# CDMT-EHR: A Continuous-Time Diffusion Framework for Generating Mixed-Type Time-Series Electronic Health Records

arXiv: https://arxiv.org/abs/2603.23719 · HTML: https://mlsanity.com/paper/2603.23719/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 35 references that are in the map.
- First posted on arXiv 24 Mar 2026.

## Builds on (most cited first, 21 of 21)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [Simple and Effective Masked Diffusion Language Models](https://mlsanity.com/paper/2406.07524/) (2024)
- [Deep EHR: A Survey of Recent Advances in Deep Learning Techniques for Electronic Health Record (EHR) Analysis](https://mlsanity.com/paper/1706.03446/) (2017)
- [CoDi: Co-evolving Contrastive Diffusion Models for Mixed-type Tabular Synthesis](https://mlsanity.com/paper/2304.12654/) (2023)
- [TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation](https://mlsanity.com/paper/2410.20626/) (2024)
- [CorGAN: Correlation-Capturing Convolutional Generative Adversarial Networks for Generating Synthetic Healthcare Records](https://mlsanity.com/paper/2001.09346/) (2020)
- [EVA: Generating Longitudinal Electronic Health Records Using Conditional Variational Autoencoders](https://mlsanity.com/paper/2012.10020/) (2020)
- [Reliable Generation of Privacy-preserving Synthetic Electronic Health Record Time Series via Diffusion Models](https://mlsanity.com/paper/2310.15290/) (2023)
- [MedDiff: Generating Electronic Health Records using Accelerated Denoising Diffusion Model](https://mlsanity.com/paper/2302.04355/) (2023)
- [EHRDiff: Exploring Realistic EHR Synthesis with Diffusion Models](https://mlsanity.com/paper/2303.05656/) (2023)
- [Synthesizing Mixed-type Electronic Health Records using Diffusion Models](https://mlsanity.com/paper/2302.14679/) (2023)
- [Continuous Diffusion for Mixed-Type Tabular Data](https://mlsanity.com/paper/2312.10431/) (2023)
- [TimEHR: Image-based Time Series Generation for Electronic Health Records](https://mlsanity.com/paper/2402.06318/) (2024)
- [TarDiff: Target-Oriented Diffusion Guidance for Synthetic Electronic Health Record Time Series Generation](https://mlsanity.com/paper/2504.17613/) (2025)
- [Synthesizing Multimodal Electronic Health Records via Predictive Diffusion Models](https://mlsanity.com/paper/2406.13942/) (2024)
- [Guided Discrete Diffusion for Electronic Health Record Generation](https://mlsanity.com/paper/2404.12314/) (2024)

## Abstract (arXiv)

Electronic health records (EHRs) are invaluable for clinical research, yet privacy concerns severely restrict data sharing. Synthetic data generation offers a promising solution, but EHRs present unique challenges: they contain both numerical and categorical features that evolve over time. While diffusion models have demonstrated strong performance in EHR synthesis, existing approaches predominantly rely on discrete-time formulations, which suffer from finite-step approximation errors and coupled training-sampling step counts. We propose a continuous-time diffusion framework for generating mixed-type time-series EHRs with three contributions: (1) continuous-time diffusion with a bidirectional gated recurrent unit backbone for capturing temporal dependencies, (2) unified Gaussian diffusion via learnable continuous embeddings for categorical variables, enabling joint cross-feature modeling, and (3) a factorized learnable noise schedule that adapts to per-feature-per-timestep learning difficulties. Experiments on two large-scale intensive care unit datasets demonstrate that our method outperforms existing approaches in downstream task performance, distribution fidelity, and discriminability, while requiring only 50 sampling steps compared to 1,000 for baseline methods. Classifier-free guidance further enables effective conditional generation for class-imbalanced clinical scenarios.