The team, which includes Health Data Sciences Institute director Georgia Tourassi and ORNL researchers Blair Christian and Shamimul Hasan, developed a knowledge graph that presents information in such a way that tools can be constructed to obtain meaningful information from various reams of unstructured text.
By applying unsupervised machine learning and large-scale graph analytic methodologies, the team was able to find clinical trials with related goals.
The digital tool development was aimed at addressing complex challenges relevant to medical conditions such as cancer and Lyme disease as part of The Opportunity Project (TOP) Health Sprint.
ORNL TOP Project lead Ioana Danciu said: “One of the major obstacles facing cancer trial eligibility is the unstructured nature of the data.
“Artificial intelligence and natural language processing tools refine and advances the process of matching cancer patients to promising clinical trials.”
The SmartClinicalTrials capability of ORNL builds upon an existing collaboration between National Cancer Institute (NCI) and DOE in which researchers of ORNL lead a pilot effort to expand cancer surveillance capabilities for predicting the clinical course and outcomes for different types of cancer.
Danciu further added: “These tools make it possible to store the information and continuously add new information in a manner that allows for computational analysis and knowledge discovery down the road.”
The research was also enabled by ORNL’s AI Initiative that harnesses the laboratory’s suite of computing capabilities and user facilities to accelerate scientific breakthroughs.