Forced detection Monte Carlo algorithms for accelerated blood vessel image simulations.

Area of Science:

• Medical Imaging
• Computational Physics
• Biomedical Optics

Background:

• Monte Carlo (MC) simulations are crucial for modeling light transport in biological tissues.
• Simulating light interactions in complex media like tissue-embedded objects can be computationally intensive.
• Variance reduction techniques are needed to improve the efficiency of MC simulations.

Purpose of the Study:

• To introduce two novel forced detection (FD) variance reduction Monte Carlo algorithms.
• To accelerate image simulations of tissue-embedded objects, particularly those with matched refractive indices.
• To evaluate the performance of these FD algorithms compared to traditional methods.

Main Methods:

• Developed two FD Monte Carlo algorithms that force a fraction of photon weight to the detector at each scattering event.
• The fractional weight is determined by the probability of reaching the detector without further interaction.
• Applied algorithms to a tissue model with blood vessels using two imaging setups.

Main Results:

• FD algorithms achieved identical results to brute force simulations for the tested tissue model.
• Simulations were accelerated by two orders of magnitude using the FD methods.
• The algorithms demonstrated effectiveness for objects embedded in tissue with matched refractive indices.

Conclusions:

• The presented FD algorithms offer a significant acceleration for MC image simulations in biomedical optics.
• These methods provide a powerful tool for efficiently simulating light transport in complex biological tissues.
• Future work will explore extending these techniques to scenarios with refractive index mismatches.