Rebecka Weegar1, Alicia Pérez2, Arantza Casillas2, Maite Oronoz2. 1. Department of Computer and Systems Sciences, DSV, Stockholm University, Borgarfjordsgatan 12, Kista, Sweden. rebeckaw@dsv.su.se. 2. IXA (UPV/EHU), University of the Basque Country, M. Lardizabal 1, Donostia, 20080, Spain.
Abstract
BACKGROUND: Text mining and natural language processing of clinical text, such as notes from electronic health records, requires specific consideration of the specialized characteristics of these texts. Deep learning methods could potentially mitigate domain specific challenges such as limited access to in-domain tools and data sets. METHODS: A bi-directional Long Short-Term Memory network is applied to clinical notes in Spanish and Swedish for the task of medical named entity recognition. Several types of embeddings, both generated from in-domain and out-of-domain text corpora, and a number of generation and combination strategies for embeddings have been evaluated in order to investigate different input representations and the influence of domain on the final results. RESULTS: For Spanish, a micro averaged F1-score of 75.25 was obtained and for Swedish, the corresponding score was 76.04. The best results for both languages were achieved using embeddings generated from in-domain corpora extracted from electronic health records, but embeddings generated from related domains were also found to be beneficial. CONCLUSIONS: A recurrent neural network with in-domain embeddings improved the medical named entity recognition compared to shallow learning methods, showing this combination to be suitable for entity recognition in clinical text for both languages.
BACKGROUND: Text mining and natural language processing of clinical text, such as notes from electronic health records, requires specific consideration of the specialized characteristics of these texts. Deep learning methods could potentially mitigate domain specific challenges such as limited access to in-domain tools and data sets. METHODS: A bi-directional Long Short-Term Memory network is applied to clinical notes in Spanish and Swedish for the task of medical named entity recognition. Several types of embeddings, both generated from in-domain and out-of-domain text corpora, and a number of generation and combination strategies for embeddings have been evaluated in order to investigate different input representations and the influence of domain on the final results. RESULTS: For Spanish, a micro averaged F1-score of 75.25 was obtained and for Swedish, the corresponding score was 76.04. The best results for both languages were achieved using embeddings generated from in-domain corpora extracted from electronic health records, but embeddings generated from related domains were also found to be beneficial. CONCLUSIONS: A recurrent neural network with in-domain embeddings improved the medical named entity recognition compared to shallow learning methods, showing this combination to be suitable for entity recognition in clinical text for both languages.
Entities:
Keywords:
Clinical text mining; Medical named entity recognition; Recurrent neural network; Unstructured electronic health records