State-of-the-art session key generation on priority-based adaptive neural machine (PANM) in telemedicine

Area of Science:

• Computer Science
• Artificial Intelligence
• Medical Informatics

Background:

• Telemedicine offers remote healthcare but requires secure data communication.
• Digitization in healthcare necessitates robust security protocols for patient data privacy.
• Existing methods for session key generation in telemedicine may face challenges in efficiency and security.

Purpose of the Study:

• To propose and validate a novel, state-of-the-art session key generation technique for telemedicine applications.
• To enhance the security and privacy of patient data transmitted through telemedicine platforms.
• To improve the efficiency of session key generation using soft computing and neural networks.

Main Methods:

• Implementation of priority-oriented neural machines with soft computing, specifically Artificial Neural Networks (ANN).
• Application of the Hebbian learning rule for synchronization between patient and doctor neural machines.
• Validation through statistical testing of various session key sizes and mathematical hardness assessments.

Main Results:

• The proposed technique significantly reduced session key generation time across different key sizes (e.g., 4.011 ms for 56 bits to 6.105 ms for 1024 bits).
• Hebbian learning facilitated faster synchronization between neural machines, requiring fewer iterations.
• The derived value-based function and partial validations confirmed the technique's effectiveness and security.

Conclusions:

• The proposed session key generation method is suitable for telemedicine, ensuring patient data privacy and security.
• The technique offers strong protection against various data attacks in public networks.
• Partial key transmission enhances security by preventing intruders from decoding bit patterns.