Saved in:
Bibliographic Details
Main Authors: Ibrahim, Mahmoud, Elen, Bart, Sun, Chang, Ertaylan, Gökhan, Dumontier, Michel
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.19728
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909864123432960
author Ibrahim, Mahmoud
Elen, Bart
Sun, Chang
Ertaylan, Gökhan
Dumontier, Michel
author_facet Ibrahim, Mahmoud
Elen, Bart
Sun, Chang
Ertaylan, Gökhan
Dumontier, Michel
contents 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 \textit{Enhanced 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 $Δ_{TRTS} \leq 0.014$ AUROC, while preserving training utility ($Δ_{TSTR} \approx 0.01$). 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.
format Preprint
id arxiv_https___arxiv_org_abs_2510_19728
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enabling Granular Subgroup Level Model Evaluations by Generating Synthetic Medical Time Series
Ibrahim, Mahmoud
Elen, Bart
Sun, Chang
Ertaylan, Gökhan
Dumontier, Michel
Machine Learning
Artificial Intelligence
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 \textit{Enhanced 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 $Δ_{TRTS} \leq 0.014$ AUROC, while preserving training utility ($Δ_{TSTR} \approx 0.01$). 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.
title Enabling Granular Subgroup Level Model Evaluations by Generating Synthetic Medical Time Series
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.19728