Motion-robust T 2 ∗ $$ {\mathrm{T}}_2^{\ast } $$ quantification from low-resolution gradient echo brain MRI with...

Area of Science:

• Medical Imaging
• Biophysics
• Computational Neuroscience

Background:

• Quantitative T2* mapping using gradient echo MRI is sensitive to subject motion.
• Motion-induced magnetic field inhomogeneities can lead to signal loss and inaccurate T2* quantification.
• Effective motion correction is essential for reliable T2* mapping.

Purpose of the Study:

• To extend the PHIMO (Physics-Informed Motion correction) method for improved T2* quantification.
• To enhance motion correction robustness for challenging motion patterns and varying magnetic field inhomogeneities.
• To evaluate the performance of the extended method, PHIMO+, in motion detection and image quality.

Main Methods:

• Extension of the learning-based PHIMO method to PHIMO+.
• Incorporation of acquisition knowledge into PHIMO+ to improve reconstruction.
• Comprehensive evaluation using simulated and real motion data, assessing motion detection accuracy and image quality.

Main Results:

• PHIMO+ demonstrated superior qualitative and quantitative performance compared to baseline learning-based methods.
• PHIMO+ achieved comparable results to a state-of-the-art conventional method that requires redundant data acquisition.
• The method showed high accuracy in motion detection and improved image quality.

Conclusions:

• PHIMO+ offers competitive motion correction for T2* quantification.
• The method significantly reduces acquisition time (over 40%) compared to existing state-of-the-art techniques.
• PHIMO+ is a promising tool for motion-robust T2* quantification in research and clinical settings.