基于双编码器的方法用于生物医学文档级实体识别和关系提取
IEEE transactions on computational biology and bioinformatics
|August 14, 2025
概括
我们开发了BioECR,这是一个有效的方法,用于生物医学文档级别的命名实体识别,核心引用解析和关系提取. 它实现了最先进的结果,并大大缩短了处理时间.
科学领域:
- 生物医学自然语言处理
- 计算生物学 计算生物学
- 生物信息学是一种生物信息学.
背景情况:
- 准确的生物医学实体和关系提取对于知识发现至关重要.
- 由于跨越句子的关系,需要在文档层面进行分析.
- 目前的方法在速度和处理复杂实体方面面临挑战.
研究的目的:
- 引入BioECR,这是一个端到端的系统,用于文档级生物医学命名实体识别,核心引用解析和关系提取.
- 提高复杂和嵌套的生物医学实体的识别.
- 为了提高关系提取任务的效率.
主要方法:
- 使用了带有生物医学实体类型和描述的双编码器结构,用于高效的嵌套实体识别.
- 采用组合图卷积神经网络来减少噪音和选择性地融合信息.
- 集成实体类型聚类,以解决跨多个实体类型的核心引用错误.
主要成果:
- 在三个基准数据集 (CDR,GDA,BioRED) 的所有子任务中实现了最先进的性能.
- 在识别复杂实体和关系方面显著改进.
- 与现有方法相比,推断时间缩短了约60%.
结论:
- 生物ECR为文档级生物医学文本挖掘提供了高效和高效的解决方案.
- 提出的方法成功地解决了嵌套实体,杂图形处理和核心引用解析方面的挑战.
- 这一进步促进了大规模的生物医学知识图表构建和下游应用.
相关概念视频
Methods of Documentation V: CBE
980
Charting by Exception, or CBE, is a method of documentation used in healthcare, particularly in nursing, that focuses on documenting only significant or abnormal findings rather than recording every detail. This approach aims to streamline the documentation process, improve efficiency, and ensure that healthcare providers can quickly identify deviations from normalcy in patient assessments.
In CBE, healthcare professionals establish predefined standards of practice that define what constitutes...
In CBE, healthcare professionals establish predefined standards of practice that define what constitutes...
980
Transducer Mechanism: Enzyme-Linked Receptors
2.8K
Enzyme-linked receptors are cell-surface receptors acting as an enzyme or associating with an enzyme intracellularly. They make excellent drug targets. Drugs can bind to the extracellular ligand-binding domain or directly affect their enzymatic domain and alter their activity.
Major types that are helpful drug targets include:
Major types that are helpful drug targets include:
2.8K
Methods of Documentation VII: EMR
913
Electronic Medical Records (EMRs) primarily center around electronically documenting patients' health information within a single healthcare organization or practice. They contain essential clinical data related to a patient's medical history, diagnoses, medications, treatment plans, lab results, and other pertinent information relevant to the specific encounter or episode of care. EMRs are designed to streamline documentation and workflow processes within individual healthcare...
913


