article · IEEE Access
This research presents a new image encryption algorithm designed to protect colour medical images, particularly for cloud storage. The process involves three main stages: initially, Fibonacci Q-matrices manipulate image data. Next, a unique Substitution box (S-box) transformation, developed in the Galois field (2⁸), is applied to each RGB channel. Finally, an encryption key generated from a hyperchaotic system, modelling a memristive coupled neural network, is used. The algorithm demonstrates high security, resistance to occlusion and noise attacks, and an extensive key space of 2⁶¹¹². It also achieves a high encryption rate of 16.65 Mbps through parallel processing, encrypting 16 images in approximately 0.1 seconds.
As medical images are increasingly stored in cloud systems, ensuring their security and patient privacy is critical. This research offers a robust encryption method to protect sensitive medical data from unauthorised access and breaches, which is vital for maintaining trust and compliance in healthcare.
This encryption algorithm is designed for integration into cloud storage systems to secure sensitive medical images, such as CT scans. It could be used by healthcare providers or cloud service providers to protect patient data from breaches and unauthorised access. The reported performance metrics suggest it is an applied research solution with potential for real-world deployment in secure medical imaging workflows.
AI-generated from the published abstract. Always read the original work before citing.
This work introduces a novel image encryption algorithm specifically designed for the protection of color medical images, particularly crucial in the era of cloud storage, where security and privacy are paramount. The initial phase of the encryption process leverages a substantial array of Fibonacci Q-matrices to manipulate image data intricately. Following this, a unique Substitution box (S-box) transformation, developed in the Galois field (28), is applied to each of the RGB channels, further enhancing the security layers. The final phase employs an encryption key generated from a hyperchaotic system of differential equations that models a memristive coupled neural network, offering a high degree of unpredictability and resistance to attacks. Performance metrics show high security [Peak Signal-to-Noise Ratio (PSNR) of 8.21 dB, Mean Absolute Error (MAE) of 81.72, and ≈ 0 Pixel Cross-Correlation (PCC)], resistivity to occlusion and noise attacks, an enormous key space of 26112, and with savvy parallel processing techniques, a high encryption rate of 16.65 Mbps. The integration of this encryption algorithm into cloud storage systems is of significant importance, as it ensures secure and confidential handling of sensitive medical images, addressing the growing concerns about data breaches and unauthorized access in the healthcare sector. Compared to previous color image encryption techniques, this work provides a large key space of 26112 and a small encryption time per image of around 0.1 second for encrypting 16 images at once. Thus, this approach is suitable for medical imaging techniques, such as Computed Tomography (CT) scans that generate multiple images of the tissues for the diagnosis of the patient.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1109/access.2024.3433499
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
