On the noise correlation matrix for multiple radio frequency coils

Area of Science:

• Physics
• Statistical Physics
• Electromagnetism

Background:

• Receiver coil noise is a critical factor in Magnetic Resonance Imaging (MRI) sensitivity.
• Understanding noise correlation between coils is essential for improving signal-to-noise ratio (SNR).
• Previous theories on noise correlation presented conflicting viewpoints.

Purpose of the Study:

• To theoretically derive and experimentally verify noise correlation formulas for multiple receiver coils.
• To reconcile conflicting arguments regarding spectral and total noise correlation.
• To elucidate the relationship between impedance, inductance, and noise correlation using statistical physics principles.

Main Methods:

• Application of the general fluctuation-dissipation theorem to derive spectral noise correlation.
• Utilizing the canonical partition function to derive total current noise correlation.
• Employing the Kramers-Kronig relation to connect spectral and total noise characteristics.
• Experimental verification using two-coil arrays.

Main Results:

• A prototypic correlation formula for spectral noise, dependent on the real part of the inverse impedance matrix, was derived.
• A distinct formula for total current noise correlation, dependent on the inverse inductance matrix, was established.
• The theoretical derivations were experimentally validated, confirming the derived relationships.
• The study reconciles previous conflicting arguments by distinguishing between spectral and total noise correlation.

Conclusions:

• Noise correlation in multi-coil receiver systems can be accurately described using statistical physics.
• The derived formulas provide a unified framework for understanding spectral and total noise correlation.
• Experimental validation supports the theoretical models, offering practical implications for MRI coil design and performance optimization.