Bleeding Entity Recognition in Electronic Health Records: A Comprehensive Analysis of End-to-End Systems
release_jwhasrhkwrfwzdqzrd6ceulnou
by
Avijit Mitra, Bhanu Pratap Singh Rawat, David McManus, Alok Kapoor, Hong Yu
2021 Volume 2020, p860-869
Abstract
A bleeding event is a common adverse drug reaction amongst patients on anticoagulation and factors critically into a clinician's decision to prescribe or continue anticoagulation for atrial fibrillation. However, bleeding events are not uniformly captured in the administrative data of electronic health records (EHR). As manual review is prohibitively expensive, we investigate the effectiveness of various natural language processing (NLP) methods for automatic extraction of bleeding events. Using our expert-annotated 1,079 de-identified EHR notes, we evaluated state-of-the-art NLP models such as biLSTM-CRF with language modeling, and different BERT variants for six entity types. On our dataset, the biLSTM-CRF surpassed other models resulting in a macro F1-score of 0.75 whereas the performance difference is negligible for sentence and document-level predictions with the best macro F1-scores of 0.84 and 0.96, respectively. Our error analyses suggest that the models' incorrect predictions can be attributed to variability in entity spans, memorization, and missing negation signals.
In text/plain
format
Archived Files and Locations
application/pdf 397.9 kB
file_qcvquk4ysnezln42k6j6fckyja
|
europepmc.org (repository) web.archive.org (webarchive) |
Open Access Publication
Not in DOAJ
In ISSN ROAD
In Keepers Registry
ISSN-L:
1559-4076
access all versions, variants, and formats of this works (eg, pre-prints)