AI-based, secure and privacy-preserving synthetic data generation platform in transfusion-dependent β-thalassemia applied to the Webthal® dataset

This is a Cohort Study study (also classified as: Validation Study).

November 3, 2025BloodOpen Access

View Full Paper

Ask AI

Bookmark

Share

Key Points

To implement a secure platform for generating synthetic clinical data in transfusion-dependent β-thalassemia.
Developed an AI-based synthetic data generation platform using generative models for multimodal data.
Trained a CT-WGAN on data from a cohort of 779 transfusion-dependent β-thalassemia patients.
Evaluated synthetic data quality using the SAFE framework for statistical fidelity and privacy preservability.
Synthetic data was consistent with original cohorts, showing comparable mortality rates and statistical outcomes.
Achieved high fidelity metrics (Clinical Synthetic Fidelity up to 0.91) and maintained privacy across datasets.
Demonstrated significant improvements in data statistical power through augmentation.

We only have this paper's title and publication details.

AI-based platform generates high-quality synthetic data in transfusion-dependent β-thalassemia, suggesting potential for precision medicine research.

No takes yet. Share an insight, caveat, or question.

Gianesin et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Share