Abstract:In the fast-evolving field of medical image analysis, Deep Learning (DL)-based methods have achieved tremendous success. However, these methods require plaintext data for training and inference stages, raising privacy concerns, especially in the sensitive area of medical data. To tackle these concerns, this paper proposes a novel framework that uses surrogate images for analysis, eliminating the need for plaintext images. This approach is called Frequency-domain Exchange Style Fusion (FESF). The framework includes two main components: Image Hidden Module (IHM) and Image Quality Enhancement Module~(IQEM). The~IHM performs in the frequency domain, blending the features of plaintext medical images into host medical images, and then combines this with IQEM to improve and create surrogate images effectively. During the diagnostic model training process, only surrogate images are used, enabling anonymous analysis without any plaintext data during both training and inference stages. Extensive evaluations demonstrate that our framework effectively preserves the privacy of medical images and maintains diagnostic accuracy of DL models at a relatively high level, proving its effectiveness across various datasets and DL-based models.

What problem does this paper attempt to address?

This paper attempts to address the issue of how to effectively diagnose and analyze medical images using deep learning (DL) technology while protecting patient privacy. Specifically, existing deep learning methods require plaintext data for training and inference, which raises serious privacy concerns when dealing with sensitive medical data. To tackle this challenge, the paper proposes a new framework called Frequency-domain Exchange Style Fusion (FESF), which generates proxy images to replace plaintext images, allowing model training and inference without exposing the original medical images. ### Main Issues and Solutions: 1. **Privacy Protection**: Existing deep learning methods require plaintext medical images for training and inference, which may lead to patient privacy breaches. The paper embeds the information of plaintext images into host images to generate proxy images, thus avoiding the direct use of plaintext images. 2. **Maintaining Diagnostic Accuracy**: Although the generated proxy images are used for training and inference, it is essential to ensure that these images maintain sufficient diagnostic accuracy. The paper introduces an Image Hiding Module (IHM) and an Image Quality Enhancement Module (IQEM) to ensure that the generated proxy images protect privacy while having high diagnostic value. ### Specific Methods: - **Image Hiding Module (IHM)**: Operates in the frequency domain to embed the features of plaintext images into host images, generating composite images. - **Image Quality Enhancement Module (IQEM)**: Further optimizes the quality of composite images to make them closer to host images, improving the practicality and diagnostic accuracy of proxy images. ### Experimental Validation: The paper conducts extensive experiments on multiple medical image datasets, including tasks such as brain tumor detection, cervical cancer, retinal disease, and skin cancer. The experimental results show that the FESF framework can protect privacy while maintaining high diagnostic accuracy. ### Conclusion: The FESF framework proposed in the paper effectively addresses the privacy protection issue in medical image analysis while maintaining the diagnostic performance of deep learning models. This method provides a new solution for data sharing and privacy protection in the medical field.

Plaintext-Free Deep Learning for Privacy-Preserving Medical Image Analysis via Frequency Information Embedding

In-depth Analysis of Privacy Threats in Federated Learning for Medical Data

Privacy Risks Analysis and Mitigation in Federated Learning for Medical Images

Privacy enhancing and generalizable deep learning with synthetic data for mediastinal neoplasm diagnosis

A separable privacy-preserving technique based on reversible medical data hiding in plaintext encrypted images using neural network

Privacy-Preserving Deep Learning With Learnable Image Encryption on Medical Images

FedDP: Privacy-preserving method based on federated learning for histopathology image segmentation

PRIMIS: Privacy-preserving medical image sharing via deep sparsifying transform learning with obfuscation

Privacy-Preserving Medical Image Classification through Deep Learning and Matrix Decomposition

Privacy Preserving Federated Learning in Medical Imaging with Uncertainty Estimation

Vision Through the Veil: Differential Privacy in Federated Learning for Medical Image Classification

Distributed Federated Learning-Based Deep Learning Model for Privacy MRI Brain Tumor Detection

Bridging the Gap: Differentially Private Equivariant Deep Learning for Medical Image Analysis

Medical imaging deep learning with differential privacy

A Fog-Based Privacy-Preserving Federated Learning System for Smart Healthcare Applications

Towards Privacy-preserving Explanations in Medical Image Analysis

End-to-end privacy preserving deep learning on multi-institutional medical imaging

Improving privacy-preserving multi-faceted long short-term memory for accurate evaluation of encrypted time-series MRI images in heart disease

Adversarial Medical Image with Hierarchical Feature Hiding

On Differentially Private 3D Medical Image Synthesis with Controllable Latent Diffusion Models

Towards Privacy-Preserving Medical Imaging: Federated Learning with Differential Privacy and Secure Aggregation Using a Modified ResNet Architecture