Publication date: Jul 15, 2019
This paper presents a conditional random fields (CRF) method that enables the capture of specific high-order label transition factors to improve clinical named entity recognition performance. Consecutive clinical entities in a sentence are usually separated from each other, and the textual descriptions in clinical narrative documents frequently indicate causal or posterior relationships that can be used to facilitate clinical named entity recognition. However, the CRF that is generally used for named entity recognition is a first-order model that constrains label transition dependency of adjoining labels under the Markov assumption.
Based on the first-order structure, our proposed model utilizes non-entity tokens between separated entities as an information transmission medium by applying a label induction method. The model is referred to as precursor-induced CRF because its non-entity state memorizes precursor entity information, and the model’s structure allows the precursor entity information to propagate forward through the label sequence.
We compared the proposed model with both first- and second-order CRFs in terms of their F-scores, using two clinical named entity recognition corpora (the i2b2 2012 challenge and the Seoul National University Hospital electronic health record). The proposed model demonstrated better entity recognition performance than both the first- and second-order CRFs and was also more efficient than the higher-order model.
The proposed precursor-induced CRF which uses non-entity labels as label transition information improves entity recognition F score by exploiting long-distance transition factors without exponentially increasing the computational time. In contrast, a conventional second-order CRF model that uses longer distance transition factors showed even worse results than the first-order model and required the longest computation time. Thus, the proposed model could offer a considerable performance improvement over current clinical named entity recognition methods based on the CRF models.
Open Access PDF
Lee, W. and Choi, J. Precursor-induced conditional random fields: connecting separate entities by induction for improved clinical named entity recognition. 05007. 2019 BMC Med Inform Decis Mak (19):1.
- Cost-aware active learning for named entity recognition in clinical text.
- Integrating biomedical research and electronic health records to create knowledge-based biologically meaningful machine-readable embeddings.
- Analysis and prediction of unplanned intensive care unit readmission using recurrent neural networks with long short-term memory.
- S17.3 The NIH All of us program: Building a national research program to advance precision medicine
- Pathology Machine Learning in Practice: Clinical trial and translational research applications