Nature Communications (Jul 2021)

A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions

  • Dani Kiyasseh,
  • Tingting Zhu,
  • David Clifton

DOI
https://doi.org/10.1038/s41467-021-24483-0
Journal volume & issue
Vol. 12, no. 1
pp. 1 – 11

Abstract

Read online

Deep learning algorithms trained on data streamed temporally from different clinical sites and from a multitude of physiological sensors are generally affected by a degradation in performance. To mitigate this, the authors propose a continual learning strategy that employs a replay buffer.