Details of Research Outputs

TitleINFORMATION PROCESSING SYSTEM FOR CHINESE CLINICAL TEXT
Author (Name in English or Pinyin)
LIU, Jiali
Author (Name in Chinese)刘甲莉
Degree TypeMaster of Philosophy
Supervisor(s) (Name in English or Pinyin)CUI, Shuguang ; CHANG, Tsung Hui
Date Issued2020
Degree GrantorThe Chinese University of Hong Kong
Place of ConferralShenzhen
Degree DisciplineComputer and Information Engineering
languageEnglish
Abstract

The clinical texts are free text document that contains a comprehensive record of the patient’s medical activities and, thus, a large amount of medical information. Extracting medical information from clinical texts is very significant, for building a precision medicine system. In this thesis, we propose to establish a set of complete and feasible annotation guidelines for the named entity and basic relation and build the corresponding corpus based on these guidelines. We developed a Chinese clinical text information processing system, which consists of named entity recognizer and relation extractor, to extract critical medical information from clinical texts. Finally, we visualized the medical information extracted. We tested this system on our dataset. The system has performed very well on our dataset. The F1 value of named entity recognizer is 94.1%, the F1 value of relation extractor is 86.5%, and the F1 value of the whole system is 80.9%. We standardized and summarized the medical information in clinical texts. We anticipate that the developed annotation guideline, corpus, and information extraction system will play a fundamental role in important downstream tasks in precision medicine. 临床文本是自由文本文档,其中包含患者医疗活动的全面记录,因此包含大量医疗信息。从临床文本中提取医学信息对于建立精准医疗系统非常重要。本文提出建立一套完整的,可行的命名实体和实体间基本关系标注规范,并在规范的基础上建立相应的语料库。然后,我们开发了中文临床文本信息处理系统,包括命名实体识别器和关系提取器,从而可以从临床文本中提取关键医学信息。最后,我们将提取的医学信息可视化。我们在自己建立的数据集上测试了该系统。该系统在我们的数据集上表现十分出色。命名识别体的F1值为86.5%,中文临床文本信息处理系统整体的F1值为80.9%。我们对临床资料中的医学信息进行了标准化和总结,我们期望开发的注释指南,语料库和信息处理系统将在精准医疗重要的下游任务中发挥重要作用。

LibraryUniversity Library
Location Theses & Dissertations Collection
Call NumberM.Phil. L58 2020
Document TypeThesis
Identifierhttps://irepository.cuhk.edu.cn/handle/3EPUXD0A/2733
LinksPRIMO
CollectionSchool of Science and Engineering
Recommended Citation
GB/T 7714
LIU, Jiali. INFORMATION PROCESSING SYSTEM FOR CHINESE CLINICAL TEXT[D]. The Chinese University of Hong Kong, Shenzhen,2020.
Files in This Item:
File Name/Size DocType File Type Version Access License
刘甲莉.pdf(13029KB)Thesis-- No AccessCC BY-NC-SA
Related Services
Usage statistics
Google Scholar
Similar articles in Google Scholar
[LIU, Jiali]'s Articles
Baidu academic
Similar articles in Baidu academic
[LIU, Jiali]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[LIU, Jiali]'s Articles
Terms of Use
No data!
Social Bookmark/Share
Please consult the service desk at the University Library regarding access of the hard copy.
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.