MedBlindTuner: Towards Privacy-preserving Fine-tuning on Biomedical Images with Transformers and Fully Homomorphic Encryption

TL;DR

MedBlindTuner introduces a privacy-preserving framework for training medical image models using fully homomorphic encryption and data-efficient transformers, ensuring data privacy without sacrificing accuracy.

Contribution

This work is the first to combine data-efficient image transformers with fully homomorphic encryption for privacy-preserving medical image analysis.

Findings

Achieves comparable accuracy to non-encrypted models

Enables training on encrypted medical images

Provides a secure solution for outsourcing ML in healthcare

Abstract

Advancements in machine learning (ML) have significantly revolutionized medical image analysis, prompting hospitals to rely on external ML services. However, the exchange of sensitive patient data, such as chest X-rays, poses inherent privacy risks when shared with third parties. Addressing this concern, we propose MedBlindTuner, a privacy-preserving framework leveraging fully homomorphic encryption (FHE) and a data-efficient image transformer (DEiT). MedBlindTuner enables the training of ML models exclusively on FHE-encrypted medical images. Our experimental evaluation demonstrates that MedBlindTuner achieves comparable accuracy to models trained on non-encrypted images, offering a secure solution for outsourcing ML computations while preserving patient data privacy. To the best of our knowledge, this is the first work that uses data-efficient image transformers and fully homomorphic encryption in this domain.