Deep Learning Strategies for Automatic Detection of Medication and Adverse Drug Events from Electronic Health Records
暂无分享,去创建一个
Background. Detecting the occurrence of adverse drug events (ADEs) and related medical information is an integral step towards the prevention of future critical ADE incidents threatening the public. Electronic health records (EHRs) of patients, a valuable source for potential ADE signals, are unstructured reports comprised of non-medical descriptive text and complex medical terminology1. Therefore, detecting a complete phrase that represents a named entity signaling an incidence is challenging. Recent research, leveraging popular deep learning approaches for the detection of medical information from EHRs, has outperformed state-of-the-art conditional random fields (CRF)2. An integration of recurrent neural networks (RNN) with CRF has also shown good performance3.
[1] Hong Yu,et al. Structured prediction models for RNN based sequence labeling in clinical text , 2016, EMNLP.
[2] Elke A. Rundensteiner,et al. One Size Does Not Fit All: An Ensemble Approach Towards Information Extraction from Adverse Drug Event Narratives , 2018, HEALTHINF.
[3] Hong Yu,et al. Bidirectional RNN for Medical Event Detection in Electronic Health Records , 2016, NAACL.