Medical knowledge embedding based on recursive neural network for multi-disease diagnosis

作者：

Highlights：

• Topological structure of Huffman tree for representing first-order logic knowledge.

• A recursive neural knowledge network (RNKN) is proposed for multi-disease diagnosis.

• Discriminative weight learning method of RNKN based on back-propagation mechanism.

• Our study results confirmed the interpretability of knowledge embedding.

摘要

•Topological structure of Huffman tree for representing first-order logic knowledge.•A recursive neural knowledge network (RNKN) is proposed for multi-disease diagnosis.•Discriminative weight learning method of RNKN based on back-propagation mechanism.•Our study results confirmed the interpretability of knowledge embedding.

论文关键词：Electronic medical records,First-order logic,Knowledge embedding,Recursive neural network

论文评审过程：Received 23 June 2018, Revised 16 September 2019, Accepted 26 November 2019, Available online 28 November 2019, Version of Record 23 January 2020.

论文官网地址：https://doi.org/10.1016/j.artmed.2019.101772