Trajectory-Ordered Objectives for Self-Supervised Representation Learning of Temporal Healthcare Data Using Transformers: Model Development and Evaluation Study.

Journal: JMIR medical informatics
Published Date:

Abstract

BACKGROUND: The growing availability of electronic health records (EHRs) presents an opportunity to enhance patient care by uncovering hidden health risks and improving informed decisions through advanced deep learning methods. However, modeling EHR sequential data, that is, patient trajectories, is challenging due to the evolving relationships between diagnoses and treatments over time. Significant progress has been achieved using transformers and self-supervised learning. While BERT-inspired models using masked language modeling (MLM) capture EHR context, they often struggle with the complex temporal dynamics of disease progression and interventions.

Authors

  • Ali Amirahmadi
    Center for Applied Intelligent Systems Research, Halmstad University, Sweden.
  • Farzaneh Etminani
    Center for Applied Intelligent Systems Research, Halmstad University, Halmstad, Sweden.
  • Jonas Björk
    Department of Clinical Sciences, Lund University, Sweden.
  • Olle Melander
    Department of Clinical Sciences, Lund University, Malmö, Sweden.
  • Mattias Ohlsson
    Department of Astronomy and Theoretical Physics, Lund University, Lund, Sweden.