College of Medicine and Biological information Engineering, Northeastern University, Shenyang, China.
Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Shenyang, China.
Med Phys. 2022 Feb;49(2):1209-1215. doi: 10.1002/mp.15350. Epub 2021 Nov 23.
Monte Carlo (MC) simulation is the most frequently used method to numerically model the light propagation in biological tissues because of its high flexibility and precision. Although MC simulation is assumed to be capable of achieving any desired precision, larger number of photons are always necessary for more precise simulation, leading to its major limitation of intensive computation. In this work, the authors present a way to adapt generative adversarial networks (GAN) to accelerate MC simulation.
The pix2pix network, a variant of GAN, was investigated to reconstruct precise MC simulation results from the results roughly modeled by small amount of photons, thus the computation time was expected to be significantly saved. The proposed method was tested on single-layer embedded tumor models to derive the absorption distribution maps.
The results demonstrate that the absorption distribution maps reconstructed from the simulation of only 10 000 photons were very similar to those modeled by using 1 000 000 photons, based on the criterion of peak signal to noise ratio (PSNR) and percentage difference of power coupling efficiencies, and the simulation process was proved to be accelerated by approximately 102 times.
For the first time, GAN was adapted to save computation time of MC simulation of light propagation. By achieving MC simulation with acceptable quality, the proposed method can speed up the computation by hundreds of times.
由于其灵活性和精度高，蒙特卡罗（MC）模拟是数值建模生物组织中光传播最常用的方法。尽管 MC 模拟被认为能够达到任何所需的精度，但为了更精确的模拟，总是需要更多数量的光子，这导致其主要的计算密集型限制。在这项工作中，作者提出了一种利用生成对抗网络（GAN）加速 MC 模拟的方法。
研究了 pix2pix 网络，这是 GAN 的一种变体，它可以从少量光子大致建模的结果中重建精确的 MC 模拟结果，从而有望显著节省计算时间。该方法在单层嵌入式肿瘤模型上进行了测试，以得出吸收分布图。
结果表明，基于峰值信噪比（PSNR）和功率耦合效率差异百分比的标准，从仅模拟 10000 个光子的模拟中重建的吸收分布图与使用 1000000 个光子建模的吸收分布图非常相似，并且模拟过程被证明可以加速约 102 倍。
首次将 GAN 应用于节省光传播 MC 模拟的计算时间。通过实现具有可接受质量的 MC 模拟，该方法可以将计算速度提高数百倍。
