Reconstructing the patient’s natural history from electronic health records
作者：
Highlights：
• Our NLP framework obtained a correct result in 85 % of instances for building the patient’s medical timeline from Spanish EHRs.
• Our rule-based annotators have shown adequate performance for recognizing lung cancer stage grouping and TNM codes from Spanish clinical texts.
• Our rule-based Temporal Tagger outperformed SUTime and HeidelTime for annotation of date expressions, written in Spanish clinical texts.
• Our Temporal Reasoning System outperformed TIPSem in linking events to time expressions at sentence level.
摘要
•Our NLP framework obtained a correct result in 85 % of instances for building the patient’s medical timeline from Spanish EHRs.•Our rule-based annotators have shown adequate performance for recognizing lung cancer stage grouping and TNM codes from Spanish clinical texts.•Our rule-based Temporal Tagger outperformed SUTime and HeidelTime for annotation of date expressions, written in Spanish clinical texts.•Our Temporal Reasoning System outperformed TIPSem in linking events to time expressions at sentence level.
论文关键词：Electronic Health Records,Natural Language Processing,Temporal Reasoning
论文评审过程：Received 29 October 2019, Revised 6 April 2020, Accepted 6 April 2020, Available online 3 May 2020, Version of Record 11 May 2020.
论文官网地址：https://doi.org/10.1016/j.artmed.2020.101860
