Tran Tung, Kavuluru Ramakanth, Kilicoglu Halil
University of Kentucky, United States.
National Library of Medicine, United States.
ACM Trans Comput Healthc. 2021 Mar;2(2). doi: 10.1145/3423209.
Preventable adverse events as a result of medical errors present a growing concern in the healthcare system. As drug-drug interactions (DDIs) may lead to preventable adverse events, being able to extract DDIs from drug labels into a machine-processable form is an important step toward effective dissemination of drug safety information. Herein, we tackle the problem of jointly extracting mentions of drugs and their interactions, including interaction , from drug labels. Our deep learning approach entails composing various intermediate representations, including graph-based context derived using graph convolutions (GCs) with a novel attention-based gating mechanism (holistically called GCA), which are combined in meaningful ways to predict on all subtasks jointly. Our model is trained and evaluated on the 2018 TAC DDI corpus. Our GCA model in conjunction with transfer learning performs at 39.20% F1 and 26.09% F1 on entity recognition (ER) and relation extraction (RE), respectively, on the first official test set and at 45.30% F1 and 27.87% F1 on ER and RE, respectively, on the second official test set. These updated results lead to improvements over our prior best by up to 6 absolute F1 points. After controlling for available training data, the proposed model exhibits state-of-the-art performance for this task.
医疗差错导致的可预防不良事件在医疗系统中日益受到关注。由于药物相互作用（DDIs）可能导致可预防的不良事件，能够将药物相互作用从药品标签中提取为机器可处理的形式，是有效传播药物安全信息的重要一步。在此，我们解决了从药品标签中联合提取药物及其相互作用（包括相互作用类型）提及内容的问题。我们的深度学习方法需要构建各种中间表示，包括使用带有新型注意力门控机制的图卷积（GCs）推导的基于图的上下文（整体称为GCA），这些表示以有意义的方式组合在一起，以便对所有子任务进行联合预测。我们的模型在2018年TAC DDI语料库上进行训练和评估。我们的GCA模型结合迁移学习，在第一个官方测试集上实体识别（ER）和关系提取（RE）的F1值分别为39.20%和26.09%，在第二个官方测试集上ER和RE的F1值分别为45.30%和27.87%。这些更新后的结果比我们之前的最佳结果提高了多达6个绝对F1点。在控制了可用训练数据后，所提出的模型在该任务中表现出了最先进的性能。
