Medical Concept Embedding with Time-Aware Attention
This work addresses the need for better healthcare analytics by improving medical concept embeddings for researchers and practitioners, though it is incremental as it builds on existing embedding methods with a novel temporal adaptation.
The paper tackled the problem of embedding medical concepts from Electronic Medical Records by incorporating temporal information, which previous models missed, and demonstrated that their time-aware attention model outperformed five state-of-the-art baselines in clustering and nearest neighbor search tasks.
Embeddings of medical concepts such as medication, procedure and diagnosis codes in Electronic Medical Records (EMRs) are central to healthcare analytics. Previous work on medical concept embedding takes medical concepts and EMRs as words and documents respectively. Nevertheless, such models miss out the temporal nature of EMR data. On the one hand, two consecutive medical concepts do not indicate they are temporally close, but the correlations between them can be revealed by the time gap. On the other hand, the temporal scopes of medical concepts often vary greatly (e.g., \textit{common cold} and \textit{diabetes}). In this paper, we propose to incorporate the temporal information to embed medical codes. Based on the Continuous Bag-of-Words model, we employ the attention mechanism to learn a "soft" time-aware context window for each medical concept. Experiments on public and proprietary datasets through clustering and nearest neighbour search tasks demonstrate the effectiveness of our model, showing that it outperforms five state-of-the-art baselines.