Multi-Scattering

Multi-Scattering simulates light propagation through scattering and absorbing media to model photon transport for applications in biomedical optics and environmental sciences.

Key Features:

• Monte Carlo simulation: Employs the Monte Carlo method by tracking billions of photon packets through simulated cubic volumes to model light transport phenomena.
• GPU acceleration: Uses general-purpose computing on graphics processing units (GPUs) to accelerate computations, achieving up to 200× speedup per GPU and up to 800× when using four GPUs.
• High-throughput photon transport: Demonstrates computing the transport path of one billion photons in ~10 seconds for optical depth (OD) = 10 and ~20 minutes for OD = 500 for anisotropy factor g = 0.86.
• Lorenz–Mie theory integration: Incorporates Lorenz–Mie theory to generate scattering phase functions from spherical particles.
• Image formation modeling: Simulates image formation using virtual collecting lenses and detection matrices that mimic camera objectives and sensor arrays.
• Experimental validation: Includes validation through comparisons between simulated results and experimental data as reported in Part II of the publication series.

Scientific Applications:

• Biomedical optics: Predicts photon transport in tissues and blood samples for biomedical imaging and analysis.
• Environmental optics: Models light propagation through clouds, smoke, fog, turbid liquids, and spray systems.
• Imaging system evaluation: Assesses camera-based detection and image formation in scattering media.

Methodology:

Monte Carlo photon-packet tracking in cubic volumes, general-purpose GPU acceleration (including multi‑GPU use), Lorenz–Mie–generated scattering phase functions, and simulation of virtual collecting lenses and detection matrices; example parameters include anisotropy factor g = 0.86 and optical depths (OD) = 10 and 500.