Radiance based method for accurate determination of volume scattering parameters using GPU-accelerated Monte Carlo

Area of Science:

• Optics and Photonics
• Biophysics
• Computer Graphics

Background:

• Volume scattering models are vital across disciplines like biology and lighting.
• Parameter estimation typically uses inverse methods fitting simulations to experimental data.
• Scattered intensity distributions are commonly used but can yield poor estimates for samples with long optical path lengths.

Purpose of the Study:

• To introduce and evaluate an inverse procedure for estimating volume scattering parameters using scattered radiance distributions.
• To compare the efficacy of radiance versus intensity distributions for parameter estimation across varying optical path lengths.
• To leverage graphics processing units (GPUs) for efficient simulation and analysis.

Main Methods:

• Developed an inverse procedure fitting simulations to scattered radiance distributions.
• Utilized modern graphics processing units (GPUs) for computationally efficient simulations.
• Conducted an in-depth simulation study comparing radiance and intensity distribution fitting.

Main Results:

• For moderate optical path lengths, intensity distributions provide sufficient information for accurate parameter estimation.
• For longer optical path lengths, intensity distributions become insufficient, leading to poor parameter estimates.
• The proposed radiance distribution-based inverse method demonstrates superior accuracy for samples with longer optical path lengths.

Conclusions:

• Radiance distribution-based inverse methods are better suited for accurate volume scattering parameter estimation in samples with longer optical path lengths.
• The choice of data distribution (intensity vs. radiance) is critical for accurate parameter estimation, especially with increasing optical path length.
• Efficient GPU implementation enables detailed studies on the impact of data distribution on scattering parameter estimation.