Simultaneous auto-calibration and gradient delays estimation (SAGE) in non-Cartesian parallel MRI using low-rank...

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Physics
• Image Reconstruction

Background:

• Non-Cartesian MRI trajectories are sensitive to gradient timing delays, causing data inconsistencies and corrupted auto-calibration information.
• Existing methods struggle to correct these delays without compromising image quality or requiring additional calibration scans.

Purpose of the Study:

• To develop a method for correcting gradient timing delays in non-Cartesian MRI.
• To simultaneously recover corruption-free auto-calibration data for parallel imaging.
• To achieve these goals without requiring additional calibration scans.

Main Methods:

• A novel method, Simultaneous Auto-calibration and Gradient delays Estimation (SAGE), is proposed.
• SAGE leverages the inherent low-rank property of calibration matrices derived from multi-channel k-space data.
• Gradient delays are estimated by minimizing the rank of the calibration matrix, using the Gauss-Newton method for non-linear problem solving.

Main Results:

• SAGE accurately estimates gradient timing delays, even at low signal-to-noise ratios (SNR) of 5.
• The method effectively removes artifacts caused by gradient timing delays.
• Image quality is restored across various non-Cartesian trajectories, including center-out radial, projection reconstruction, and spiral.

Conclusions:

• The proposed low-rank based method simultaneously estimates gradient timing delays.
• SAGE provides accurate auto-calibration data, leading to improved image quality.
• This approach eliminates the need for additional calibration scans in non-Cartesian MRI.