Rapid motion estimation and correction using self-encoded FID navigators in 3D radial MRI

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging (MRI)

Background:

• Head motion during 3D radial MRI degrades image quality.
• Accurate motion compensation is crucial for robust neuroanatomical imaging.
• Existing methods may not adequately address continuous or large head movements.

Purpose of the Study:

• To develop a self-navigated motion compensation strategy for 3D radial MRI.
• To compensate for continuous head motion using self-encoded FID navigators.
• To measure rigid body motion parameters with high temporal resolution.

Main Methods:

• A forward model simulated motion effects on self-encoded FID navigator signals.
• 3D gradient-echo MRI with a Kooshball trajectory was used in volunteers.
• Rigid body motion parameters were estimated using least-squares fitting and compared to external tracking.

Main Results:

• Self-encoded FID navigators achieved high accuracy (mean absolute errors of ~0.7 mm and ~0.8°).
• Retrospective correction significantly improved image quality for abrupt and continuous motion.
• Image quality after correction was comparable to results using external tracking.

Conclusions:

• Self-encoded FID navigators provide rapid, accurate motion parameters for 3D radial MRI.
• This self-navigated approach enables continuous correction of head movements.
• The method is suitable for robust neuroanatomical imaging in subjects with significant motion.