Scholarlink

<p>GPU-based Monte Carlo (MC) simulations are highly valued for their potential to improve both the computational efficiency and accuracy of radiotherapy. However, in proton therapy, these methods often simplify human tissues as water for nuclear reactions, disregarding their true elemental composition and thereby potentially compromising calculation accuracy. Consequently, this study developed the program gMCAP (GPU-based proton MC Algorithm for Proton therapy), incorporating precise discrete interactions, and established a refined nuclear reaction model (REFINED) that considers the actual materials of the human body. Compared to the approximate water model (APPROX), the REFINED model demonstrated an improvement in calculation accuracy of 3%. In particular, in high-density tissue regions, the maximum dose deviation between the REFINED and APPROX models was up to 15%. In summary, the gMCAP program can efficiently simulate 1 million protons within 1&#xa0;s while significantly enhancing dose calculation accuracy in high-density tissues, thus providing a more precise and efficient engine for proton radiotherapy dose calculations in clinical practice.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

gMCAP: a GPU-based Monte Carlo proton transport program for high-density tissues with precise nuclear reaction models

  • 期刊论文

  • 发表日期：

    2025年3月20日

查看全文
  • Xi-Yu Luo,
  • Liang Sun,
  • Zhen Wu,
  • Rui Qiu,
  • Shou-Ping Xu,
  • Hui Zhang,
  • Jun-Li Li

摘要

GPU-based Monte Carlo (MC) simulations are highly valued for their potential to improve both the computational efficiency and accuracy of radiotherapy. However, in proton therapy, these methods often simplify human tissues as water for nuclear reactions, disregarding their true elemental composition and thereby potentially compromising calculation accuracy. Consequently, this study developed the program gMCAP (GPU-based proton MC Algorithm for Proton therapy), incorporating precise discrete interactions, and established a refined nuclear reaction model (REFINED) that considers the actual materials of the human body. Compared to the approximate water model (APPROX), the REFINED model demonstrated an improvement in calculation accuracy of 3%. In particular, in high-density tissue regions, the maximum dose deviation between the REFINED and APPROX models was up to 15%. In summary, the gMCAP program can efficiently simulate 1 million protons within 1 s while significantly enhancing dose calculation accuracy in high-density tissues, thus providing a more precise and efficient engine for proton radiotherapy dose calculations in clinical practice.