PSyGenTAB: A Privacy-Preserving Framework for Synthetic Clinical Tabular Data Generation via Constrained Optimization

2026-06-16

AI Summary

This work addresses the scarcity of high-quality clinical data that hinders medical AI development, as existing synthetic data methods struggle to simultaneously preserve privacy and clinical utility. The authors formulate the generation of clinical tabular data as a constrained optimization problem, explicitly embedding configurable privacy constraints directly into the generative model training for the first time. By employing an augmented Lagrangian method, the approach enables controllable trade-offs between privacy and utility. The resulting synthetic data effectively retains critical clinical feature correlations and rare diagnostic patterns, achieving performance on par with real data in downstream clinical tasks while substantially reducing record-level re-identification risk and demonstrating strong robustness against membership inference attacks.

Abstract

The development of medical AI is constrained by limited access to high-quality clinical data due to institutional silos and strict privacy regulations such as HIPAA and GDPR. Synthetic data generation offers a potential solution, but existing methods lack principled mechanisms to explicitly manage the privacy-utility trade-off, often degrading clinically meaningful patterns or risking patient re-identification. We present PSyGenTAB, a privacy-preserving generative framework that formulates synthetic healthcare data generation as a constrained optimization problem solved using the Augmented Lagrangian Method. By embedding configurable privacy constraints directly into model training, PSyGenTAB enforces minimum privacy thresholds while maximizing clinical data utility. Across multiple clinically motivated benchmarks, PSyGenTAB preserves inter-feature clinical relationships and minority-class diagnostic patterns essential for reliable health AI. Downstream evaluation using Train-on-Synthetic, Test-on-Real and Train-on-Real, Test-on-Synthetic protocols shows that models trained on synthetic data achieve performance comparable to those trained on real patient records. Privacy auditing further demonstrates reduced exact record reproduction and strong resilience to membership inference attacks. These results establish PSyGenTAB as a principled framework for balancing privacy protection and clinical utility in synthetic healthcare data, supporting secure cross-institutional AI development.

Problem

• synthetic clinical data
• privacy-utility trade-off
• patient re-identification
• clinical data utility
• privacy-preserving

Innovation

• constrained optimization
• privacy-preserving synthetic data
• clinical tabular data
• augmented Lagrangian method
• privacy-utility trade-off