Neural identification of Type 1 Diabetes Mellitus for care and forecasting of risk events

作者：

Highlights：

• With RHOON we perform identification of the dynamics of blood glucose.

• 35 min of blood glucose were predicted with RHONN.

• Training for the identification and prediction network was done online.

• RHONN predictions are compared with three methods.

摘要

•With RHOON we perform identification of the dynamics of blood glucose.•35 min of blood glucose were predicted with RHONN.•Training for the identification and prediction network was done online.•RHONN predictions are compared with three methods.

论文关键词：Type 1 Diabetes Mellitus,Neural network,Identification,Multi-step ahead predictor,RHONN

论文评审过程：Received 25 July 2020, Revised 6 November 2020, Accepted 5 June 2021, Available online 12 June 2021, Version of Record 22 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115367