Estimating Fiber Orientation Distribution Functions in 3D-Polarized Light Imaging

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Biophysics

Background:

• Human brain connectome research necessitates multiscale imaging approaches.
• Integrating data across different scales and modalities is crucial for comprehensive brain mapping.
• Current methods face challenges in bridging microscopic and macroscopic brain data.

Purpose of the Study:

• To establish a scalable concept for integrating microscopic 3D-Polarized Light Imaging (3D-PLI) data with macroscopic Diffusion Magnetic Resonance Imaging (dMRI).
• To enable data fusion across spatial scales for large-scale human brain connectome analysis.
• To validate a novel method for bridging micro- to macro-scale brain data.

Main Methods:

• Developed orientation distribution functions (pliODFs) from high-resolution 3D-PLI vector data using spherical harmonics expansion.
• Utilized high-performance computing and supercomputing for data processing.
• Validated the approach using simulated 3D-PLI data with defined fiber patterns and real human and seal brain tissue data.

Main Results:

• Successfully bridged spatial scales from microscopic 3D-PLI to meso- or macroscopic dimensions.
• Demonstrated the feasibility of data fusion between 3D-PLI derived pliODFs and dMRI data for large-scale datasets.
• Validated the method's effectiveness with both simulated and real biological data.

Conclusions:

• The established concept effectively integrates micro- and macro-scale neuroimaging data.
• This approach facilitates comprehensive human brain connectome research by enabling multiscale data fusion.
• The method holds significant potential for advancing our understanding of brain structure and connectivity.