A Mathematically Generated Noise Technique for Ultrasound Systems

Area of Science:

• Medical Imaging
• Cybersecurity
• Information Theory

Background:

• Ultrasound systems are vulnerable to cyberattacks, necessitating robust patient data protection.
• Current methods for generating random bits rely on hardware noise, which is difficult to extract efficiently in resource-limited systems.
• Software-based random bit generation offers greater accessibility for developers.

Purpose of the Study:

• To develop and evaluate a novel software-based algorithm for generating random bits for ultrasound systems.
• To compare the performance of the proposed algorithm against hardware-based methods, specifically using CPU frequency.
• To assess the security and quality of randomized ultrasound images generated using the new algorithm.

Main Methods:

• A mathematical noise function was employed to generate random bits from a raw bitmap image.
• Time-series analysis of generated random bits was performed, verifying statistical properties like mean, median, and mode.
• Ultrasound phantom images were randomized using exclusive mixing with the generated bits.
• Image quality was assessed using Peak Signal-to-Noise Ratio (PSNR) and Mean Square Error (MSE).

Main Results:

• The proposed algorithm achieved a min entropy estimate of 7.156616/8 bit, outperforming the conventional GetSystemTime method.
• Statistical analysis confirmed the unpredictability and stability of the generated random bits.
• Randomized ultrasound phantom images showed comparable or improved quality metrics compared to hardware noise methods.

Conclusions:

• The proposed software-based algorithm provides a superior and more accessible method for generating random bits in ultrasound systems.
• This approach enhances the security of patient information against cyber threats.
• The algorithm offers a practical solution for improving the stability and unpredictability of random bit generation in medical devices.