Keyed Nonlinear Transform: Lightweight Privacy-Enhancing Feature Sharing for Medical Image Analysis

TL;DR

The paper introduces Keyed Nonlinear Transform (KNT), a lightweight, privacy-preserving feature transformation for medical image analysis that significantly reduces re-identification risk with minimal performance impact.

Contribution

KNT provides a practical, key-conditioned obfuscation method for feature sharing in split inference, enhancing privacy without retraining or substantial overhead.

Findings

KNT reduces re-identification AUC from 0.635 to 0.586.

KNT introduces only 0.15 ms CPU overhead.

KNT maintains classification accuracy within 1.0 percentage point.

Abstract

Feature sharing via split inference offers a lightweight alternative to federated learning for resource-constrained hospitals, but transmitted features still leak patient identity information and lack practical mechanisms for controlled feature sharing. We propose Keyed Nonlinear Transform (KNT), a drop-in feature transformation that applies key-conditioned obfuscation to intermediate representations. KNT reduces re-identification AUC from 0.635 to 0.586, corresponding to a 36% reduction in above-chance identity signal, while introducing only 0.15 ms CPU overhead, without backbone retraining, and preserving classification performance within 1.0 pp. Our analysis shows that KNT's nonlinear transform prevents closed-form inversion and shifts recovery to iterative gradient-based optimization under full key compromise, substantially increasing inversion difficulty. The same transform generalizes to dense prediction tasks, incurring only a 4.4 pp Dice reduction on skin-lesion segmentation without retraining. These results position KNT as a practical and efficient privacy layer for split inference deployments.