Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AI-Driven Secure Data Sharing: A Trustworthy and Privacy-Preserving Approach

Al Amin, Kamrul Hasan, Sharif Ullah, Liang Hong

TL;DR

This work tackles secure, privacy-preserving data sharing for medical imaging in cloud environments, where applying DNNs to encrypted data often harms performance. It proposes a learnable block-pixel encryption scheme combined with Vision Transformer (ViT) to securely scramble data with per-client keys while preserving discriminative features for classification. The approach achieves about 94% validation accuracy on encrypted MRI and histopathology datasets and demonstrates robustness against integrity and confidentiality attacks, outperforming traditional DNNs in encrypted settings and offering favorable compute efficiency. The framework enables practical, privacy-preserving cloud-based medical image analysis with strong security guarantees and potential for multimodal extension in the future.

Abstract

In the era of data-driven decision-making, ensuring the privacy and security of shared data is paramount across various domains. Applying existing deep neural networks (DNNs) to encrypted data is critical and often compromises performance, security, and computational overhead. To address these limitations, this research introduces a secure framework consisting of a learnable encryption method based on the block-pixel operation to encrypt the data and subsequently integrate it with the Vision Transformer (ViT). The proposed framework ensures data privacy and security by creating unique scrambling patterns per key, providing robust performance against adversarial attacks without compromising computational efficiency and data integrity. The framework was tested on sensitive medical datasets to validate its efficacy, proving its ability to handle highly confidential information securely. The suggested framework was validated with a 94\% success rate after extensive testing on real-world datasets, such as MRI brain tumors and histological scans of lung and colon cancers. Additionally, the framework was tested under diverse adversarial attempts against secure data sharing with optimum performance and demonstrated its effectiveness in various threat scenarios. These comprehensive analyses underscore its robustness, making it a trustworthy solution for secure data sharing in critical applications.

AI-Driven Secure Data Sharing: A Trustworthy and Privacy-Preserving Approach

TL;DR

This work tackles secure, privacy-preserving data sharing for medical imaging in cloud environments, where applying DNNs to encrypted data often harms performance. It proposes a learnable block-pixel encryption scheme combined with Vision Transformer (ViT) to securely scramble data with per-client keys while preserving discriminative features for classification. The approach achieves about 94% validation accuracy on encrypted MRI and histopathology datasets and demonstrates robustness against integrity and confidentiality attacks, outperforming traditional DNNs in encrypted settings and offering favorable compute efficiency. The framework enables practical, privacy-preserving cloud-based medical image analysis with strong security guarantees and potential for multimodal extension in the future.

Abstract

In the era of data-driven decision-making, ensuring the privacy and security of shared data is paramount across various domains. Applying existing deep neural networks (DNNs) to encrypted data is critical and often compromises performance, security, and computational overhead. To address these limitations, this research introduces a secure framework consisting of a learnable encryption method based on the block-pixel operation to encrypt the data and subsequently integrate it with the Vision Transformer (ViT). The proposed framework ensures data privacy and security by creating unique scrambling patterns per key, providing robust performance against adversarial attacks without compromising computational efficiency and data integrity. The framework was tested on sensitive medical datasets to validate its efficacy, proving its ability to handle highly confidential information securely. The suggested framework was validated with a 94\% success rate after extensive testing on real-world datasets, such as MRI brain tumors and histological scans of lung and colon cancers. Additionally, the framework was tested under diverse adversarial attempts against secure data sharing with optimum performance and demonstrated its effectiveness in various threat scenarios. These comprehensive analyses underscore its robustness, making it a trustworthy solution for secure data sharing in critical applications.
Paper Structure (20 sections, 6 equations, 4 figures, 2 tables, 1 algorithm)

This paper contains 20 sections, 6 equations, 4 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. Secure Data Sharing
  4. Privacy-Preserving Medical Image Classification
  5. Proposed Approach
  6. Learnable Encryption based on Block-Pixel Operation
  7. ViT Architecture
  8. Embedding Layer
  9. Transformer Encoder
  10. Algorithmic Representation of the Proposed Approach
  11. Robustness Evaluation Through Adversarial Attacks
  12. Adversarial Noise Injection Bit-Flip and Gaussian Noise Attacks
  13. Reconstruction Attacks: Leading Bit and Minimum Difference Attacks
  14. RESULTS AND DISCUSSION
  15. Experimental Setup
  16. ...and 5 more sections

Figures (4)

  • Figure 1: Overview of ViT integrated privacy-preserving secure medical data sharing and classification framework.
  • Figure 2: Original image and encrypted images generated using the learnable encryption method based on block-pixel operation with three different keys (K1, K2, K3).
  • Figure 3: Training and Validation Accuracy and Loss Curves for the proposed Framework.
  • Figure 4: Transformation Pipeline: Original, Encrypted, and Post-Attack Images for Confidentiality Evaluation.