Analysis and correction of gradient nonlinearity bias in apparent diffusion coefficient measurements

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Diffusion Tensor Imaging (DTI)

Background:

• MRI gradient nonlinearity causes spatially dependent b-values, leading to significant (10-20%) non-uniformity errors in apparent diffusion coefficient (ADC) measurements.
• These errors are particularly problematic over clinically relevant field-of-views and affect quantitative diffusion analysis.

Purpose of the Study:

• To develop a practical correction procedure for MRI systems that effectively reduces apparent diffusion coefficient (ADC) bias.
• The procedure aims to minimize ADC errors caused by gradient nonlinearity with minimal measurements for any anisotropic medium.

Main Methods:

• A bias analysis was performed considering spatial and time-domain cross-terms for diffusion and imaging gradients.
• The proposed correction involves rotating the gradient nonlinearity tensor into the diffusion gradient frame, approximating spatial bias using the b-matrix's Euclidean norm.
• Correction efficiency was numerically evaluated across various model diffusion tensor anisotropies and orientations.

Main Results:

• The developed correction terms accounted for 75-95% of the ADC bias for fractional anisotropy (FA) values between 0.3-0.9.
• Residual non-uniformity errors in ADC were found to increase with anisotropic diffusion.
• The approximation avoids the need for full diffusion tensor measurement and diagonalization for corrected ADC derivation.

Conclusions:

• A simplified correction algorithm using three orthogonal diffusion measurements is sufficient to control ADC non-uniformity errors in clinical MRI studies.
• Optimal ADC bias reduction for anisotropic media is achieved using non-laboratory-based diffusion gradients.