Correction of Physiological Artifacts in Multi-Echo fMRI Data-Evaluation of Possible RETROICOR Implementations

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging
• Physiological Signal Processing

Background:

• Physiological noise (e.g., cardiac and respiratory variations) significantly impacts functional magnetic resonance imaging (fMRI) data quality.
• Multi-echo (ME) fMRI offers potential benefits but requires robust artifact mitigation strategies.
• Retrospective correction techniques are crucial for enhancing the reliability of fMRI studies.

Purpose of the Study:

• To evaluate the efficacy of Retrospective Image Correction (RETROICOR) in reducing physiological artifacts in multi-echo fMRI data.
• To compare two RETROICOR implementations: correction on individual echoes (RTC_ind) versus composite multi-echo data (RTC_comp).
• To assess the impact of varying acquisition parameters, including multiband acceleration and flip angles, on RETROICOR performance.

Main Methods:

• Acquired multi-echo fMRI data from 50 healthy participants on a 3T Siemens Prisma scanner.
• Applied two RETROICOR correction methods (RTC_ind and RTC_comp) to the fMRI data.
• Analyzed data quality using metrics such as temporal signal-to-noise ratio (tSNR), signal fluctuation sensitivity (SFS), and variance of residuals across different acquisition parameters (multiband factors 2, 4, 6, 8; flip angles 30°, 45°, 60°).

Main Results:

• Both RETROICOR implementations significantly improved data quality (tSNR, SFS, residuals) across tested parameters.
• Optimal performance was observed with moderate multiband acceleration (factors 4 and 6) and lower flip angles (45°).
• Differences between RTC_ind and RTC_comp were minimal, indicating comparable efficacy.
• While RETROICOR improved data quality even at higher acceleration (multiband factor 8), overall quality was reduced compared to lower acceleration factors.

Conclusions:

• RETROICOR is an effective method for mitigating physiological artifacts in multi-echo fMRI.
• Both RTC_ind and RTC_comp are viable for practical applications in fMRI data processing.
• Optimizing acquisition parameters, such as flip angle and acceleration, in conjunction with RETROICOR is essential for maximizing fMRI data quality and reliability.