Deep Learning-based Propensity Scores for Confounding Control in Comparative Effectiveness Research: A Large-scale, Real-world Data Study.

Journal: Epidemiology (Cambridge, Mass.)
PMID: 33591049

Abstract

BACKGROUND: Due to the non-randomized nature of real-world data, prognostic factors need to be balanced, which is often done by propensity scores (PSs). This study aimed to investigate whether autoencoders, which are unsupervised deep learning architectures, might be leveraged to compute PS.

Authors

  • Janick Weberpals
    From the Data Science, Pharmaceutical Research and Early Development Informatics (pREDi), Roche Innovation Center Munich (RICM), Penzberg, Germany.
  • Tim Becker
    xValue GmbH, Willich, Germany, on behalf of Data Science IV, Pharmaceutical Research and Early Development Informatics (pREDi), Roche Innovation Center Munich (RICM), Penzberg, Germany.
  • Jessica Davies
    F. Hoffmann-La Roche Ltd, Welwyn Garden City, United Kingdom.
  • Fabian Schmich
    From the Data Science, Pharmaceutical Research and Early Development Informatics (pREDi), Roche Innovation Center Munich (RICM), Penzberg, Germany.
  • Dominik Rüttinger
    Early Clinical Development Oncology, Pharmaceutical Research and Early Development (pRED), Roche Innovation Center Munich (RICM), Penzberg, Germany.
  • Fabian J Theis
    Institute of Computational Biology, Helmholtz Zentrum München, German Research Center for Environmental Health, Munich, Germany.
  • Anna Bauer-Mehren
    Center for Biomedical Informatics Research, Stanford University, Stanford, California, USA.

Keywords

External Resources

Popular Topics
Recent Journals