. biomarkers. Current work involves building connections and collaboration with AI resources, using existing data to pilot projects to generate a gap analysis and build data framework and create a functional query ready radiation oncology study platform in the NIH AI ecosystem to allow for streamlining of analysis. Algorithms are being developed at the clinical, imaging and proteomic level for future aggregation. Future protocol will be aimed at validating promising algorithms.