Hoang to discuss drug-drug interaction research at AMIA

Linh Hoang
Linh Hoang
Jodi Schneider
Jodi Schneider, Associate Professor
Nigel Bosch
Nigel Bosch, Assistant Professor

PhD student Linh Hoang will present her research with Assistant Professor Jodi Schneider and Assistant Professor Nigel Bosch at the AMIA (American Medical Informatics Association) Annual Symposium, which will be held virtually from November 14-18. The symposium showcases the latest innovations from the community of biomedical informatics researchers and practitioners.

Hoang will present the paper, "Automatically Classifying the Evidence Type of Drug-Drug Interaction Research Papers as a Step Toward Computer Supported Evidence Curation," which she coauthored with Schneider, Bosch, Richard D. Boyce and Britney Stottlemyer of the University of Pittsburgh, and Mathias Brochhausen of the University of Florida, Gainesville.

"Drug-drug interactions (DDIs) are often a patient safety concern," said Hoang. "However, knowledge bases that discuss DDIs are known for being incomplete and inconsistent with one another. Drug experts, who develop DDI knowledge, need to search for and review evidence from biomedical literature before synthesizing the evidence into clinically useful recommendations. We believe that computer-supported evidence assessment could help drug experts be more efficient and objective in assessing DDI evidence from the literature."

In their paper, the researchers tested the feasibility of using natural language processing (NLP) and machine learning (ML) to classify clinical DDI papers into the formally defined evidence types present in an existing ontology, DIDEO.

"The results suggested that it is feasible to accurately automate the classification of DDI evidence types, which could be a key component of a computerized decision support to help experts be more objective in assessing DDI evidence," said Hoang. "In addition, using existing knowledge about DDI evidence provided by the ontology was a two-way benefit: the NLP/ML tool design helps identify evidence types that do not exist in the current ontology, and the ontology provides the structure that helps construct our NLP/ML tool development more efficiently."

Hoang's research interests include information management, knowledge discovery, and data analytics. She holds a master's in information systems from the University of Surrey in England and a bachelor's in information technology from the Hanoi University of Science and Technology in Vietnam.