A prospective approach to correct for inter-image head rotation in fMRI

Area of Science:

• Neuroimaging
• Biophysics
• Medical Physics

Background:

• Head motion during functional MRI (fMRI) can compromise the accuracy of brain activation mapping.
• Physiological signals detected in fMRI are susceptible to corruption by even small head movements.
• Accurate interpretation of fMRI results relies on minimizing motion-induced artifacts.

Purpose of the Study:

• To develop and validate a real-time prospective correction method for head rotation in fMRI.
• To assess the efficacy of the proposed method in improving the quality of fMRI data.
• To demonstrate the feasibility of dynamic motion correction in vivo.

Main Methods:

• Utilized orbital navigator echoes for dynamic detection and correction of rotational motion.
• Approximated head motion as a rigid body, decomposing it into linear and rotational components.
• Implemented prospective correction for both through-plane and in-plane inter-image head rotation.
• Validated the method using phantom experiments and in vivo imaging studies in volunteers.
• Employed a modified receiver operating characteristic method to evaluate motion-corrected fMRI data.

Main Results:

• The developed method achieved dynamic detection and correction of rotational motion in under 100 ms.
• Phantom experiments confirmed accurate correction of rotational motion across a range of +/-0.36 to +/-12 degrees.
• In vivo studies demonstrated the feasibility of real-time prospective rotational motion correction.
• Motion-corrected fMRI sensorimotor studies showed superior performance compared to uncorrected data.

Conclusions:

• Real-time prospective correction of head rotation using orbital navigator echoes is feasible and effective in fMRI.
• This technique significantly improves the accuracy and reliability of fMRI activation maps.
• The method offers a valuable tool for enhancing the quality of neuroimaging studies by mitigating motion artifacts.