Accelerated calibrationless parallel transmit mapping using joint transmit and receive low-rank tensor completion.

Area of Science:

• Magnetic Resonance Imaging
• Medical Physics
• Biomedical Engineering

Background:

• Parallel imaging techniques accelerate MRI acquisition by using multiple receiver coils.
• Transmit field mapping is crucial for parallel transmit MRI but traditionally requires long acquisition times.
• Calibrationless methods aim to reconstruct images without the need for explicit coil sensitivity maps.

Purpose of the Study:

• To evaluate a calibrationless parallel imaging algorithm for reconstructing undersampled parallel transmit field maps.
• To assess the performance of different low-rank tensor completion approaches for transmit field mapping.
• To determine the feasibility of high acceleration factors in transmit array mapping.

Main Methods:

• Three joint transmit and receive low-rank tensor completion algorithms were evaluated: virtual coils, joint-receiver coils, and transmit low rank.
• The methods were tested using synthetic data and in vivo brain and body measurements.
• Performance was assessed at various noise levels and acceleration rates on an 8-channel parallel transmit 7 Tesla system.

Main Results:

• The virtual coils method failed at acceleration rates > 2, while joint-receiver coils succeeded up to acceleration factor 4.
• The transmit low-rank method demonstrated robust performance, enabling up to eightfold acceleration with normalized RMS errors below 0.1.
• This indicates superior performance of the transmit low-rank approach in challenging conditions.

Conclusions:

• Calibrationless parallel imaging methods are feasible for high-factor undersampling in transmit array mapping.
• The transmit low-rank tensor completion algorithm significantly advances the potential for accelerated transmit field mapping.
• This research paves the way for faster and more efficient parallel transmit MRI protocols.