TY - JOUR
T1 - Classifying relations in clinical narratives using segment graph convolutional and recurrent neural networks (Seg-GCRNs)
AU - Li, Yifu
AU - Jin, Ran
AU - Luo, Yuan
N1 - Funding Information:
This work was supported in part by NIH Grant 1R21LM012618-01.
PY - 2019/3/1
Y1 - 2019/3/1
N2 - We propose to use segment graph convolutional and recurrent neural networks (Seg-GCRNs), which use only word embedding and sentence syntactic dependencies, to classify relations from clinical notes without manual feature engineering. In this study, the relations between 2 medical concepts are classified by simultaneously learning representations of text segments in the context of sentence syntactic dependency: preceding, concept1, middle, concept2, and succeeding segments. Seg-GCRN was systematically evaluated on the i2b2/VA relation classification challenge datasets. Experiments show that Seg-GCRN attains state-of-the-art micro-averaged F-measure for all 3 relation categories: 0.692 for classifying medical treatment-problem relations, 0.827 for medical test-problem relations, and 0.741 for medical problem-medical problem relations. Comparison with the previous state-of-the-art segment convolutional neural network (Seg-CNN) suggests that adding syntactic dependency information helps refine medical word embedding and improves concept relation classification without manual feature engineering. Seg-GCRN can be trained efficiently for the i2b2/VA dataset on a GPU platform.
AB - We propose to use segment graph convolutional and recurrent neural networks (Seg-GCRNs), which use only word embedding and sentence syntactic dependencies, to classify relations from clinical notes without manual feature engineering. In this study, the relations between 2 medical concepts are classified by simultaneously learning representations of text segments in the context of sentence syntactic dependency: preceding, concept1, middle, concept2, and succeeding segments. Seg-GCRN was systematically evaluated on the i2b2/VA relation classification challenge datasets. Experiments show that Seg-GCRN attains state-of-the-art micro-averaged F-measure for all 3 relation categories: 0.692 for classifying medical treatment-problem relations, 0.827 for medical test-problem relations, and 0.741 for medical problem-medical problem relations. Comparison with the previous state-of-the-art segment convolutional neural network (Seg-CNN) suggests that adding syntactic dependency information helps refine medical word embedding and improves concept relation classification without manual feature engineering. Seg-GCRN can be trained efficiently for the i2b2/VA dataset on a GPU platform.
KW - Bidirectional long short-term memory networks
KW - Graph convolutional networks
KW - Medical relation classification
KW - Natural language processing
UR - http://www.scopus.com/inward/record.url?scp=85060826879&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060826879&partnerID=8YFLogxK
U2 - 10.1093/jamia/ocy157
DO - 10.1093/jamia/ocy157
M3 - Article
C2 - 30590613
AN - SCOPUS:85060826879
VL - 26
SP - 262
EP - 268
JO - Journal of the American Medical Informatics Association : JAMIA
JF - Journal of the American Medical Informatics Association : JAMIA
SN - 1067-5027
IS - 3
ER -