Temporal SNR optimization through RF coil combination in fMRI: The more, the better?

Area of Science:

• Magnetic Resonance Imaging
• Neuroimaging
• Biophysics

Background:

• Functional MRI (fMRI) reconstruction often uses channel-by-channel processing with covariance Sum-of-Squares (covSoS) for multi-channel RF coils.
• While covSoS offers near-optimal Signal-to-Noise Ratio (SNR), its impact on temporal SNR (tSNR), crucial for fMRI data quality, is less understood.

Purpose of the Study:

• To investigate tSNR optimality in voxel-wise RF coil combination methods for fMRI.
• To evaluate the effects of tSNR-optimized coil combinations on Blood-Oxygen-Level-Dependent (BOLD) sensitivity.

Main Methods:

• Developed an analytical solution for optimal RF coil combination based on the extended Krueger-Glover model, incorporating thermal and physiological noise.
• Experimentally compared the proposed method against covSoS in four volunteers at 7 Tesla (7T).

Main Results:

• The novel tSNR-optimized method demonstrated significant improvements in tSNR compared to covSoS.
• Surprisingly, the enhanced tSNR did not translate to higher BOLD sensitivity.
• Further exploration of t-score based methods yielded similar observations after statistical correction.

Conclusions:

• RF coil combinations optimized solely for tSNR in fMRI may not enhance BOLD sensitivity.
• Caution is advised against purely tSNR-driven, data-dependent RF coil combinations without unbiased assumptions.
• Future research may need to consider both tSNR and BOLD sensitivity in RF coil combination strategies.