Modeling asynchronous event sequences with RNNs

Stephen Wu, Sijia Liu, Sunghwan Sohn, Sungrim Moon, Chung il Wi, Young J Juhn, Hongfang D Liu

Research output: Contribution to journalArticle

8 Scopus citations

Abstract

Sequences of events have often been modeled with computational techniques, but typical preprocessing steps and problem settings do not explicitly address the ramifications of timestamped events. Clinical data, such as is found in electronic health records (EHRs), typically comes with timestamp information. In this work, we define event sequences and their properties: synchronicity, evenness, and co-cardinality; we then show how asynchronous, uneven, and multi-cardinal problem settings can support explicit accountings of relative time. Our evaluation uses the temporally sensitive clinical use case of pediatric asthma, which is a chronic disease with symptoms (and lack thereof) evolving over time. We show several approaches to explicitly incorporating relative time into a recurrent neural network (RNN) model that improve the overall classification of patients into those with no asthma, those with persistent asthma, those in long-term remission, and those who have experienced relapse. We also compare and contrast these results with those in an inpatient intensive care setting.

Original languageEnglish (US)
Pages (from-to)167-177
Number of pages11
JournalJournal of Biomedical Informatics
Volume83
DOIs
StatePublished - Jul 1 2018

    Fingerprint

Keywords

  • Asthma
  • Deep learning
  • Electronic health records
  • Temporal data

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this