Use of BERT (Bidirectional Encoder Representations from Transformers)-Based Deep Learning Method for Extracting Evidences in Chinese Radiology Reports: Development of a Computer-Aided Liver Cancer Diagnosis Framework
单位：[1]School of Biomedical Engineering, Capital Medical University, Beijing, China[2]Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Capital Medical University, Beijing, China[3]Department of Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing, China医技科室影像中心放射科首都医科大学附属北京友谊医院[4]Ministry of Education Key Laboratory of Bioinformatics, Research Department of Bioinformatics at the Beijing National Research Center for Information Science and Technology, Center for Synthetic and Systems Biology, Department of Automation, Tsinghua University, Beijing, China
Background: Liver cancer is a substantial disease burden in China. As one of the primary diagnostic tools for detecting liver cancer, dynamic contrast-enhanced computed tomography provides detailed evidences for diagnosis that are recorded in free-text radiology reports. Objective: The aim of our study was to apply a deep learning model and rule-based natural language processing (NLP) method to identify evidences for liver cancer diagnosis automatically. Methods: We proposed a pretrained, fine-tuned BERT (Bidirectional Encoder Representations from Transformers)-based BiLSTM-CRF (Bidirectional Long Short-Term Memory-Conditional Random Field) model to recognize the phrases of APHE (hyperintense enhancement in the arterial phase) and PDPH (hypointense in the portal and delayed phases). To identify more essential diagnostic evidences, we used the traditional rule-based NLP methods for the extraction of radiological features. APHE, PDPH, and other extracted radiological features were used to design a computer-aided liver cancer diagnosis framework by random forest. Results: The BERT-BiLSTM-CRF predicted the phrases of APHE and PDPH with an F1 score of 98.40% and 90.67%, respectively. The prediction model using combined features had a higher performance (F1 score, 88.55%) than those using APHE and PDPH (84.88%) or other extracted radiological features (83.52%). APHE and PDPH were the top 2 essential features for liver cancer diagnosis. Conclusions: This work was a comprehensive NLP study, wherein we identified evidences for the diagnosis of liver cancer from Chinese radiology reports, considering both clinical knowledge and radiology findings. The BERT-based deep learning method for the extraction of diagnostic evidence achieved state-of-the-art performance. The high performance proves the feasibility of the BERT-BiLSTM-CRF model in information extraction from Chinese radiology reports. The findings of our study suggest that the deep learning-based method for automatically identifying evidences for diagnosis can be extended to other types of Chinese clinical texts.
基金：
National Natural Science Foundation of ChinaNational Natural Science Foundation of China (NSFC) [81701792, 81971707]; National Key Research and Development Program of China [2018YFC0910404]
语种：
外文
被引次数：
WOS：
PubmedID：
中科院(CAS)分区：
出版当年[2020]版：
大类|2 区医学
小类|1 区医学：信息2 区卫生保健与服务
最新[2025]版：
大类|2 区医学
小类|2 区卫生保健与服务2 区医学：信息
JCR分区：
出版当年[2019]版：
Q1HEALTH CARE SCIENCES & SERVICESQ1MEDICAL INFORMATICS
最新[2023]版：
Q1HEALTH CARE SCIENCES & SERVICESQ1MEDICAL INFORMATICS
第一作者单位：[1]School of Biomedical Engineering, Capital Medical University, Beijing, China[2]Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Capital Medical University, Beijing, China
通讯作者：
通讯机构：[1]School of Biomedical Engineering, Capital Medical University, Beijing, China[2]Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Capital Medical University, Beijing, China[*1]School of Biomedical Engineering, Capital Medical University No 10, Xitoutiao, Youanmen, Fengtai District Beijing China
推荐引用方式(GB/T 7714)：
Liu Honglei,Zhang Zhiqiang,Xu Yan,et al.Use of BERT (Bidirectional Encoder Representations from Transformers)-Based Deep Learning Method for Extracting Evidences in Chinese Radiology Reports: Development of a Computer-Aided Liver Cancer Diagnosis Framework[J].JOURNAL of MEDICAL INTERNET RESEARCH.2021,23(1):doi:10.2196/19689.
APA：
Liu, Honglei,Zhang, Zhiqiang,Xu, Yan,Wang, Ni,Huang, Yanqun...&Chen, Hui.(2021).Use of BERT (Bidirectional Encoder Representations from Transformers)-Based Deep Learning Method for Extracting Evidences in Chinese Radiology Reports: Development of a Computer-Aided Liver Cancer Diagnosis Framework.JOURNAL of MEDICAL INTERNET RESEARCH,23,(1)
MLA：
Liu, Honglei,et al."Use of BERT (Bidirectional Encoder Representations from Transformers)-Based Deep Learning Method for Extracting Evidences in Chinese Radiology Reports: Development of a Computer-Aided Liver Cancer Diagnosis Framework".JOURNAL of MEDICAL INTERNET RESEARCH 23..1(2021)
医学