Bidirectional LSTM-CRF for Clinical Concept Extraction

TL;DR

This paper presents a bidirectional LSTM-CRF model utilizing general-purpose word embeddings for clinical concept extraction, achieving competitive results without relying on extensive domain-specific features.

Contribution

The study introduces a neural network approach for clinical concept extraction that reduces dependence on handcrafted features and domain resources.

Findings

Achieved competitive performance on i2b2/VA dataset

Outperformed traditional feature-based methods

Demonstrated effectiveness of general-purpose embeddings

Abstract

Extraction of concepts present in patient clinical records is an essential step in clinical research. The 2010 i2b2/VA Workshop on Natural Language Processing Challenges for clinical records presented concept extraction (CE) task, with aim to identify concepts (such as treatments, tests, problems) and classify them into predefined categories. State-of-the-art CE approaches heavily rely on hand crafted features and domain specific resources which are hard to collect and tune. For this reason, this paper employs bidirectional LSTM with CRF decoding initialized with general purpose off-the-shelf word embeddings for CE. The experimental results achieved on 2010 i2b2/VA reference standard corpora using bidirectional LSTM CRF ranks closely with top ranked systems.