Optimizing neural networks for medical data sets: A case study on neonatal apnea prediction

作者：

Highlights：

• Optimal machine learning models for predicting the presence of apnea episodes after the first week of admission at NICU.

• A generic framework for stepwise empirical tuning of neural networks, selection of algorithms with their hyper parametric optimization.

• The results indicate that optimized Multilayer Perceptron outperforms the other conventional machine learning models in predicting neonatal apnea.

• The investigated models can help neonatologist for decision making as a diagnostic tool.

摘要

•Optimal machine learning models for predicting the presence of apnea episodes after the first week of admission at NICU.•A generic framework for stepwise empirical tuning of neural networks, selection of algorithms with their hyper parametric optimization.•The results indicate that optimized Multilayer Perceptron outperforms the other conventional machine learning models in predicting neonatal apnea.•The investigated models can help neonatologist for decision making as a diagnostic tool.

论文关键词：Deep network architectures,Multi-layer perceptron,Optimizing neural network,Deep belief networks,Deep autoencoders

论文评审过程：Received 16 January 2018, Revised 9 February 2019, Accepted 24 July 2019, Available online 25 July 2019, Version of Record 30 July 2019.

论文官网地址：https://doi.org/10.1016/j.artmed.2019.07.008