Efficient and accurate EAP imaging from multi-shell dMRI with micro-structure adaptive convolution kernels and dual...

Area of Science:

• Neuroimaging
• Diffusion MRI
• Computational neuroscience

Background:

• Advanced computational techniques are used to image the Ensemble Average Propagator (EAP) in brain white matter.
• Methods like MAP-MRI and MAPL describe the low-frequency spectrum of EAP and compute scalar indices (RTOP, RTPP, RTAP).
• Current non-parametric methods require large-scale optimization problems for EAP representation.

Purpose of the Study:

• To propose a novel semi-parametric approach for more efficient and accurate EAP imaging.
• To develop a method that reduces computational complexity compared to existing techniques.
• To enable rapid and precise estimation of white matter microstructural properties.

Main Methods:

• Approximating EAP using spherical convolution of a Micro-Structure adaptive Gaussian kernel and a non-parametric orientation histogram.
• Employing dual Fourier domain Integral Transforms for analytical computation of scalar indices.
• Developing the MiSFIT (Microstructure-informed fitting) approach for semi-parametric EAP analysis.

Main Results:

• MiSFIT significantly reduces optimization complexity by fitting only 2-3 kernel parameters.
• The method achieves high time efficiency, processing multi-shell diffusion MRI data in approximately one minute.
• MiSFIT provides accurate estimates of RTOP, RTPP, and RTAP, comparable to MAPL across various white matter configurations.

Conclusions:

• MiSFIT offers a computationally efficient and accurate alternative for white matter EAP imaging.
• The semi-parametric approach effectively captures low-frequency EAP responses for robust scalar index estimation.
• This method advances the analysis of white matter microstructure using diffusion MRI data.