Validation of a Natural Language Machine Learning Model for Safety Literature Surveillance.

Journal: Drug safety
Published Date:

Abstract

INTRODUCTION: As part of routine safety surveillance, thousands of articles of potential interest are manually triaged for review by safety surveillance teams. This manual triage task is an interesting candidate for automation based on the abundance of process data available for training, the performance of natural language processing algorithms for this type of cognitive task, and the small number of safety signals that originate from literature review, resulting in its lower risk profile. However, deep learning algorithms introduce unique risks and the validation of such models for use in Good Pharmacovigilance Practice remains an open question.

Authors

  • Jiyoon Park
    Global Patient Safety, Chief Medical Office, AstraZeneca, Gaithersburg, MD, USA.
  • Malek Djelassi
    Enterprise AI Services, IGNITE IT, AstraZeneca, Mölndal, Sweden.
  • Daniel Chima
    Global Patient Safety, Chief Medical Office, AstraZeneca, Gaithersburg, MD, USA.
  • Robert Hernandez
    Enterprise AI Services, IGNITE IT, AstraZeneca, Cambridge, UK.
  • Vladimir Poroshin
    Enterprise AI Services, IGNITE IT, AstraZeneca, Cambridge, UK.
  • Ana-Maria Iliescu
    Global Patient Safety, Chief Medical Office, AstraZeneca, Mölndal, Sweden.
  • Douglas Domalik
    Global Patient Safety, Chief Medical Office, AstraZeneca, Gaithersburg, MD, USA.
  • Noel Southall
    National Center for Advancing Translational Sciences (NCATS) , National Institutes of Health , 9800 Medical Center Drive , Rockville , Maryland 20850 , United States.