A chaos-based visual encryption mechanism for clinical EEG signals

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cryptography

Background:

• Clinical electroencephalography (EEG) signals require robust security measures.
• Existing encryption methods may lack sufficient unpredictability for sensitive biological data.
• Chaos theory offers a promising avenue for enhancing data randomization and security.

Purpose of the Study:

• To develop and evaluate a chaos-based visual encryption mechanism for clinical EEG signals.
• To enhance the unpredictability and security of EEG data transmission and storage.
• To demonstrate the effectiveness of chaotic sequences in EEG signal encryption.

Main Methods:

• Utilized a 1D chaotic scrambler for randomizing EEG signal values.
• Implemented a permutation scheme and chaotic address scanning order for encryption.
• Employed multiplication with a 1D chaotic signal to scramble input EEG data.
• Applied chaotic address scanning order encryption to randomized values.

Main Results:

• Complete signal recovery achieved with correct deciphering parameters.
• Achieved extremely low Percent Root-Mean-Square Difference (PRD) values: 4.33 x 10(-15)% for control and 4.11 x 10(-15)% for alcoholic EEG signals.
• Demonstrated signal unrecoverability with minute input parameter errors (e.g., 0.00000001% initial point error).

Conclusions:

• The developed chaos-based visual encryption mechanism provides a highly secure method for clinical EEG signals.
• The method offers strong data protection due to the inherent unpredictability of chaotic sequences.
• The sensitivity to initial parameters ensures data integrity and prevents unauthorized access.