摘要
Electronic medical data contains biochemical, imaging, pathological information during diagnosis and treatment. The pathology report is a kind of highly liberalized unstructured textual data, which is the basis and gold standard of cancer diagnosis and is very important for the prognosis and treatment of patients. The application of information extraction technology to pathological reports can obtain structured data that can be understood and analyzed by computers, helping pathologists make appropriate decisions. In this work, we proposed an attention-based graph convolutional network (GCN) for converting unstructured pathological reports into a structured form suitable for computer analysis to improve the current pathologist's workflow, collected medical data from different platforms, and provided more accurate assistance for diagnosis and treatment. We used pathology reports data from TCGA (The Cancer Genome Atlas) database with fine-grained annotations on 3632 pathology reports including four types of cancers. Our method performs better in our pathology report dataset with higher F1 score than traditional methods and deep learning methods. The results indicate that our method is robust, thus may work with other types of cancer pathology report.
| 源语言 | 英语 |
|---|---|
| 主期刊名 | Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020 |
| 编辑 | Taesung Park, Young-Rae Cho, Xiaohua Tony Hu, Illhoi Yoo, Hyun Goo Woo, Jianxin Wang, Julio Facelli, Seungyoon Nam, Mingon Kang |
| 出版商 | Institute of Electrical and Electronics Engineers Inc. |
| 页 | 2395-2402 |
| 页数 | 8 |
| ISBN(电子版) | 9781728162157 |
| DOI | |
| 出版状态 | 已出版 - 16 12月 2020 |
| 活动 | 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020 - Virtual, Seoul, 韩国 期限: 16 12月 2020 → 19 12月 2020 |
丛书
| 姓名 | Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020 |
|---|
会议
| 会议 | 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020 |
|---|---|
| 国家/地区 | 韩国 |
| 市 | Virtual, Seoul |
| 时期 | 16/12/20 → 19/12/20 |
联合国可持续发展目标
此成果有助于实现下列可持续发展目标:
-
可持续发展目标 3 良好健康与福祉
学术指纹
探究 'Structured Information Extraction of Pathology Reports with Attention-based Graph Convolutional Network' 的科研主题。它们共同构成独一无二的学术指纹。引用此
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver