AUTHOR=Li Xiaowu , Peng Huiling 

TITLE=Chaotic medical image encryption method using attention mechanism fusion ResNet model

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.1226154

DOI=10.3389/fnins.2023.1226154

ISSN=1662-453X

ABSTRACT=With the rapid advancement of artificial intelligence (AI) technology, ensuring the privacy and security of patient medical images has emerged as a critical concern in current research on image privacy protection. However, traditional methods for encrypting medical images have been criticized for their limited flexibility and inadequate security. In light of this, the present study proposes a novel chaotic medical image encryption method that leverages the attention mechanism fused with the ResNet model (referred to as AT-ResNet-CM) to overcome these limitations.Initially, the ResNet model is employed as the underlying network to construct the encryption and decryption framework. The ResNet's residual structure and jump connections are utilized to effectively extract profound information from medical images and accelerate the convergence of the model. Subsequently, the output of the ResNet model is encrypted using a logistic chaotic system, which introduces randomness and complexity to the encryption process. This enhances the security of the resulting encrypted network. Additionally, an attention mechanism is introduced to augment the model's response to the region of interest within the medical image, further strengthening the security of the encrypted network.Experimental simulations and analyses have been conducted to evaluate the performance of the proposed approach. The results demonstrate that the proposed method surpasses alternative models in terms of encryption effectiveness, with a horizontal correlation coefficient of 0.0021 and information entropy of 0.9887. Furthermore, the incorporation of the attention mechanism significantly improves the encryption performance of the model, reducing the horizontal correlation coefficient to 0.0010 and increasing the information entropy to 0.9965. These findings validate the efficacy of the proposed method for medical image encryption tasks, as it offers enhanced security and flexibility compared to existing approaches.In conclusion, the proposed chaotic medical image encryption method, AT-ResNet-CM, presents a promising solution to address the 1 Li et al.limitations of traditional encryption techniques in protecting patient medical images. By leveraging the attention mechanism fused with the ResNet model, the method achieves improved security and flexibility. The experimental results substantiate the superiority of the proposed method in terms of encryption effectiveness, horizontal correlation coefficient, and information entropy.