Fundamental understanding of complex effects of wave propagation in turbid media requires accurate computational techniques to take into account the effects of multiple scattering of light. We present an open-source Monte Carlo algorithm designed for energy-efficient processors, surpassing existing solutions in both accuracy and performance. Our implementation optimizes photon transport simulations using Apple’s low-power, high-performance M-family chips. Additionally, we explore integrating Machine Learning (ML) techniques to efficiently create a forward solver for the Radiative Transport Equation, identifying top-performing ML models. Our open-source software package integrates ML, optimizing photon transport simulations and facilitating customization for specific applications. Extensive validation against common solvers in biomedical imaging demonstrates comparable accuracy with significantly reduced computational time and energy consumption. This approach maintains accuracy while drastically reducing computational time and energy consumption, offering a promising emerging concept/solution for simulating light propagation in turbid media.
|