Saved in:
Bibliographic Details
Main Authors: Nafis, Nazia, Esnaola, Inaki, Martinez-Perez, Alvaro, Villa-Uriol, Maria-Cruz, Osmani, Venet
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.18544
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911684606558208
author Nafis, Nazia
Esnaola, Inaki
Martinez-Perez, Alvaro
Villa-Uriol, Maria-Cruz
Osmani, Venet
author_facet Nafis, Nazia
Esnaola, Inaki
Martinez-Perez, Alvaro
Villa-Uriol, Maria-Cruz
Osmani, Venet
contents Generating synthetic tabular health data is challenging, and evaluating their quality is equally, if not more, complex. This systematic review highlights the critical importance of rigorous evaluation of synthetic health data to ensure reliability, clinical relevance, and appropriate use. From an initial identification of 2067 relevant papers published in the last ten years, 134 studies were selected for detailed analysis. Our review identifies key challenges, including lack of consensus on evaluation methods, inconsistent application of evaluation metrics, limited involvement of domain experts, inadequate reporting of dataset characteristics, and limited reproducibility of results. In response, we provide a structured consolidation of synthetic data generation and evaluation methods into taxonomies, alongside practical guidelines to support more robust and standardised evaluation practices. These findings aim to support the responsible development and use of synthetic health data, aligned with emerging expectations around transparency, reproducibility, and governance, ultimately enabling the community to fully harness its transformative potential and accelerate innovation.
format Preprint
id arxiv_https___arxiv_org_abs_2504_18544
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Critical Challenges and Guidelines in Evaluating Synthetic Tabular Data: A Systematic Review
Nafis, Nazia
Esnaola, Inaki
Martinez-Perez, Alvaro
Villa-Uriol, Maria-Cruz
Osmani, Venet
Machine Learning
Artificial Intelligence
Computers and Society
Generating synthetic tabular health data is challenging, and evaluating their quality is equally, if not more, complex. This systematic review highlights the critical importance of rigorous evaluation of synthetic health data to ensure reliability, clinical relevance, and appropriate use. From an initial identification of 2067 relevant papers published in the last ten years, 134 studies were selected for detailed analysis. Our review identifies key challenges, including lack of consensus on evaluation methods, inconsistent application of evaluation metrics, limited involvement of domain experts, inadequate reporting of dataset characteristics, and limited reproducibility of results. In response, we provide a structured consolidation of synthetic data generation and evaluation methods into taxonomies, alongside practical guidelines to support more robust and standardised evaluation practices. These findings aim to support the responsible development and use of synthetic health data, aligned with emerging expectations around transparency, reproducibility, and governance, ultimately enabling the community to fully harness its transformative potential and accelerate innovation.
title Critical Challenges and Guidelines in Evaluating Synthetic Tabular Data: A Systematic Review
topic Machine Learning
Artificial Intelligence
Computers and Society
url https://arxiv.org/abs/2504.18544