Octopus: A Novel Approach for Health Data Masking and Retrieving Using Physical Unclonable Functions and Machine...

Area of Science:

• Health Informatics
• Cybersecurity
• Machine Learning

Background:

• Smart health equipment and mobile applications facilitate remote health monitoring via the Internet of Medical Things (IoMT).
• The interconnected nature of IoMT presents significant security and confidentiality challenges due to accessibility and unpredictability.

Purpose of the Study:

• To enhance privacy and security for healthcare devices within IoMT systems.
• To develop a robust method for masking sensitive health data and mitigating security breaches.

Main Methods:

• Implementation of Octopus and Physically Unclonable Functions (PUFs) for data masking and privacy preservation.
• Application of machine learning (ML) techniques for secure data retrieval and network security enhancement.

Main Results:

• The proposed technique successfully masks health data, providing privacy to healthcare devices.
• Machine learning algorithms demonstrated high efficacy in retrieving masked data and reducing security breaches.
• The integrated approach achieved an accuracy rate of 99.45% in securing health data.

Conclusions:

• The combination of Octopus, PUFs, and ML offers a highly accurate and effective solution for securing IoMT data.
• This technique significantly enhances the confidentiality and integrity of sensitive health information in remote monitoring systems.