Chinese Medical Sciences Journal ›› 2019, Vol. 34 ›› Issue (2): 133-139.doi: 10.24920/003589
Special Issue: 医学人工智能
• Original Article • Previous Articles Next Articles
Li Peilin, Yuan Zhenming, Tu Wenbo, Yu Kai, Lu Dongxin
Received:
2019-03-29
Accepted:
2019-04-24
Published:
2019-06-30
Online:
2019-05-14
Based on the research status of deep learning, the paper discussed and built two application scenes of bi-directional long short-term memory combined conditional random field (BiLSTM-CRF) model in NER and MRE tasks. Validation on the I2B2 2010 public dataset showed better performance than the baseline methods in the two task. |
Li Peilin, Yuan Zhenming, Tu Wenbo, Yu Kai, Lu Dongxin. Medical Knowledge Extraction and Analysis from Electronic Medical Records Using Deep Learning[J].Chinese Medical Sciences Journal, 2019, 34(2): 133-139.
Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks
Figure 1.
LSTM neuron structure. The tanh is a classical neural network nonlinear activation function. The input gate, the forgetting gate, and the output gate in the LSTM unit are defined as i, f, and o, respectively. Ct is the state of the storage unit at the current time, and equation (4) represents the process of the state transition of the memory unit. The current state is calculated by the previous time state Ct-1, the result of the forgotten gate ft, and the input gate it of the current time LSTM unit."
Table 1
Categories of medical relations and descriptions"
Category | Description |
---|---|
TrIP | Treatment improves medical problems. |
TrWP | Treatment worsens medical problems. |
TrCP | Treatment causes medical problems. |
TrAP | Treatment is applied to medical problems. |
TrNAP | Treatment is not applied to medical problems. |
TeRP | Tests reveal medical problems. |
TeCP | In order to prove medical problems, need to be checked. |
PIP | The relation between medical problems. |
Figure 2.
Text and labeling example of electronic medical record (EMR). The example has three parts. The first part is the sentence containing the entity in the medical record, where “CNIS” is the abbreviation of calcineurin inhibitors (CnIs) and steroids treatment for vascular disease. The middle part shows the line:column number of the entities in the record and its category. The last part is the relation, TeRP, between the entities (CNIS and carotid stenosis) in the sentence."
Figure 3.
BiLSTM-CRF structure of named entity recognition(NER). “l1”, “r1”, “e1”, etc., represents the different network layers in the model. The CRF layer is a combination label, where “B”, “I”, “E”, “S”, “O” are word segmentation labels and “Te”, “P”, “Tr” are entity category labels. BiLSTM, bidirectional long short-term memory; CRF, conditional random field."
Table 3
The F1-measure of models in MRE"
Models | TrIP | TrWP | TrCP | TrAP | TrNAP | TeRP | TeCP | PIP | Total |
---|---|---|---|---|---|---|---|---|---|
SVM | 0.23 | 0.05 | 0.496 | 0.806 | 0.17 | 0.872 | 0.45 | 0.87 | 0.737 |
ME | 0.216 | 0.02 | 0.502 | 0.814 | 0.193 | 0.859 | 0.393 | 0.91 | 0.731 |
DNN+CRF | 0.225 | 0.03 | 0.534 | 0.86 | 0.225 | 0.916 | 0.451 | 0.96 | 0.752 |
BiLSTM-CRF | 0.251 | 0.11 | 0.572 | 0.903 | 0.35 | 0.931 | 0.503 | 0.98 | 0.775 |
[1] |
Wu JW, Guan Y, Lv XB . Entity relation extraction from electronic medical records based on deep learning. Intell comput appl 2014; 4(3):35-8. doi: .
doi: 10.3969/j.issn.2095-2163.2014.03.009 |
[2] |
Chen L, Li Y, Chen W , et al. Utilizing Soft Constraints to Enhance Medical Relation Extraction from the History of Present Illness in Electronic Medical Records. J Biomed Inform 2018; 87:108-17. doi: .
doi: 10.1016/j.jbi.2018.09.013 |
[3] |
Xue NW, Shen LB . Chinese word segmentation as LMR tagging. Proceedings of the second SIGHAN workshop on Chinese language processing. 2003, Jul 11-12; Sapporo, Japan. Stroudsburg, PA, USA: Association for computation Linguistics; 2003. 17:176-9. doi: .
doi: 10.3115/1119250.1119278 |
[4] |
Finkel JR, Grenager T, Manning C . Incorporating non-local information into information extraction systems by Gibbs sampling. Proceedings of the 43rd annual meeting on association for computational linguistics 2005; 363-70. doi:
doi: 10.3115/1219840.1219885 |
[5] |
Ye F, Chen W, Zhou GG , et al. Intelligent recognition of named entities in electronic medical records. Chin J of Biomed Eng 2011; 30(2):256-62. Chinese. doi: .
doi: 10.3969/j.issn.0258-8021.2011.02.014 |
[6] |
Li W, Zhao DZ, Li B , et al. Entity recognition of medical records combined with CRF and rules. App Res Comput 2015; 32(4):1082-6. doi: .
doi: 10.3969/j.issn.1001-3695.2015.04.029 |
[7] |
Bollegala D, Matsuo Y, Ishizuka M. Relation adaptation: learning to extract novel relations with minimum supervision. Proceedings of the 22nd International Joint Conference of Artificial Intelligence (IJCAI). 2011 Jul 16-22; Barcelona, Spain. Menlo Park, California, USA: AAAI Press; 2011. p. 2205-10. doi: doi:.
