Accelerated MR whole brain imaging with sheared voxel imaging using aliasing separation gradients

Area of Science:

• Medical Imaging
• Physics

Background:

• Accelerating Magnetic Resonance Imaging (MRI) data acquisition is crucial for reducing scan times and improving patient comfort.
• Undersampling phase encoding (PE) lines is a common strategy to achieve faster MRI, but it typically leads to aliasing artifacts.

Purpose of the Study:

• To propose a novel aliasing separation method for accelerating MRI data acquisition by undersampling PE lines.
• To achieve acceleration through modifications in voxel shape using additional encoding gradients.

Main Methods:

• A 3D imaging technique employing undersampling in both PE directions with aliasing separation gradients is introduced.
• Reconstruction involves a fast process using shift correction or gridding.
• The method's characteristics and acceleration capabilities were evaluated using phantom and in vivo experiments.
• Further acceleration was explored by integrating parallel acquisition techniques.

Main Results:

• The proposed technique results in an anisotropic resolution pattern due to voxel shearing, which scales with the acceleration factor.
• Analysis in both image and k-space domains, visualized with grid phantom imaging, confirmed this characteristic.
• In vivo experiments demonstrated 6x acceleration with the proposed method alone.
• Combined with parallel imaging, acceleration reached approximately 12x.

Conclusions:

• The application of separation gradients in both PE directions effectively separates aliasing artifacts in 3D MRI.
• High acceleration factors were achieved for whole-brain 3D MRI.
• A slight blurring effect due to voxel shearing was observed as a trade-off for the increased acceleration.