Revisiting equivalent optical properties for cerebrospinal fluid to improve diffusion-based modeling accuracy in the...

Area of Science:

• Biomedical Optics
• Neuroimaging Techniques
• Computational Modeling

Background:

• Diffusion approximation (DA) is widely used in functional near-infrared spectroscopy (fNIRS) despite known limitations, particularly concerning cerebrospinal fluid (CSF) optical properties.
• Existing studies often use empirical CSF optical properties not optimized for minimizing DA modeling errors, potentially leading to inaccuracies.

Purpose of the Study:

• To directly quantify the accuracy of DA solutions in brain models by comparing them with gold-standard mesh-based Monte Carlo (MMC) solutions.
• To derive updated recommendations for CSF optical properties in DA simulations to minimize modeling errors.

Main Methods:

• DA solutions were generated for five-layer head and Colin27 atlas models, with independent sweeping of CSF absorption and reduced scattering coefficients.
• Errors in surface fluence, total brain sensitivity, and brain energy deposition were computed by comparing DA solutions to MMC solutions using literature CSF optical properties.
• Optimized CSF optical properties were identified by minimizing these errors.

Main Results:

• Previously recommended CSF properties caused significant errors (8.7%–52%) in multiple tested metrics.
• Infinite solutions were found to match DA and MMC results for single metrics by sweeping both CSF absorption and scattering coefficients.
• Simultaneous minimization of multiple metrics across various source/detector separations led to updated recommendations for CSF optical properties.

Conclusions:

• Updated CSF optical properties recommendations reduce model mismatches between DA and MMC solutions without compromising computational speed.
• It is possible to eliminate mismatches for specific metrics of interest by adjusting CSF optical properties in DA simulations.