DeepAISE – An interpretable and recurrent neural survival model for early prediction of sepsis

作者：

Highlights：

• Sepsis is among the leading causes of morbidity, mortality, and cost overruns in the Intensive Care Unit (ICU).

• Early prediction of sepsis can improve situational awareness and facilitate timely, protective interventions.

• DeepAISE (Deep Artificial Intelligence Sepsis Expert), a recurrent neural survival model for the early prediction of sepsis.

• DeepAISE provides rationale for alerts by tracking the top features contributing to the sepsis score as a function of time.

摘要

•Sepsis is among the leading causes of morbidity, mortality, and cost overruns in the Intensive Care Unit (ICU).•Early prediction of sepsis can improve situational awareness and facilitate timely, protective interventions.•DeepAISE (Deep Artificial Intelligence Sepsis Expert), a recurrent neural survival model for the early prediction of sepsis.•DeepAISE provides rationale for alerts by tracking the top features contributing to the sepsis score as a function of time.

论文关键词：Deep learning,Sepsis,Artificial intelligence,Interpretability,Transfer learning

论文评审过程：Received 1 June 2020, Revised 13 January 2021, Accepted 9 February 2021, Available online 13 February 2021, Version of Record 26 February 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102036