Adverse Drug Event Detection from Electronic Health Records Using Hierarchical Recurrent Neural Networks with Dual-Level Embedding

• Published in: Drug safety Vol. 42; no. 1; pp. 113 - 122
• Main Authors: Wunnava, Susmitha, Qin, Xiao, Kakar, Tabassum, Sen, Cansu, Rundensteiner, Elke A., Kong, Xiangnan
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.01.2019; Springer Nature B.V
• Subjects: Artificial intelligence; Cardiovascular disease; Deep Learning - standards; Deep Learning - trends; Drug Safety and Pharmacovigilance; Drug-Related Side Effects and Adverse Reactions - diagnosis; Drug-Related Side Effects and Adverse Reactions - epidemiology; Electronic health records; Electronic Health Records - standards; Electronic Health Records - trends; Electronic medical records; Embedded systems; Embedding; Humans; Information retrieval; Innovations; International conferences; Language; Linguistics; Long short-term memory; Machine learning; Machine Learning - standards; Machine Learning - trends; Medicine; Medicine & Public Health; Natural language processing; Neural networks; Neural Networks, Computer; NLP Challenge for Detecting Medication and Adverse Drug Events from Electronic Health Records (MADE 1.0); Original Research Article; Pharmacology; Pharmacology/Toxicology; Pharmacovigilance; Recurrent neural networks; Semantics; Terminology

Introduction Adverse drug event (ADE) detection is a vital step towards effective pharmacovigilance and prevention of future incidents caused by potentially harmful ADEs. The electronic health records (EHRs) of patients in hospitals contain valuable information regarding ADEs and hence are an import...