Abstract
Electronic Medical Records (EMRs) are the foundation of modern medical information systems. Despite the benefits of EMRs, the exhausting process of constructing EMRs decreases the efficiency of medical consultation. Therefore, it becomes an emerging research field to automatically extract EMRs from medical dialogues. In Chinese medical dialogues, the phenomena of omission and reference are extremely common, leading to strong contextual relevance among utterances. However, recent studies on converting Chinese medical dialogues to EMRs lack a reliable mechanism to effectively exploit the contextual relevance information among utterances. Moreover, they neglect the frequency imbalance of different items and treat these items indiscriminately, which eventually degrade the overall system performance. In this article, we proposed a Contextual-Aware Information Extractor (CANE), which employs a local-to-global mechanism over utterances to model the contextual relevance among utterances. Furthermore, an adaptive objective is introduced to alleviate the frequency imbalance of items by dynamically assigning weights to each sample. Experimental results indicate that CANE outperforms previous state-of-the-arts with considerable improvements (+6.11% and +3.39% on F1-score).
- [1] . 2016. Layer normalization. arXiv:1607.06450. https://arxiv.org/abs/1607.06450.Google Scholar
- [2] . 2014. Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv:1406.1078. https://arxiv.org/abs/1406.1078.Google Scholar
- [3] . 2014. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv:1412.3555. https://arxiv.org/abs/1412.3555.Google Scholar
- [4] . 2019. Extracting symptoms and their status from clinical conversations. arXiv:1906.02239. https://arxiv.org/abs/1906.02239.Google Scholar
- [5] . 2018. An automated medical scribe for documenting clinical encounters. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. 11–15.
DOI: Google ScholarCross Ref
- [6] . 2017. Convolutional sequence to sequence learning. arXiv:1705.03122. https://arxiv.org/abs/1705.03122.Google Scholar
- [7] . 2016. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770–778.
DOI: Google ScholarCross Ref
- [8] . 1997. Long short-term memory. Neural Computation 9, 8 (1997), 1735–1780.
DOI: Google ScholarCross Ref
- [9] . 2015. Bidirectional LSTM-CRF models for sequence tagging. arXiv:1508.01991. https://arxiv.org/abs/1508.01991.Google Scholar
- [10] . 2014. A convolutional neural network for modelling sentences. arXiv:1404.2188. https://arxiv.org/abs/1404.2188.Google Scholar
- [11] . 2019. A method for stochastic optimization. arXiv:1412.6980 434. https://arxiv.org/abs/1412.6980.Google Scholar
- [12] . 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data.Google Scholar
- [13] . 2017. Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision. 2980–2988. https://openaccess.thecvf.com/content_ICCV_2017/papers/Lin_Focal_Loss_for_ICCV_2017_paper.pdf.Google Scholar
Cross Ref
- [14] . 2019. Enhancing dialogue symptom diagnosis with global attention and symptom graph. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 5036–5045.
DOI: Google ScholarCross Ref
- [15] . 2017. A survey of deep neural network architectures and their applications. Neurocomputing 234 (2017), 11–26.
DOI: Google ScholarCross Ref
- [16] . 2020. Auxiliary learning for relation extraction. IEEE Transactions on Emerging Topics in Computational Intelligence (2020), 1–10. Google Scholar
Cross Ref
- [17] . 2016. End-to-end sequence labeling via bi-directional lstm-cnns-crf. arXiv:1603.01354. https://arxiv.org/abs/1603.01354.Google Scholar
- [18] . 2013. Distributed representations of words and phrases and their compositionality. Advances in Neural Information Processing Systems 26 (2013), 3111–3119. https://proceedings.neurips.cc/paper/2013/file/9aa42b31882ec039965f3c4923ce901b-Paper.pdf.Google Scholar
Digital Library
- [19] . 2016. Neural belief tracker: Data-driven dialogue state tracking. arXiv:1606.03777. https://arxiv.org/abs/1606.03777.Google Scholar
- [20] . 2020. A contextual hierarchical attention network with adaptive objective for dialogue state tracking. arXiv:2006.01554.
DOI: Google ScholarCross Ref
- [21] . 2016. Allocation of physician time in ambulatory practice: A time and motion study in 4 specialties. Annals of Internal Medicine 165, 11 (2016), 753–760.
DOI: Google ScholarCross Ref
- [22] . 2014. Sequence to sequence learning with neural networks. Advances in Neural Information Processing Systems 27 (2014), 3104–3112. https://proceedings.neurips.cc/paper/2014/file/a14ac55a4f27472c5d894ec1c3c743d2-Paper.pdf.Google Scholar
Digital Library
- [23] . 1997. Introduction to multi-layer feed-forward neural networks. Chemometrics and Intelligent Laboratory Systems 39, 1 (1997), 43–62.
DOI: Google ScholarCross Ref
- [24] . 2011. 2010 i2b2/VA challenge on concepts, assertions, and relations in clinical text. Journal of the American Medical Informatics Association 18, 5 (2011), 552–556.
DOI: Google ScholarCross Ref
- [25] . 2017. Attention is all you need. Advances in Neural Information Processing Systems 30 (2017), 5998–6008. https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf.Google Scholar
- [26] . 2018. To combat physician burnout and improve care, fix the electronic health record. Harvard Business Review (2018).Google Scholar
- [27] . 2019. Transferable multi-domain state generator for task-oriented dialogue systems. arXiv:1905.08743. https://arxiv.org/abs/1905.08743.Google Scholar
- [28] . 2020. MIE: A medical information extractor towards medical dialogues. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 6460–6469.
DOI: Google ScholarCross Ref
- [29] . 2020. Condition aware and revise transformer for question answering. In Proceedings of The Web Conference 2020. 2377–2387. Google Scholar
Cross Ref
- [30] . 2018. Global-locally self-attentive dialogue state tracker. arXiv:1805.09655. https://arxiv.org/abs/1805.09655.Google Scholar
- [31] . 2016. Attention-based bidirectional long short-term memory networks for relation classification. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 207–212. https://aclanthology.org/P16-2034.pdf.Google Scholar
Cross Ref
Index Terms
Contextual-Aware Information Extractor with Adaptive Objective for Chinese Medical Dialogues
Recommendations
EMR Usage and Nurse Documentation Burden in a Medical Intensive Care Unit
Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management. Health, Operations Management, and DesignAbstractElectronic medical records (EMRs) are a standard documentation system that contains vital patient health history. Although EMR system was implemented to improve nurse documentation, many healthcare workers feel the burden of EMR documentation. Due ...
Using Latent Class Analysis to Identify Sophistication Categories of Electronic Medical Record Systems in U.S. Acute Care Hospitals
Many believe that electronic medical record (EMR) systems hold promise for improving the quality of health care services. The body of research on this topic is still in the early stages, however, in part because of the challenge of measuring the ...
Nurses' Views on Electronic Medical Records (EMR) in Turkey: An Analysis According to Use, Quality and User Satisfaction
Electronic medical records are generally used by nurses in hospitals. However, studies investigating views on and evaluations of electronic medical records by nurses are limited in Turkey and in other countries around the world. Thus, in this study, ...






Comments