Abstract
One broad goal of biomedical informatics is to generate fully-synthetic, faithfully representative electronic health records (EHRs) to facilitate data sharing between healthcare providers and researchers and promote methodological research. A variety of methods existing for generating synthetic EHRs, but they are not capable of generating unstructured text, like emergency department (ED) chief complaints, history of present illness, or progress notes. Here, we use the encoder–decoder model, a deep learning algorithm that features in many contemporary machine translation systems, to generate synthetic chief complaints from discrete variables in EHRs, like age group, gender, and discharge diagnosis. After being trained end-to-end on authentic records, the model can generate realistic chief complaint text that appears to preserve the epidemiological information encoded in the original record-sentence pairs. As a side effect of the model’s optimization goal, these synthetic chief complaints are also free of relatively uncommon abbreviation and misspellings, and they include none of the personally identifiable information (PII) that was in the training data, suggesting that this model may be used to support the de-identification of text in EHRs. When combined with algorithms like generative adversarial networks (GANs), our model could be used to generate fully-synthetic EHRs, allowing healthcare providers to share faithful representations of multimodal medical data without compromising patient privacy. This is an important advance that we hope will facilitate the development of machine-learning methods for clinical decision support, disease surveillance, and other data-hungry applications in biomedical informatics.
Highlights
The wide adoption of electronic health record (EHR) systems has led to the creation of large amounts of healthcare data
Because they contain personally identifiable patient information, much of which is protected under the Health Insurance Portability and Accountability Act (HIPAA), these data are often difficult for providers to share with investigators outside their organizations, limiting their feasibility for use in research
By training one neural network to generate fake records and another to discriminate those fakes from the real records, the model is able to learn the distribution of both count- and binary-valued variables in the EHRs, which can be used to produce patient-level records that preserve the analytic properties of the data without sacrificing patient privacy
Summary
The wide adoption of electronic health record (EHR) systems has led to the creation of large amounts of healthcare data. We explore the use of encoder–decoder models, a kind of deep learning algorithm, to generate natural-language text for EHRs, filling an existing gap and increasing the feasibility of using generative models like Choi et al.’s GAN to create highquality healthcare datasets for secondary uses Like their name implies, encoder–decoder models comprise 2 (conventionally recurrent) neural networks: one that encodes the input sequence as a single dense vector, and one that decodes this vector into the target sequence.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have