Team 11: Natural language processing tool for clinical trial inclusion criteria

In order to match cancer patients with clinical trials that match their diagnosis and molecular features of their tumor, physicians need to know what are the inclusion and exclusion criteria for the available clinical trials.  All clinical trials that provide a drug therapy are required to register their clinical trial at However, often time the exclusion and inclusion criteria is embedded into free form text fields.  The goal of this project is use natural language processing to extract the criteria necessary to match patients with clinical trials.  These characteristics of patient disease can be stored into a database that would include information including: required gender, tumor type, tissue type, gene mutational status, such as BRAF wildtype (WT) or mutated in BRCA2 and molecular markers, such Her2, ER, PR or PSA.

Team Lead: Jeffrey Gagan, Pathology,