Segment convolutional neural networks (Seg-CNNs) for classifying relations in clinical notes
Segment convolutional neural networks (Seg-CNNs) for classifying relations in clinical notes
复制标题
DOI：
10.1093/jamia/ocx090
复制
发表时间：
2018-01-01
影响因子：
6.4
通讯作者：
Starren, Justin
中科院分区：
文献类型：
--
作者：
Luo, Yuan;Cheng, Yu;Starren, Justin
我们提出了分段卷积神经网络（Seg-CNN），用于对临床笔记中的关系进行分类。Seg-CNN仅使用单词嵌入特征，而无需手动特征工程。与典型的CNN模型不同，两个概念之间的关系是通过同时学习句子中文本片段的单独表示来识别的：preceding，concept（1），middle，concept（2），successful。我们在i2 b2/VA关系分类挑战数据集上评估Seg-CNN。我们发现Seg-CNN在总体评估中实现了最先进的微平均F测量值0.742，在分类医疗问题-治疗关系时为0.686，在医疗问题-测试关系时为0.820，在医疗问题-医疗问题关系时为0.702。我们展示了学习段级表示的好处。我们发现，医学领域的词嵌入有助于提高关系分类。Seg-CNN可以在图形处理单元（GPU）平台上快速训练i2 b2/VA数据集。这些结果支持使用在文本段上计算的CNN对医疗关系进行分类，因为它们显示了最先进的性能，同时不需要手动特征工程。
We propose Segment Convolutional Neural Networks (Seg-CNNs) for classifying relations from clinical notes. Seg-CNNs use only word-embedding features without manual feature engineering. Unlike typical CNN models, relations between 2 concepts are identified by simultaneously learning separate representations for text segments in a sentence: preceding, concept(1), middle, concept(2), and succeeding. We evaluate Seg-CNN on the i2b2/VA relation classification challenge dataset. We show that Seg-CNN achieves a state-of-the-art micro-average F-measure of 0.742 for overall evaluation, 0.686 for classifying medical problem-treatment relations, 0.820 for medical problem-test relations, and 0.702 for medical problem-medical problem relations. We demonstrate the benefits of learning segment-level representations. We show that medical domain word embeddings help improve relation classification. Seg-CNNs can be trained quickly for the i2b2/VA dataset on a graphics processing unit (GPU) platform. These results support the use of CNNs computed over segments of text for classifying medical relations, as they show state-of-the-art performance while requiring no manual feature engineering.