Abstract:Telemedicine and teleconsultations have become rampant today, consequent to the pandemic scenario as well as the availability of seamless internet connectivity. In this context, maintaining the integrity and confidentiality of a medical record has become more significant than ever before. A novel data-hiding methodology based upon Grain 128a algorithm and two-dimensional logistic mapping for the security of biomedical images in applications of telemedicine is proposed in this research article. In this endeavor, the name of the medical practitioner and diagnosis results are embedded into the MRI brain image of a patient with astrocytoma. A highlight of this work is the use of a multilevel system for enhanced security. Initially, Grain 128a is used for deriving the initial factors of the logistic mapping using a 512-bit confidential key. Grain 128a is able to withstand various types of single-key attacks. Logistic Mapping is used to generate two sequences, one for the X and the other for the Y direction. Grain 128a algorithm and logistic mapping are combined to enhance the security as well as the sensitivity of the data hiding system for a single variation in the secret key. A bit diffusion and confusion process is used to identify the pixel locations in which the name of the practitioner and the diagnosis result must be embedded. The doctor's name and diagnosis result are converted into binary and introduced into the least significant bit positions (LSBs) of those pixel locations. Finally, encryption is done on the data-hided image using Elliptical Curve Cryptography (ECC) for enhanced security. The system performance is analyzed using different measures such as histogram analysis, structural similarity index (SSIM), universal image quality index (UIQI), Pearson's correlation coefficient (PCC), entropy, peak signal-to-noise ratio (PSNR), etc. The novel system is compared with two existing techniques and observed that it has better performance in terms of PSNR and SSIM. The security of the encryption system is measured by means of histogram analysis, correlation coefficient, PSNR, entropy, etc. The performance of the encryption system has been compared with the current state of art technique and it is found that the proposed system outperforms the current state of art techniques in every performance measure.

Integrating Deep Learning and Homomorphic Encryption for Secure Image Transmission

Applying Deep Neural Networks over Homomorphic Encrypted Medical Data

Enhancing Security of Medical Images Using Deep Learning, Chaotic Map, and Hash Table

Using chaos to encrypt images with reconstruction through deep learning model for smart healthcare

A deep learning signed medical image based on cryptographic techniques

Application of Homomorphic Encryption in Medical Imaging

Enhancing medical image security through machine learning and dual watermarking-based technique

Exploring AI-driven Innovations in Image Communication Systems for Enhanced Medical Imaging Applications

A Deep Cryptographic Framework for Securing the Healthcare Network from Penetration

DeepEDN: A Deep-Learning-Based Image Encryption and Decryption Network for Internet of Medical Things

Securing medical image privacy in cloud using deep learning network

Secure Convolution Neural Network Inference Based on Homomorphic Encryption

Hybrid data encryption model integrating multi-objective adaptive genetic algorithm for secure medical data communication over cloud-based healthcare systems

Deep learning-based encryption for secure transmission digital images: A survey

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

A Novel Medical Image Encryption Scheme Based on Deep Learning Feature Encoding and Decoding

A Privacy Protection Framework for Medical Image Security without Key Dependency Based on Visual Cryptography and Trusted Computing

A secure data hiding system in biomedical images using grain 128a algorithm, logistic mapping and elliptical curve cryptography

Blockchain with deep learning-enabled secure healthcare data transmission and diagnostic model

Federated Learning with Homomorphic Encryption for Ensuring Privacy in Medical Data

Blockchain and homomorphic encryption based privacy-preserving model aggregation for medical images