Prediction of fetal state from the cardiotocogram recordings using neural network models

作者：

Highlights：

• Cardiotocography and uterine activity are used for prediction of the fetal state.

• Multi-layer of sub ANFIS topology is proposed for 21 input variables.

• Deep stacked sparse auto-encoders and deep-ANFIS are applied based on a UCI data set.

• Deep learning is more precise for predicting fetal state.

摘要

•Cardiotocography and uterine activity are used for prediction of the fetal state.•Multi-layer of sub ANFIS topology is proposed for 21 input variables.•Deep stacked sparse auto-encoders and deep-ANFIS are applied based on a UCI data set.•Deep learning is more precise for predicting fetal state.

论文关键词：Deep stacked sparse auto-encoders,Fetal state,Neural network,Adaptive neuro fuzzy inference network

论文评审过程：Received 9 August 2018, Revised 2 February 2019, Accepted 17 March 2019, Available online 19 March 2019, Version of Record 23 March 2019.

论文官网地址：https://doi.org/10.1016/j.artmed.2019.03.005