Aliasing artifact reduction in spiral real-time MRI

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging

Background:

• Spiral real-time MRI is prone to aliasing artifacts originating from signals outside the field of view (FOV).
• These artifacts are particularly problematic in midsagittal views during speech production imaging.

Purpose of the Study:

• To identify the cause of spiral aliasing artifacts in real-time MRI.
• To propose and evaluate two novel methods for mitigating these artifacts: large FOV (LF) and estimation-subtraction (ES).

Main Methods:

• Simulations were conducted to pinpoint the artifact's origin.
• The LF method involves reconstructing with a larger FOV to encompass artifact sources.
• The ES method estimates and subtracts artifact signals from multicoil k-space data.

Main Results:

• Artifact origin identified as gradient nonlinearity and imperfect spiral anti-aliasing.
• Both LF and ES methods significantly reduced artifacts, improving Likert scores by 1.25 and 1.35 levels, respectively.
• Reconstruction times: baseline (160.69 ms), LF (526.43 ms), ES (171.47 ms).

Conclusions:

• Both LF and ES methods successfully mitigate spiral aliasing artifacts.
• The estimation-subtraction (ES) method offers superior artifact reduction and is more time-efficient than the large FOV (LF) approach.