Attention-gated graph convolutions for extracting drug interaction information from drug labels

Tung Tran, Ramakanth Kavuluru, Halil Kilicoglu

Research output: Contribution to journalArticlepeer-review


Preventable adverse events as a result of medical errors present a growing concern in the healthcare system. As drug-drug interactions (DDIs) may lead to preventable adverse events, being able to extract DDIs from drug labels into a machine-processable form is an important step toward effective dissemination of drug safety information. Herein, we tackle the problem of jointly extracting mentions of drugs and their interactions, including interaction outcome, from drug labels. Our deep learning approach entails composing various intermediate representations, including graph-based context derived using graph convolutions (GCs) with a novel attention-based gating mechanism (holistically called GCA), which are combined in meaningful ways to predict on all subtasks jointly. Our model is trained and evaluated on the 2018 TAC DDI corpus. Our GCA model in conjunction with transfer learning performs at 39.20% F1 and 26.09% F1 on entity recognition (ER) and relation extraction (RE), respectively, on the first official test set and at 45.30% F1 and 27.87% F1 on ER and RE, respectively, on the second official test set. These updated results lead to improvements over our prior best by up to 6 absolute F1 points. After controlling for available training data, the proposed model exhibits state-of-The-Art performance for this task.

Original languageEnglish (US)
Article number2637-8051
JournalACM Transactions on Computing for Healthcare
Issue number2
StatePublished - Mar 2021
Externally publishedYes


  • Drug-drug interactions
  • Multi-Task learning
  • Neural networks
  • Relation extraction

ASJC Scopus subject areas

  • Software
  • Medicine (miscellaneous)
  • Information Systems
  • Biomedical Engineering
  • Computer Science Applications
  • Health Informatics
  • Health Information Management


Dive into the research topics of 'Attention-gated graph convolutions for extracting drug interaction information from drug labels'. Together they form a unique fingerprint.

Cite this