doi: 10.5591/978-1-57735-516-8/IJCAI11-368 |
[8] |
Suchanek FM, Ifrim G, Weikum G . Combining linguistic and statistical analysis to extract relations from web documents. Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining. 2006 Aug 20-23; Philadelphia, USA. New York: ACM; 2006. p. 712-7. doi: .
doi: 10.1145/1150402.1150492 |
[9] |
Qin B, Liu AA , Liu T. Unsupervised Chinese Open Entity Relation Extraction. J Comput Res Dev 2015; 52(5):1029-35. Chinese. doi:.
doi: 10.7544/issn1000-1239.2015.20131550 |
[10] |
Uzuner O, Mailoa J, Ryan R , et al. Semantic relations for problem-oriented medical records. Artif Intell Med 2010; 50(2):63-73. doi:
doi: 10.1016/j.artmed.2010.05.006 |
[11] |
Zhou G, Su J, Zhang J , et al. Exploring various knowledge in relation extraction. ACL '05 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics. 2005 Jun 25-30; Michigan, USA. Stroudsburg, PA, USA: Association for computational Linguistics; 2005. p. 427-34. doi: .
doi: 10.3115/1219840.1219893 |
[12] |
Uzuner O, South BR, Shen S , et al. 2010 i2b2/VA challenge on concepts, assertions, and relations in clinical text. J Am Med Inform Assn 2011; 18(5):552-6. doi:
doi: 10.1136/amiajnl-2011-000203 |
[13] |
Rink B, Harabagiu S, Roberts K . Automatic extraction of relations between medical concepts in clinical texts. J Am Med Inform Assn 2011; 18(5):594-600. doi:
doi: 10.1136/amiajnl-2011-000153 |
[14] |
Demner-Fushman D, Mork JG, Shooshan SE , et al. UMLS content views appropriate for NLP processing of the biomedical literature vs. clinical text. J Biomed Inform 2010; 43(4):587-94. doi:
doi: 10.1016/j.jbi.2010.02.005 |
[15] |
Lv X, Yi G, Yang J , et al. Clinical relation extraction with deep learning. Int J Inf Tech Decis 2016; 9(7):237-48. doi: .
doi: 10.14257/ijhit.2016.9.7.22 |
[16] |
Pennington J, Socher R , Manning C. Glove: Global vectors for word representation. Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). 2014 Oct 25-29; Doha, Qatar. p. 1532-43. doi: .
doi: 10.3115/v1/D14-1162 |
[17] | Yin W, Kann K, Yu M , et al. Comparative study of CNN and RNN for natural language Processing 2017; arXiv: 1702. 01923. |
[18] |
Yang JF, Yu QB, Guan Y , et al. A Review of research on name recognition and entity relation extraction of electronic medical records. Acta Automatica Sinica 2014; 40(8):1537-62. doi: .
doi: 10.3724/SP.J.1004.2014.01537 |
[19] |
Jiang M, Chen Y, Liu M , et al. A study of machine-learning-based approaches to extract clinical entities and their assertions from discharge summaries. J Am Med Inform Assn 2011; 18(5):601-6. doi:
doi: 10.1136/amiajnl-2011-000163 |
[20] |
De Bruijn B, Cherry C, Kiritchenko S , et al. Machine-learned solutions for three stages of clinical information extraction: the state of the art at i2b2 2010. J Am Med Inform Assn 2011; 18(5):557-62. doi:
doi: 10.1136/amiajnl-2011-000150 |
[1] | Wei Ba, Shuhao Wang, Cancheng Liu, Yuefeng Wang, Huaiyin Shi, Zhigang Song. Histopathological Diagnosis System for Gastritis Using Deep Learning Algorithm [J]. Chinese Medical Sciences Journal, 2021, 36(3): 204-209. |
[2] | Lianyan Xu, Ke Yan, Le Lu, Weihong Zhang, Xu Chen, Xiaofei Huo, Jingjing Lu. External and Internal Validation of a Computer Assisted Diagnostic Model for Detecting Multi-Organ Mass Lesions in CT images [J]. Chinese Medical Sciences Journal, 2021, 36(3): 210-217. |
[3] | Jiazheng Li, Lei Tang. Radiomics in Antineoplastic Agents Development: Application and Challenge in Response Evaluation [J]. Chinese Medical Sciences Journal, 2021, 36(3): 187-195. |
[4] | Dasheng Li,Dawei Wang,Nana Wang,Haiwang Xu,He Huang,Jianping Dong,Chen Xia. An Insight of the First Community Infected COVID-19 Patient in Beijing by Imported Case: Role of Deep Learning-Assisted CT Diagnosis [J]. Chinese Medical Sciences Journal, 2021, 36(1): 66-71. |
[5] | Shi Ying-huan,Wang Qian. The Artificial Intelligence-Enabled Medical Imaging: Today and Its Future [J]. Chinese Medical Sciences Journal, 2019, 34(2): 71-75. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||
|
Supervised by National Health & Family Plan Commission of PRC
9 Dongdan Santiao, Dongcheng district, Beijing, 100730 China
Tel: 86-10-65105897 Fax:86-10-65133074
E-mail: cmsj@cams.cn www.cmsj.cams.cn
Copyright © 2018 Chinese Academy of Medical Sciences
All right reserved.
京公安备110402430088 京ICP备06002729号-1