An ensemble of neural models for nested adverse drug events and medication extraction with subwords

Abstract

Objective This article describes an ensembling system to automatically extract adverse drug events and drug related entities from clinical narratives, which was developed for the 2018 n2c2 Shared Task Track 2. Materials and Methods We designed a neural model to tackle both nested (entities embedded in other entities) and polysemous entities (entities annotated with multiple semantic types) based on MIMIC III discharge summaries. To better represent rare and unknown words in entities, we further tokenized the MIMIC III data set by splitting the words into finer-grained subwords. We finally combined all the models to boost the performance. Additionally, we implemented a featured-based conditional random field model and created an ensemble to combine its predictions with those of the neural model. Results Our method achieved 92.78% lenient micro F1-score, with 95.99% lenient precision, and 89.79% lenient recall, respectively. Experimental results showed that combining the predictions of either multiple models, or of a single model with different settings can improve performance. Discussion Analysis of the development set showed that our neural models can detect more informative text regions than feature-based conditional random field models. Furthermore, most entity types significantly benefit from subword representation, which also allows us to extract sparse entities, especially nested entities. Conclusion The overall results have demonstrated that the ensemble method can accurately recognize entities, including nested and polysemous entities. Additionally, our method can recognize sparse entities by reconsidering the clinical narratives at a finer-grained subword level, rather than at the word level.

New and Increasing Rates of Adverse Events Can be Found in Unstructured Text in Electronic Health Records using the Shakespeare Method

“…We used EHRs for critical care admissions within an adult hospital, Beth Israel Deaconess Medical Center, Boston, MA the Medical Information Mart for Intensive Care III (MIMIC-III) [35, 44], which used one medical record system in 2001-2008 and another afterwards. We received the real dates, within several weeks, for the earlier data.…”

Adverse drug events and medication relation extraction in electronic health records with ensemble deep learning methods

“…To perform end-to-end RE, we build a pipeline system. We utilize the ensemble of state-of-the-art BiLSTM-CRF 31 models and simpler feature-based CRF models for detection of named entities 32 . The former model is able to recognize nested entities inside sentences which are essentially entities embedded into other entities.…”

Big data augmentated business trend identification: the case of mobile commerce

“…The approach based on bare keywords without semantic multipliers (such as word2vec based associative/synonymous series of terms) is even weaker than the use of official taxonomies, as the results are ultra-sensitive to meaningless fluctuations in word usage patterns, homonyms, and lexical hypes (see application of these approaches in Goldberg and Levy 2014 ; Rong 2014 ; Vidra 2015 ; Kim et al 2020 ). To find, download and process the most relevant documents that potentially contain information about emerging S&T areas, researchers apply iterative processes from general search categories and terms to more concrete words and phrases (Ju et al 2020 ; Huang et al 2015 ).…”