International Journal of Biomedical Data Mining
Open Access
ISSN: 2090-4924
ISSN: 2090-4924
Zhijing Li, Yu Long, Xuan Wang, Qinghua Zheng and Chen Li
Temporal information extraction is important to understand the text in clinical documents. Extracting temporal courses of clinical events from a patient’s electronic health records may present structured information of the patient’s symptoms, diagnosis, and treatments etc. in the actually occurring order. We have developed the clinic temporal information extractor, an easy-to-use standalone application, which can automatically extract such information and mitigate the difficulty of analyzing tremendous clinical data. This extractor is able to extract temporal expressions and events with their attributes and relations. The system has been tested on the colon cancer data and brain cancer data of Semeval 2017 Clinical TempEval data and achieves the top performance comparable to the state of the art. It achieves 0.62 F-measure for the temporal expression extraction, 0.73 for event extraction and 0.59 for document time relation extraction
Published Date: 2021-03-31; Received Date: 2021-02-26