arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.09469cs.CLcs.AI
收藏

NOWJ @BioCreative IX ToxHabits: 一种用于检测临床文本中物质使用及上下文信息的集成深度学习方法

NOWJ @BioCreative IX ToxHabits: An Ensemble Deep Learning Approach for Detecting Substance Use and Contextual Information in Clinical Texts

University of Engineering and Technology(工程大学);University of Engineering(工程大学);Technology, Vietnam National University(技术,越南国家大学)

Huu-Huy-Hoang Tran, Gia-Bao Duong, Quoc-Viet-Anh Tran, Thi-Hai-Yen Vuong, Hoang-Quynh Le

AI总结:

本文提出了一种集成深度学习方法,用于在西班牙临床文本中检测有毒物质使用及上下文信息,实现了较高的F1和精度指标。

摘要:

从非结构化电子健康记录中提取药物使用信息仍然是临床自然语言处理中的主要挑战。尽管大语言模型显示出进展,但其在临床NLP中的应用受到对信任、控制和效率的担忧的限制。为此,我们提出了NOWJ在BioCreative IX的ToxHabits共享任务中的提交。该任务旨在检测西班牙临床文本中的有毒物质使用及上下文属性,这是一个领域特定且资源有限的环境。我们提出了一种多输出集成系统,同时处理子任务1-ToxNER和子任务2-ToxUse。我们的系统整合了BETO与CRF层进行序列标注,采用多样化的训练策略,并利用句子过滤来提高精度。我们的最佳运行在触发检测上实现了0.94 F1和0.97精度,在论证检测上实现了0.91 F1。

英文摘要:

Extracting drug use information from unstructured Electronic Health Records remains a major challenge in clinical Natural Language Processing. While Large Language Models demonstrate advancements, their use in clinical NLP is limited by concerns over trust, control, and efficiency. To address this, we present NOWJ submission to the ToxHabits Shared Task at BioCreative IX. This task targets the detection of toxic substance use and contextual attributes in Spanish clinical texts, a domain-specific, low-resource setting. We propose a multi-output ensemble system tackling both Subtask 1 - ToxNER and Subtask 2 - ToxUse. Our system integrates BETO with a CRF layer for sequence labeling, employs diverse training strategies, and uses sentence filtering to boost precision. Our top run achieved 0.94 F1 and 0.97 precision for Trigger Detection, and 0.91 F1 for Argument Detection.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