Obstetric Diagnosis Assistant via Knowledge Powered Attention and Information-Enhanced Strategy

The obstetric Electronic Medical Records (EMRs) contain a large amount of medical data and health information. The obstetric EMRs play a vital role in improving the quality of the diagnosis assistant service. In this paper, we treat the diagnosis assistant as a multi-label classification task and propose a Knowledge powered Attention and Information-Enhanced (KAIE) model for the obstetric diagnosis assistant. In order to make most of the information in EMRs, we propose to utilize the numerical information and chief complaint information to enhance the text representation. In addition to the use of information in EMRs, we integrate external knowledge from the COKG medical knowledge graph into the model. Specifically, we propose a multi-way attention mechanism for the generation of knowledge-aware representations based on text representations. Experiment results show that our approach is able to bring about +1.37 F1 score improvements upon the strong BERT baseline in the diagnosis assistant task.