Privacy-Aware Adversarial Defense Approach for Medical Image Classification in Distributed Healthcare Systems

A Deep Learning-Enabled Privacy-Preserving Framework for Secure Medical Image Storage in Cloud Systems Using Advanced Convolutional Architectures

With the rapid adoption of cloud computing for healthcare data storage, ensuring the privacy and security of sensitive medical images has become a critical challenge. Recent advancements in deep learning, particularly convolutional neural networks (CNNs), offer new possibilities for enhancing encryption and data protection without compromising image quality or diagnostic value. This research aims to develop a state-of-the-art, CNN-based privacy-preserving framework that leverages feature extraction and adaptive encryption methods to securely manage medical imaging data in cloud environments, aligning with emerging trends in AI-driven healthcare security and regulatory compliance. With the increasing reliance on cloud-based platforms for storing and sharing medical imaging data, privacy and security concerns have become paramount. Medical images such as CT scans, MRI, and X-rays contain sensitive patient information that must be protected from unauthorized access while ensuring usability for clinical diagnosis. This research proposes a Deep Learning-Enabled Privacy-Preserving Framework that integrates advanced convolutional neural network (CNN) architectures with adaptive encryption techniques to enhance the confidentiality of medical images stored in cloud environments. The framework extracts essential features from images while applying encryption algorithms that maintain data integrity and diagnostic value. Experimental results on diverse medical image datasets demonstrate the efficiency, robustness, and scalability of the proposed approach, making it suitable for modern healthcare applications where secure, cloud-based data management is essential. The proposed method aligns with current trends in artificial intelligence, cybersecurity, and regulatory standards for medical data protection.

Seeing Through Threats: (ADEx) Adversarial Detection through Explainability.

Deep Neural Networks (DNNs) remain vulnerable to adversarial perturbations, raising significant concerns in image processing applications, particularly in high-stakes domains such as medical imaging and security-critical systems. Most existing defense strategies are limited by domain specificity, architectural dependence, or the need for extensive retraining, making them impractical for real-world deployment. In this work, we propose ADEx, the first framework to integrate low-rank image approximation with explainability-driven analysis for the detection of adversarial samples. ADEx works by extracting a low-rank representation of the input image using Singular Value Thresholding (SVT), and identifying important image regions by computing class-specific gradient maps from the final layers of the classifier. These maps are then compared using Rank-Biased Overlap (RBO) to quantify the degree of attention drift induced by adversarial perturbations. ADEx is designed for adversarial detection in image classification systems, where class-specific gradient-based explanations are well defined. The framework operates without retraining or architectural modification and can be applied to a wide range of differentiable classifiers, provided gradient access is available for explanation generation. Extensive experiments across multiple datasets, architectures, and attack types demonstrate consistent performance, robustness to hyperparameter choices, and low sensitivity to calibration size. The method provides an interpretable and lightweight solution suitable for practical deployment.

Privacy-Preserving and Quantum-Resistant Federated Learning Framework for Fog Healthcare Systems

Quantum computing is on the horizon and will destroy existing cryptography standards, putting digital healthcare system security and patient privacy at danger. For very private and secure communication in fog healthcare settings, this article presents a new Quantum-Resistant Federated Deep Learning (QR-FDL) Framework. The QR-FDL architecture incorporates a lattice-based technique (like Kyber) into the model aggregation stage of Federated Learning (FL), which is a kind of Post-Quantum Cryptography (PQC). We provide a solution that uses Differential Privacy (DP) to prevent attacks based on gradients in data reconstruction and guarantees quantum-era security for weights of models with gradients sent between the central server and decentralized fog nodes. The practical use of QR-FDL is shown by its rapid convergence rate and excellent classification accuracy of 91.5% in an empirical assessment including a medical imaging job, such as tumor classification. Critically, we measure the cryptographic overhead and demonstrate that, even if PQC causes a regulated latency rise the total FL round time is still tolerable for real-time fog settings. In order to implement quantum-secure along with privacy-preserving Deep Learning in mission-critical healthcare communications, this study presents the first proven, end-to-end solution. In addition to showing that federated medical facilities can be securely encrypted from end to end, this architecture uses differential privacy during model training to protect patients’ personal information. Through the research, we demonstrate its practical implementation in the near future for it is hard to balance the precision of modeled obstructions with cryptographic overhead that still remains acceptable. The results of this research lay the groundwork for medical AI systems of the future that are both secure and respectful of patients’ privacy.

A Comprehensive Review on Adversarial Attacks and Detection Techniques in Deep Learning Models for Image Analysis

The research methodology involved a systematic literature review using the Scopus database, adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, and focusing on recent advancements in attack and defence techniques.

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

Federated learning (FL) is emerging as a promising machine learning technique in the medical field for analyzing medical images, as it is considered an effective method to safeguard sensitive patient data and comply with privacy regulations. However, recent studies have revealed that the default settings of FL may inadvertently expose private training data to privacy attacks. Thus, the intensity of such privacy risks and potential mitigation strategies in the medical domain remain unclear. In this paper, we make three original contributions to privacy risk analysis and mitigation in FL for medical data. First, we propose a holistic framework, MedPFL, for analyzing privacy risks in processing medical data in the FL environment and developing effective mitigation strategies for protecting privacy. Second, through our empirical analysis, we demonstrate the higher privacy risks in FL to process medical images, where adversaries can accurately reconstruct private medical images by performing privacy attacks. Third, we illustrate that the prevalent defense mechanism of adding random noise may not always be effective in protecting medical images against privacy attacks in FL, which poses unique and pressing challenges related to protecting the privacy of medical data. Furthermore, the paper discusses several unique research questions related to the privacy protection of medical data in the FL environment. We conduct extensive experiments on several benchmark medical image datasets to analyze and mitigate the privacy risks associated with FL for medical data.

Privacy-preserving methodologies against privacy attacks on deep learning: a survey