Fuzzy clustering of gradient-echo functional MRI in the human visual cortex. Part II: quantification

Area of Science:

• Neuroimaging
• Data Analysis
• Biophysics

Background:

• Functional MRI (fMRI) is crucial for brain activity analysis.
• Distinguishing signal contributions like blood oxygen level dependent (BOLD) and flow is challenging.
• Existing methods may not fully resolve complex signal dynamics in the visual cortex.

Purpose of the Study:

• To evaluate the performance of Fuzzy Cluster Analysis (FCA) for fMRI data.
• To compare FCA with traditional correlation analysis.
• To separate and quantify flow and BOLD signal contributions in the human visual cortex.

Main Methods:

• Utilized simulated and in vivo fMRI data from the human visual cortex.
• Applied Fuzzy Cluster Analysis (FCA) as implemented in the Evident software package.
• Performed quantitative comparisons with correlation analysis and analyzed fMRI model fits.

Main Results:

• FCA demonstrated robustness in separating functional brain activation from noise in fMRI data.
• Signal enhancement in the visual cortex varied based on vascularization and radiofrequency (RF) flip angle.
• BOLD contributions were observable in larger regions and showed a flip angle dependency, distinct from inflow enhancement.

Conclusions:

• Fuzzy clustering is a robust and efficient method for fMRI data analysis.
• FCA can separate functional brain activation from noise and other signal changes.
• The method allows quantification of flow and BOLD contributions, even in areas with complex vascularization.