Computational Approach for Securing Radiology-Diagnostic Data in Connected Health Network using High-Performance...

Area of Science:

• Medical Imaging and Informatics
• Cybersecurity in Healthcare
• Health Information Technology

Background:

• Diagnostic radiology is crucial for modern medicine, enhanced by information technology for data management.
• Connected health networks improve collaboration but face challenges with high computational costs and data privacy breaches.
• Existing cryptographic methods struggle to encrypt both medical images and textual data securely and efficiently.

Purpose of the Study:

• To propose a secured radiology-diagnostic data framework for connected health networks.
• To utilize high-performance GPU-accelerated Advanced Encryption Standard (AES) for robust data encryption.
• To address the challenges of encrypting diverse medical data formats (images and text) at a lower computational cost.

Main Methods:

• Developed a framework integrating GPU-accelerated AES for encrypting diagnostic data.
• Evaluated the framework using brain MRI and CT datasets from multiple institutions.
• Tested the framework's ability to encrypt and decrypt textual data in common formats like Microsoft Word, Excel, and PDF.

Main Results:

• The framework successfully encrypted and decrypted both medical image datasets (MRI, CT) and textual medical reports.
• Encryption and decryption were achieved at a lower computational cost using standard hardware and software.
• Data quality and security levels were maintained throughout the encryption-decryption process.

Conclusions:

• The proposed framework offers a secure and efficient solution for protecting radiology diagnostic data in connected health networks.
• GPU-accelerated AES provides a viable method for encrypting diverse medical data formats without compromising quality or security.
• This approach enhances data privacy and security in healthcare while maintaining cost-effectiveness and performance.