Physiological noise effects on the flip angle selection in BOLD fMRI.

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging
• Physiological Noise Analysis

Background:

• Blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) is sensitive to noise, particularly from physiological sources.
• The choice of imaging flip angle critically influences the temporal signal-to-noise ratio (TSNR) in fMRI.
• Understanding the interplay between signal-to-noise ratio (SNR), physiological noise, and flip angle is crucial for optimizing BOLD fMRI acquisition.

Purpose of the Study:

• To investigate the feasibility of using imaging flip angles significantly below the Ernst angle in BOLD fMRI.
• To determine if reduced flip angles impact the detection of neuronal activation when physiological noise is dominant.
• To provide a method for calculating an optimal, lower flip angle under specific noise and SNR conditions.

Main Methods:

• Theoretical modeling of TSNR dependence on SNR and flip angle under varying noise conditions.
• Experimental validation using BOLD fMRI data.
• Comparison of imaging results at low flip angles versus angles near the Ernst angle using a visual-motor task in eight subjects.

Main Results:

• A nonlinear relationship between TSNR and SNR was observed when physiological noise dominates.
• Reducing flip angles well below the Ernst angle resulted in negligible TSNR loss under high SNR and dominant physiological noise.
• No significant differences in hemodynamic response, contrast-to-noise ratio, effect size, or statistical activation maps were found when imaging at 9° compared to 75°.

Conclusions:

• Low flip angles are a viable option in BOLD fMRI when physiological noise is the primary noise source and SNR is high.
• Using lower flip angles can offer benefits including reduced RF power, limited inflow effects, decreased motion artifacts, lower physiological noise, and improved tissue contrast.
• This study provides a practical approach to optimizing flip angle selection for enhanced BOLD fMRI efficiency and data quality.