Optical Transmission in Single-Layer Brain Tissues under Different Optical Source Types: Modelling and Simulation

Xi Yang, Chengpeng Chai, Yun-Hsuan Chen, Mohamad SawanBioengineering 2024 · DOI

TL;DR

This paper investigates how different types of light sources penetrate and scatter within individual layers of brain tissue. By simulating light propagation using the Monte Carlo method, the researchers provide crucial insights into optimizing optical source selection for brain imaging techniques like photoacoustic imaging, laying the groundwork for more complex multi-layer brain models.

What Problem Does This Paper Address?

Accurately imaging the human brain using optical methods, such as photoacoustic imaging (PAI), is challenging due to the complex, multi-layered structure of the brain and the way light interacts with different tissues. While previous research has often modeled the brain as multiple distinct layers, there's a need to understand how light behaves within each individual tissue type when illuminated by various light sources. This is important because different brain tissues (like scalp, skull, gray matter, etc.) have unique optical properties (absorption, scattering) that affect light penetration and diffusion. Without this granular understanding, it's difficult to select the most effective optical source for a given imaging task or to build more realistic multi-layer models. This paper aims to fill this gap by analyzing the optical transmission characteristics of six individual brain tissue types under sixteen different optical source configurations.

Methods

The researchers employed a Monte Carlo (MC) simulation approach, a standard and robust method for modeling light transport in scattering and absorbing media. They utilized the 'mcxlab' MATLAB toolbox for their simulations.

Key aspects of their methodology include:

  • Simplified Single-Layer Models: Instead of complex multi-layer models, they created six distinct, homogeneous single-layer models, each representing a specific brain tissue: scalp, skull, cerebrospinal fluid (CSF), gray matter, white matter, and blood vessel. These models were cubic volumes (14x14x14 mm³).
  • Optical Properties: The simulations used established optical properties (absorption coefficient, scattering coefficient, anisotropy factor, refractive index) for these tissues at an 800 nm wavelength, chosen for its relatively low absorption by water and hemoglobin, allowing for deeper penetration.
  • Sixteen Optical Source Types: They investigated the impact of sixteen different optical source types, including common ones like pencil and isotropic beams, as well as more complex ones like Gaussian distributions, planar, disk, ring, and Fourier-based sources. These sources were launched from a central point and propagated vertically.
  • Data Analysis:
    • Optical Energy Distribution: The normalized optical energy was analyzed to understand how light intensity varied across the tissue.
  • Propagating Depth Evaluation: The maximum depth reached by the light at specific low energy levels (1%, 0.1%, 0.01% of maximum) was determined to assess penetration.
  • Optical Field Width Evaluation: The Full Width at Half Maximum (FWHM) of the optical energy distribution was calculated in both horizontal (X and Y) directions at a specific depth (12 mm) to evaluate light diffusion and, by extension, imaging resolution.

Datasets

The study did not use external datasets for training or evaluation. Instead, it relied on:

  • Established Optical Properties: The optical properties of the six brain tissues at 800 nm were taken from existing literature (cited as reference [15]).
  • Simulated Data: The core "dataset" consists of the simulated optical energy distributions, propagation depths, and FWHM values generated by the Monte Carlo simulations for each tissue type and light source combination.

Key Insights and Ideas

  • Tissue-Specific Light Behavior: The study clearly demonstrates that different brain tissues exhibit distinct responses to optical illumination. For instance, CSF shows less attenuation and diffusion compared to denser tissues like the scalp and skull, allowing light to penetrate further.
  • Source-Dependent Distribution: The choice of optical source significantly impacts how light energy is distributed within a tissue. Some sources are highly focused (pencil beam), while others spread out more (e.g., hyperboloid Gaussian).
  • CSF's Unique Characteristics: Cerebrospinal fluid (CSF) stands out with "obvious characteristics" in its optical energy distribution, likely due to its lower scattering and absorption coefficients. This suggests it might be a more accessible layer for optical imaging.
  • White Matter's Distinctive Source Preference: While most tissues showed similar optimal sources for maximum depth, white matter exhibited a different set of preferred sources, highlighting its unique optical properties.
  • FWHM as a Resolution Indicator: The FWHM analysis reveals how much light spreads out, directly correlating with the potential resolution of an optical imaging technique. Sources that produce smaller FWHMs are generally better for higher-resolution imaging.

Key Results

  • Penetration Depth: Collimated Gaussian, planar, and disk sources consistently ranked among the top three for deepest penetration across most tissues and energy levels. White matter, however, showed a preference for collimated Gaussian, hyperboloid Gaussian, and planar sources.
  • Field Width (FWHM):
    • The hyperboloid Gaussian source generally resulted in the largest FWHM (widest spread) in the X-axis direction across all tissues, indicating significant diffusion.
  • The pencil beam was highly focused, showing minimal divergence.
  • The spatial frequency Fourier source was noted for having minimal FWHM in some tissues, suggesting it might offer better resolution in specific scenarios.
  • Tissue Differences: The study quantified differences in penetration depth and FWHM across the six tissues, providing concrete data on how each tissue layer affects optical signals. For example, the maximum FWHM at 12mm depth was 9.1 mm in scalp (hyperboloid Gaussian), while the minimum was 6.8 mm in blood vessel (spatial frequency Fourier).

Limitations and Future Work

  • Homogeneous Models: The primary limitation is the use of single-layer, homogeneous tissue models. Real brain tissues are heterogeneous and layered, meaning these simulations represent a simplification. The authors acknowledge that multi-layer models are more complex but necessary for capturing real-world scenarios.
  • Single Wavelength: The simulations were conducted at a single wavelength (800 nm). Optical properties vary with wavelength, so a broader spectral analysis could provide more comprehensive insights.
  • Future Work: The authors suggest that future research should explore the "other characteristics of the optical sources in optical intensity and distribution" to further improve illumination strategies. They also propose that these single-layer studies provide a foundation for constructing more complex, multi-layer brain models, which are crucial for clinical applications and disease diagnosis.

Summary updated

Abstract

The human brain is a complex organ controlling daily activity. Present technique models have mostly focused on multi-layer brain tissues, which lack understanding of the propagation characteristics of various single brain tissues. To better understand the influence of different optical source types on individual brain tissues, we constructed single-layer brain models and simulated optical propagation using the Monte Carlo method. Based on the optical simulation results, sixteen optical source types had different optical energy distributions, and the distribution in cerebrospinal fluid had obvious characteristics. Five brain tissues (scalp, skull, cerebrospinal fluid, gray matter, and blood vessel) had the same set of the first three optical source types with maximum depth, while white matter had a different set of the first three optical source types with maximum depth. Each brain tissue had different optical source types with the maximum and minimum full width at half maximum. The study on single-layer brain tissues under different optical source types lays the foundation for constructing complex brain models with multiple tissue layers. It provides a theoretical reference for optimizing the selection of optical source devices for brain imaging.
BibTeX
@inproceedings{yang2024optical,
  author = {Xi Yang and Chengpeng Chai and Yun-Hsuan Chen and Mohamad Sawan},
  title = {{Optical Transmission in Single-Layer Brain Tissues under Different Optical Source Types: Modelling and Simulation}},
  booktitle = {Bioengineering},
  year = {2024},
  doi = {10.3390/bioengineering11090916},
}
Figure