arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.03531cs.AIcs.CL
收藏

利用情境相关性通过零样本学习识别临床记录中的自杀文档

Leveraging Contextual Relatedness to Identify Suicide Documentation in Clinical Notes through Zero Shot Learning

The George Washington University(乔治华盛顿大学);VA Medical Center, Washington, DC(华盛顿退伍军人医疗中心);Yale School of Medicine(耶鲁大学医学院);VA Connecticut Healthcare System(康涅狄格退伍军人医疗系统);University of Southern California(南加州大学)
Terri Elizabeth Workman, Joseph L. Goulet, Cynthia A. Brandt, Allison R. Warren, Jacob Eleazer, Melissa Skanderson, Luke Lindemann, John R. Blosnich, John O Lea… 展开作者

Terri Elizabeth Workman, Joseph L. Goulet, Cynthia A. Brandt, Allison R. Warren, Jacob Eleazer, Melissa Skanderson, Luke Lindemann, John R. Blosnich, John O Leary, Qing Zeng Treitler

摘要:

在临床笔记中的电子健康记录数据中识别自杀倾向(包括自杀意念、企图和风险因素)是困难的。一个主要的困难是缺乏训练样本,因为在越来越多的接受筛查的患者中,真正的阳性病例数量很少。本文描述了一种新的方法,通过解决这个数据稀疏性问题,通过零样本学习,识别自杀的临床笔记。美国退伍军人事务部的临床记录作为数据。训练数据集标签使用自杀企图和自我伤害的诊断代码确定。与自杀倾向的目标标签相关的基本字符串用于通过将阳性训练病例缩小到包含基本字符串的病例来提供辅助信息。通过将训练文档内容映射到语义空间来训练深度神经网络。为了进行比较,我们使用相同的训练数据集标签和词袋特征训练了另一个深度神经网络。零样本学习模型在多个概率阈值下的AUC、灵敏度、特异性和阳性预测值方面优于基线模型。在应用0.90的概率阈值时,该方法识别出与记录自杀倾向的相关ICD 10 CM代码无关的笔记,准确率为94%。这种新方法可以有效地识别自杀倾向,而不需要人工注释。

英文摘要:

Identifying suicidality including suicidal ideation, attempts, and risk factors in electronic health record data in clinical notes is difficult. A major difficulty is the lack of training samples given the small number of true positive instances among the increasingly large number of patients being screened. This paper describes a novel methodology that identifies suicidality in clinical notes by addressing this data sparsity issue through zero-shot learning. U.S. Veterans Affairs clinical notes served as data. The training dataset label was determined using diagnostic codes of suicide attempt and self-harm. A base string associated with the target label of suicidality was used to provide auxiliary information by narrowing the positive training cases to those containing the base string. A deep neural network was trained by mapping the training documents contents to a semantic space. For comparison, we trained another deep neural network using the identical training dataset labels and bag-of-words features. The zero shot learning model outperformed the baseline model in terms of AUC, sensitivity, specificity, and positive predictive value at multiple probability thresholds. In applying a 0.90 probability threshold, the methodology identified notes not associated with a relevant ICD 10 CM code that documented suicidality, with 94 percent accuracy. This new method can effectively identify suicidality without requiring manual annotation.

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。

↑