Predicting life expectancy with a long short-term memory recurrent neural network using electronic medical records.

Journal: BMC medical informatics and decision making
PMID:

Abstract

BACKGROUND: Life expectancy is one of the most important factors in end-of-life decision making. Good prognostication for example helps to determine the course of treatment and helps to anticipate the procurement of health care services and facilities, or more broadly: facilitates Advance Care Planning. Advance Care Planning improves the quality of the final phase of life by stimulating doctors to explore the preferences for end-of-life care with their patients, and people close to the patients. Physicians, however, tend to overestimate life expectancy, and miss the window of opportunity to initiate Advance Care Planning. This research tests the potential of using machine learning and natural language processing techniques for predicting life expectancy from electronic medical records.

Authors

  • Merijn Beeksma
    Centre for Language Studies, Radboud University, Erasmusplein 1, 6525, HT, Nijmegen, The Netherlands. m.t.beeksma@let.ru.nl.
  • Suzan Verberne
    Leiden Institute for Advanced Computer Sciences, Leiden University, Niels Bohrweg 1, 2333, CA, Leiden, The Netherlands.
  • Antal van den Bosch
    KNAW Meertens Institute, Oudezijds Achterburgwal 185, 1012, DK, Amsterdam, The Netherlands.
  • Enny Das
    Centre for Language Studies, Radboud University, Erasmusplein 1, 6525, HT, Nijmegen, The Netherlands.
  • Iris Hendrickx
    Centre for Language Studies, Radboud University, Erasmusplein 1, 6525, HT, Nijmegen, The Netherlands.
  • Stef Groenewoud
    IQ Healthcare, Radboudumc, Mailbox 9101, 6500, HB, Nijmegen, The Netherlands.