# Enabling Granular Subgroup Level Model Evaluations by Generating Synthetic Medical Time Series

arXiv: https://arxiv.org/abs/2510.19728 · HTML: https://mlsanity.com/paper/2510.19728/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 16 references that are in the map.
- First posted on arXiv 22 Oct 2025.

## Builds on (most cited first, 7 of 7)

- [Diffusion-TS: Interpretable Diffusion for General Time Series Generation](https://mlsanity.com/paper/2403.01742/) (2024)
- [Generating Synthetic Mixed-type Longitudinal Electronic Health Records for Artificial Intelligent Applications](https://mlsanity.com/paper/2112.12047/) (2021)
- [Yet Another ICU Benchmark: A Flexible Multi-Center Framework for Clinical ML](https://mlsanity.com/paper/2306.05109/) (2023)
- [Reliable Generation of Privacy-preserving Synthetic Electronic Health Record Time Series via Diffusion Models](https://mlsanity.com/paper/2310.15290/) (2023)
- [Generative AI for Synthetic Data Across Multiple Medical Modalities: A Systematic Review of Recent Developments and Challenges](https://mlsanity.com/paper/2407.00116/) (2024)
- [Conditional Generation of Medical Time Series for Extrapolation to Underrepresented Populations](https://mlsanity.com/paper/2201.08186/) (2022)
- [A structured regression approach for evaluating model performance across intersectional subgroups](https://mlsanity.com/paper/2401.14893/) (2024)

## Abstract (arXiv)

We present a novel framework for leveraging synthetic ICU time-series data not only to train but also to rigorously and trustworthily evaluate predictive models, both at the population level and within fine-grained demographic subgroups. Building on prior diffusion and VAE-based generators (TimeDiff, HealthGen, TimeAutoDiff), we introduce textitEnhanced TimeAutoDiff, which augments the latent diffusion objective with distribution-alignment penalties. We extensively benchmark all models on MIMIC-III and eICU, on 24-hour mortality and binary length-of-stay tasks. Our results show that Enhanced TimeAutoDiff reduces the gap between real-on-synthetic and real-on-real evaluation (``TRTS gap'') by over 70%, achieving DeltaTRTS ≤ 0.014 AUROC, while preserving training utility (DeltaTSTR ≈ 0.01). Crucially, for 32 intersectional subgroups, large synthetic cohorts cut subgroup-level AUROC estimation error by up to 50% relative to small real test sets, and outperform them in 72--84% of subgroups. This work provides a practical, privacy-preserving roadmap for trustworthy, granular model evaluation in critical care, enabling robust and reliable performance analysis across diverse patient populations without exposing sensitive EHR data, contributing to the overall trustworthiness of Medical AI.