Reliable Generation of EHR Time Series via Diffusion Models

Electronic Health Records (EHRs) are rich sources of patient-level data, including laboratory tests, medications, and diagnoses, offering valuable resources for medical data analysis. However, concerns about privacy often restrict access to EHRs, hindering downstream analysis. Researchers have explored various methods for generating privacy-preserving EHR data. In this study, we introduce a new method for generating diverse and realistic synthetic EHR time series data using Denoising Diffusion Probabilistic Models (DDPM). We conducted experiments on six datasets, comparing our proposed method with eight existing methods. Our results demonstrate that our approach significantly outperforms all existing methods in terms of data utility while requiring less training effort. Our approach also enhances downstream medical data analysis by providing diverse and realistic synthetic EHR data. The Electronic Health Record (EHR) is a digital version of the patient's medical history maintained by healthcare providers. It includes information such as demographic attributes, vital signals, and lab measurements that are sensitive in nature and important for clinical research. Researchers have been utilizing statistical and machine learning (ML) methods to analyze EHR for a variety of downstream tasks such as disease diagnosis, in-hospital mortality prediction, and disease phenotyping (Shickel et al., 2018; Goldstein et al., 2017). However, due to privacy concerns, EHR data is strictly regulated, and thus the availability of EHR data is often limited, creating barriers to the development of computational models in the field of healthcare.