An Efficient and Privacy-Preserving Scheme for Disease Prediction in Modern Healthcare Systems

Area of Science:

• Health Informatics
• Cybersecurity
• Artificial Intelligence

Background:

• Mobile healthcare applications leveraging the Internet of Things (IoT) offer advanced online services.
• Disease Prediction Systems (DPS) are crucial for improving healthcare quality through faster, more accurate diagnoses.
• Protecting sensitive personal health information is a growing concern in digital health.

Purpose of the Study:

• To propose an efficient privacy-preserving (PP) scheme for patient healthcare data from IoT devices for disease prediction.
• To address challenges in multi-label disease prediction accuracy and data privacy.
• To enhance the security of patient data transfer and storage in the Health Care System (HCS).

Main Methods:

• Utilized Log of Round value-based Elliptic Curve Cryptography (LR-ECC) for secure data transfer post-authentication.
• Implemented an enhanced Herding Genetic Algorithm-based Deep Learning Neural Network (EHGA-DLNN) for disease prediction.
• Focused on handling multi-label instances to maintain prediction accuracy.

Main Results:

• The proposed PP scheme significantly enhances security during data transfer.
• EHGA-DLNN demonstrated improved disease prediction accuracy, especially for multi-label cases.
• Experimental results show superior privacy and security compared to existing methods.

Conclusions:

• The developed privacy-preserving scheme effectively secures patient data from IoT devices.
• The EHGA-DLNN model provides accurate and reliable disease predictions.
• This integrated approach offers a robust solution for secure and accurate disease prediction in modern healthcare.