A comparison between discrete and continuous time Bayesian networks in learning from clinical time series data with irregularity

作者：

Highlights：

• The conventional COPD exacerbation detection is reformulated in terms of symptom dynamics.

• Two temporal Bayesian networks are used to model the dynamics of COPD symptoms from unevenly spaced clinical time series.

• Hyperparameters and evidence type should be taken into consideration in continuous-time Bayesian models.

摘要

•The conventional COPD exacerbation detection is reformulated in terms of symptom dynamics.•Two temporal Bayesian networks are used to model the dynamics of COPD symptoms from unevenly spaced clinical time series.•Hyperparameters and evidence type should be taken into consideration in continuous-time Bayesian models.

论文关键词：Dynamic Bayesian networks,Continuous-time Bayesian networks,Point evidence,Interval evidence,Irregular time-series data,COPD

论文评审过程：Received 19 September 2017, Revised 12 September 2018, Accepted 3 October 2018, Available online 22 January 2019, Version of Record 20 March 2019.

论文官网地址：https://doi.org/10.1016/j.artmed.2018.10.002