Background: Data sharing has been a big challenge in biomedical informatics due to privacy concerns. Contextual embedding models have demonstrated very strong representative capability to describe medical concepts (and their context), and they have shown promise as an alternative way to support deep learning applications without the need to disclose original data. However, contextual embedding models acquired from individual hospitals cannot be directly combined because their embedding spaces are different and naive pooling renders combined embeddings useless.Objective: We present a novel approach to address these issues to promote sharing representation without sharing data. We can build a global model from representations learned from local private data without sacrificing privacy and synchronize information from multiple sources.
Methods:We propose a methodology that harmonizes different local contextual embeddings into a global model. We use Word2Vec to generate contextual embeddings from each source and Procrustes to fuse different vector models into one common space by using a list of corresponding pairs as anchor points. With harmonized embeddings, we performed prediction analysis.
Results:We used sequential medical events extracted from the Medical Information Mart for Intensive Care III database to evaluate the proposed methodology in predicting the next likely diagnosis of a new patient using either structured data or unstructured data. Under different experimental scenarios, we confirmed that the global model built from harmonized local models achieves more accurate prediction than local models and global model built from naive pooling.Conclusions: Such aggregation of local models using our unique harmonization can serve as the proxy for a global model, combining information from a wide range of institutions and information sources. It allows information unique to a certain hospital to become available to other sites, increasing the fluidity of information flow in health care.