Auto-Regressive Time Delayed jump neural network for blood glucose levels forecasting
作者:
Highlights:
• A novel net named Auto-Regressive Time Delayed (ARTiDe) jump neural network is presented.
• It is tested on blood glucose levels prediction using 24 h of monitoring data.
• The model exploits a univariate approach reducing the burden of data collection.
• The model outperforms many multivariate approaches at the state of the art.
• A thorough overview of glucose levels forecasting is provided.
摘要
•A novel net named Auto-Regressive Time Delayed (ARTiDe) jump neural network is presented.•It is tested on blood glucose levels prediction using 24 h of monitoring data.•The model exploits a univariate approach reducing the burden of data collection.•The model outperforms many multivariate approaches at the state of the art.•A thorough overview of glucose levels forecasting is provided.
论文关键词:Time series forecasting,Neural networks,Precision medicine,Diabetes,Bio-medical patterns
论文评审过程:Received 19 December 2019, Revised 6 June 2020, Accepted 9 June 2020, Available online 12 June 2020, Version of Record 17 June 2020.
论文官网地址:https://doi.org/10.1016/j.knosys.2020.106134