TimEHR: Image-based Time Series Generation for Electronic Health Records
This addresses the problem of generating realistic EHR data for researchers and practitioners, but it is incremental as it builds on existing GAN methods with a novel image-based approach.
The paper tackled the challenge of generating time series data from Electronic Health Records (EHRs) by proposing TimEHR, a GAN-based model that treats time series as images, and it outperformed state-of-the-art methods on fidelity, utility, and privacy metrics across three real-world datasets.
Time series in Electronic Health Records (EHRs) present unique challenges for generative models, such as irregular sampling, missing values, and high dimensionality. In this paper, we propose a novel generative adversarial network (GAN) model, TimEHR, to generate time series data from EHRs. In particular, TimEHR treats time series as images and is based on two conditional GANs. The first GAN generates missingness patterns, and the second GAN generates time series values based on the missingness pattern. Experimental results on three real-world EHR datasets show that TimEHR outperforms state-of-the-art methods in terms of fidelity, utility, and privacy metrics.