A Feasibility Study of Geometric-Decomposition Coil Compression in MRI Radial Acquisitions

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Physics
• Biomedical Engineering

Background:

• Modern MRI systems utilize receiver arrays with numerous coil elements to boost signal-to-noise ratio (SNR) and parallel imaging capabilities.
• Increasing channel counts in MRI leads to longer reconstruction times and higher computational demands, particularly for iterative reconstruction methods.
• Coil compression techniques offer a solution by reducing the number of coil channels, thereby mitigating computational burdens.

Purpose of the Study:

• To investigate the application and efficacy of geometric-decomposition coil compression (GCC) for radial MRI acquisitions.
• To evaluate GCC's performance in comparison to principal component analysis (PCA) for coil compression in radial sampling patterns.
• To assess the impact of GCC on SNR and reconstruction accuracy in high-channel MRI systems.

Main Methods:

• Geometric-decomposition coil compression (GCC) was applied to radial sampling data, including both linear-angle and golden-angle patterns.
• GCC was performed spatially along fully sampled directions, followed by an alignment step to ensure virtual coil sensitivity smoothness.
• The method was validated using both numerical simulation and in vivo MRI data.

Main Results:

• GCC demonstrated superior performance compared to conventional principal component analysis (PCA) in radial acquisitions.
• The GCC algorithm achieved a higher signal-to-noise ratio (SNR) in the reconstructed images.
• Lower normalized root mean squared error (NRMSE) values were observed with GCC, indicating improved accuracy.

Conclusions:

• Geometric-decomposition coil compression (GCC) is an effective strategy for reducing computational complexity in high-channel radial MRI.
• GCC offers significant advantages in SNR and accuracy over PCA for radial MRI reconstruction.
• The findings support the use of GCC for efficient and high-quality image reconstruction in advanced MRI techniques.