Parallel imaging reconstruction using spatial nulling maps.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Imaging Reconstruction
• Parallel Imaging Techniques

Background:

• Parallel imaging accelerates MRI acquisition but requires robust reconstruction methods.
• Existing methods like ESPIRiT rely on empirical masking and are sensitive to subspace separation.
• There is a need for more robust and less empirically dependent parallel imaging reconstruction.

Purpose of the Study:

• To develop a robust parallel imaging reconstruction method using spatial nulling maps (SNMs).
• To combine concepts from PRUNO and ESPIRiT for improved reconstruction.
• To eliminate manual masking procedures in parallel imaging reconstruction.

Main Methods:

• Developed a hybrid-domain method combining PRUNO and ESPIRiT.
• Extracted null-subspace bases to calculate image-domain SNMs.
• Reconstructed multi-channel images using an image-domain nulling system formed by SNMs.

Main Results:

• The proposed method achieved reconstruction quality comparable to ESPIRiT with manual masking.
• The method eliminated the need for masking-related manual procedures.
• The reconstruction was tolerant of null- and signal-subspace division and allowed for spatial regularization.

Conclusions:

• An efficient hybrid-domain reconstruction method using multi-channel SNMs was developed.
• The method eliminates coil sensitivity masking and is insensitive to subspace separation.
• This presents a robust parallel imaging reconstruction procedure for practical use.