Simultaneous magnetic resonance diffusion and pseudo-diffusion tensor imaging.

Area of Science:

• Diffusion magnetic resonance imaging
• Medical imaging physics

Background:

• Intravoxel incoherent motion (IVIM) models water diffusion and blood microcirculation.
• The combined IVIM diffusion tensor imaging (IVIM-DTI) model accounts for anisotropic pseudo-diffusion and anisotropic diffusion.

Purpose of the Study:

• To propose a robust IVIM-DTI approach for simultaneous diffusion and pseudo-diffusion tensor imaging.
• To evaluate the performance of one-step and two-step estimation methods for the IVIM-DTI model.

Main Methods:

• Applied conventional one-step and two-step IVIM estimation methods to the IVIM-DTI model.
• Introduced an improved one-step method using a damped Gauss-Newton algorithm and a Gaussian prior.
• Tested method sensitivity to parameter initializations using in silico and in vivo data.

Main Results:

• The one-step damped Gauss-Newton method with a Gaussian prior demonstrated superior performance.
• This improved method showed less sensitivity to noise and initial parameter choices.
• More accurate estimates of IVIM-DTI parameters were achieved compared to other methods.

Conclusions:

• One-step estimation with damped Gauss-Newton and Gaussian prior is robust for IVIM-DTI.
• This method enables simultaneous diffusion and pseudo-diffusion tensor imaging.
• The findings support the clinical application of this advanced IVIM-DTI technique.