Democratising Clinical AI through Dataset Condensation for Classical Clinical Models

TL;DR

This paper introduces a differentially private dataset condensation method that enables model-agnostic, privacy-preserving data sharing for clinical models, including non-differentiable ones like decision trees and Cox regression.

Contribution

It extends dataset condensation to non-differentiable models using zero-order optimization, facilitating privacy-preserving data sharing in healthcare.

Findings

Preserves model utility with differential privacy guarantees

Works effectively across classification and survival datasets

Enables safe, model-agnostic clinical data sharing

Abstract

Dataset condensation (DC) learns a compact synthetic dataset that enables models to match the performance of full-data training, prioritising utility over distributional fidelity. While typically explored for computational efficiency, DC also holds promise for healthcare data democratisation, especially when paired with differential privacy, allowing synthetic data to serve as a safe alternative to real records. However, existing DC methods rely on differentiable neural networks, limiting their compatibility with widely used clinical models such as decision trees and Cox regression. We address this gap using a differentially private, zero-order optimisation framework that extends DC to non-differentiable models using only function evaluations. Empirical results across six datasets, including both classification and survival tasks, show that the proposed method produces condensed datasets that preserve model utility while providing effective differential privacy guarantees - enabling model-agnostic data sharing for clinical prediction tasks without exposing sensitive patient information.