Securing Multimedia Using a Deep Learning Based Chaotic Logistic Map.

Area of Science:

• Computer Science
• Information Security
• Medical Informatics

Background:

• Telemedicine adoption surged during the pandemic, increasing the need for secure transmission of sensitive patient medical data.
• Existing security measures may not adequately protect the confidentiality and integrity of multimedia medical records.
• Cyber-attacks pose a significant threat to the privacy of digital health information.

Purpose of the Study:

• To propose a novel secure multimedia transformation approach for protecting medical data.
• To enhance the security and robustness of medical image and video data transmission.
• To develop a deep learning-based method for identifying and preventing the dissemination of fake medical multimedia data.

Main Methods:

• Integration of a lightweight encryption function utilizing a chaotic logistic map for confusion and diffusion.
• Application of the ResNet model for classifying fake medical multimedia data.
• Implementation of linear feedback shift register operations and an interactive user interface for ease of use.
• Utilization of Multilayer Perceptrons (MLP) for classifying medical data on the receiver side.

Main Results:

• The proposed approach demonstrated efficiency in securing medical data against various cyber-attacks.
• The encryption mechanism achieved high entropy levels, indicating robust data protection.
• The ResNet model effectively identified fake medical multimedia content.
• The chaotic map provided essential security properties like confusion and diffusion, enhancing encryption robustness.

Conclusions:

• The developed secure multimedia transformation approach effectively safeguards sensitive medical data in telemedicine.
• The integration of deep learning and chaotic maps offers a robust solution for medical data security and authenticity verification.
• The proposed method contributes to enhancing trust and security in digital health platforms.