Gehrmann Sebastian, Dernoncourt Franck, Li Yeran, Carlson Eric T, Wu Joy T, Welt Jonathan, Foote John, Moseley Edward T, Grant David W, Tyler Patrick D, Celi Leo A
MIT Critical Data, Laboratory for Computational Physiology, Cambridge, MA, United States of America.
Harvard SEAS, Harvard University, Cambridge, MA, United States of America.
PLoS One. 2018 Feb 15;13(2):e0192360. doi: 10.1371/journal.pone.0192360. eCollection 2018.
In secondary analysis of electronic health records, a crucial task consists in correctly identifying the patient cohort under investigation. In many cases, the most valuable and relevant information for an accurate classification of medical conditions exist only in clinical narratives. Therefore, it is necessary to use natural language processing (NLP) techniques to extract and evaluate these narratives. The most commonly used approach to this problem relies on extracting a number of clinician-defined medical concepts from text and using machine learning techniques to identify whether a particular patient has a certain condition. However, recent advances in deep learning and NLP enable models to learn a rich representation of (medical) language. Convolutional neural networks (CNN) for text classification can augment the existing techniques by leveraging the representation of language to learn which phrases in a text are relevant for a given medical condition. In this work, we compare concept extraction based methods with CNNs and other commonly used models in NLP in ten phenotyping tasks using 1,610 discharge summaries from the MIMIC-III database. We show that CNNs outperform concept extraction based methods in almost all of the tasks, with an improvement in F1-score of up to 26 and up to 7 percentage points in area under the ROC curve (AUC). We additionally assess the interpretability of both approaches by presenting and evaluating methods that calculate and extract the most salient phrases for a prediction. The results indicate that CNNs are a valid alternative to existing approaches in patient phenotyping and cohort identification, and should be further investigated. Moreover, the deep learning approach presented in this paper can be used to assist clinicians during chart review or support the extraction of billing codes from text by identifying and highlighting relevant phrases for various medical conditions.
在电子健康记录的二次分析中，一项关键任务是正确识别所研究的患者队列。在许多情况下，用于准确分类医疗状况的最有价值和相关性最强的信息仅存在于临床叙述中。因此，有必要使用自然语言处理（NLP）技术来提取和评估这些叙述。解决这个问题最常用的方法是从文本中提取一些临床医生定义的医学概念，并使用机器学习技术来识别特定患者是否患有某种疾病。然而，深度学习和NLP的最新进展使模型能够学习（医学）语言的丰富表示。用于文本分类的卷积神经网络（CNN）可以通过利用语言表示来学习文本中哪些短语与给定的医疗状况相关，从而增强现有技术。在这项工作中，我们在十个表型分析任务中，使用来自MIMIC-III数据库的1610份出院小结，将基于概念提取的方法与CNN以及NLP中其他常用模型进行比较。我们表明，在几乎所有任务中，CNN的表现都优于基于概念提取的方法，F1分数提高了26，ROC曲线下面积（AUC）提高了7个百分点。我们还通过展示和评估计算和提取预测中最突出短语的方法，来评估这两种方法的可解释性。结果表明，在患者表型分析和队列识别中，CNN是现有方法的有效替代方案，应进一步研究。此外，本文提出的深度学习方法可用于在病历审查期间协助临床医生，或通过识别和突出显示各种医疗状况的相关短语来支持从文本中提取计费代码。
