3D phase unwrapping using global expected phase as a reference: application to MRI global shimming.

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging (MRI)
• Image Processing

Background:

• MRI phase data frequently exhibits phase wrapping, causing discontinuities.
• Existing phase unwrapping methods rely on local voxel information.
• Phase wrapping poses challenges in low-signal regions and air cavities.

Purpose of the Study:

• To introduce a novel, robust 3D MRI phase unwrapping strategy.
• To address limitations of traditional local-based phase unwrapping algorithms.
• To improve phase map accuracy in challenging anatomical areas.

Main Methods:

• Developed a 3D phase unwrapping strategy utilizing global phase distribution.
• Employed continuous trigonometric functions to create an expected phase map reference.
• Guided voxel-wise phase correction using the generated reference map.
• Estimated original phase by analyzing the derivative of the wrapped phase image.

Main Results:

• Simulations demonstrated the method's effectiveness in various phase-wrapped scenarios.
• In vivo application for MRI automatic global shimming showed promising results.
• The proposed method proved more reliable and robust than traditional algorithms.
• Successfully obtained accurate phase maps in low-signal regions and air cavities (e.g., abdomen, pelvis).

Conclusions:

• The novel 3D phase unwrapping method offers superior reliability and robustness.
• This technique effectively handles phase wrapping in challenging MRI data.
• The approach has significant implications for improving MRI image quality and applications like shimming.