Patient representation learning and interpretable evaluation using clinical notes
Patient representation learning and interpretable evaluation using clinical notes
复制标题
DOI：
10.1016/j.jbi.2018.06.016
复制
发表时间：
2018-08-01
影响因子：
4.5
通讯作者：
Daelemans, Walter
中科院分区：
文献类型：
--
作者：
Sushil, Madhumita;Suster, Simon;Daelemans, Walter
我们在这项工作中有三个贡献：1。我们探索了堆栈去噪自动编码器和段落向量模型的实用性，以直接从临床笔记中学习与任务无关的密集患者表示。为了分析这些表示是否可以在任务之间转移，我们在多个监督设置中对其进行评估，以预测患者死亡率，主要诊断和手术类别以及性别。我们比较他们的表现与稀疏表示从一个词袋模型。我们观察到，学习的广义表示显着优于稀疏表示时，我们有几个积极的情况下学习，并没有强大的词汇功能。2.我们比较了从一袋词构造的特征集的模型性能，从医学概念获得的。在后一种情况下，概念代表问题、治疗和测试。我们发现，概念识别并没有提高分类性能。3.我们提出了新的技术，以促进模型的可解释性。为了理解和解释这些表示，我们探索了从自动编码器模型获得的患者表示中的最佳编码特征。此外，当我们使用这些预训练的表示作为监督输入时，我们计算两个网络的特征敏感度，以识别不同分类任务的最重要输入特征。我们成功地提取了最有影响力的功能，使用这种技术的管道。
We have three contributions in this work: 1. We explore the utility of a stacked denoising autoencoder and a paragraph vector model to learn task-independent dense patient representations directly from clinical notes. To analyze if these representations are transferable across tasks, we evaluate them in multiple supervised setups to predict patient mortality, primary diagnostic and procedural category, and gender. We compare their performance with sparse representations obtained from a bag-of-words model. We observe that the learned generalized representations significantly outperform the sparse representations when we have few positive instances to learn from, and there is an absence of strong lexical features. 2. We compare the model performance of the feature set constructed from a bag of words to that obtained from medical concepts. In the latter case, concepts represent problems, treatments, and tests. We find that concept identification does not improve the classification performance. 3. We propose novel techniques to facilitate model interpretability. To understand and interpret the representations, we explore the best encoded features within the patient representations obtained from the autoencoder model. Further, we calculate feature sensitivity across two networks to identify the most significant input features for different classification tasks when we use these pretrained representations as the supervised input. We successfully extract the most influential features for the pipeline using this technique.