|国家预印本平台
| 注册
首页|A physically constrained Energy Spectrum Superposition Method-Machine Learning coupling algorithm for obtaining the neutron spectrum of BNCT rapidly
来源：中国科学院科技论文预发布平台_logo中国科学院科技论文预发布平台

A physically constrained Energy Spectrum Superposition Method-Machine Learning coupling algorithm for obtaining the neutron spectrum of BNCT rapidly

A physically constrained Energy Spectrum Superposition Method-Machine Learning coupling algorithm for obtaining the neutron spectrum of BNCT rapidly

Hao-PengDeng 1杨竣凯 1Zhi-MengHu 2Fang-CongZhang 3Li-KaiGuo 3MinPeng 3Deng-JieXiao 4Ping-QuanWang 5HuiZhang 5Bo-WenZhou 6ChungmingPaulChu 3LinXiao 7GiuseppeGorini8

1. These authors contributed equally to this work and should be considered co-first authors;南京大学 2. Department of Nuclear Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210093, China 3. 南京大学 4. Shenzhen Polytechnic University, The Tech X Academy, Shenzhen 518000, China 5. National Institute of Metrology, Beijing 100029, China 6. Suzhou Laboratory, Suzhou 215000, China 7. School of Physics, Nanjing University, Nanjing 210093, China 8. Department of Physics, University of Milan-Bicocca, Milan 20126, Italy

准确、快速地获取中子能谱是硼中子俘获治疗图像引导放射治疗（IGRT - BNCT）面临的重大挑战之一。本文将包括深度神经网络（DNN）和随机森林（RF）算法在内的机器学习（ML）融入能谱叠加法（ESSM），以预测数据库中不存在的未知能谱，即 ESSM - ML。新算法能在高精度的基础上进一步提高能谱获取速度。DNN 和 RF 对两种物理过程的中子能谱预测均显示出较高的 R² 值和极低的均方根误差（RMSE）。在热中子、超热中子和快中子三个能区，预测值与真实值之间的偏差较小。对于束流整形组件（BSA）发射窗口处的总中子能谱，ESSM - ML 获得的总中子注量率与传统基于模拟的方法获得的总中子注量率之比为 95.3%。此外，对于 ESSM - ML 获得治疗能谱的耗时，所需时间仅为 69 秒，计算效率提高了 4500 倍。ML 模块对单个能谱的平均预测时间仅为 0.0052 秒。ESSM - ML 为实现 IGRT - BNCT 提供了理论和算法基础。

肿瘤学原子能技术基础理论计算技术、计算机技术

ESSM-MLIGRT-BNCT机器学习数据库计算效率

Hao-PengDeng,杨竣凯,Zhi-MengHu,Fang-CongZhang,Li-KaiGuo,MinPeng,Deng-JieXiao,Ping-QuanWang,HuiZhang,Bo-WenZhou,ChungmingPaulChu,LinXiao,GiuseppeGorini.A physically constrained Energy Spectrum Superposition Method-Machine Learning coupling algorithm for obtaining the neutron spectrum of BNCT rapidly[EB/OL].(2025-09-26)[2025-10-01].https://chinaxiv.org/abs/202509.00185.点此复制

Accurately and rapidly obtaining neutron energy spectrum is one of the significant challenges for Image-Guided Radiotherapy of BNCT (IGRT-BNCT). In this paper, machine learning (ML), including deep neural network (DNN) and Random Forest (RF) algorithm, is integrated into energy spectrum superposition method (ESSM) to predict unknown energy spectrum not present in the database, namely ESSM-ML. The new algorithm can further improve the speed of spectrum acquisition with high accuracy. The predictions of neutron energy spectra of two physical processes by both DNN and RF show high R values and extremely low RMSE. In the three energy regions of thermal, epithermal, and fast neutrons, the deviations between predicted and true values are low. For the total neutron energy spectrum at the Beam shaping assembly (BSA) emission window, the ratio of the total neutron fluence rate obtained by ESSM-ML to that by the traditional simulation-based method is 95.3%. Moreover, for the time consumption of treatment energy spectrum obtained by ESSM-ML, the required time is only 69s, with a 4500-fold improvement in computational efficiency. The average prediction time for a single energy spectrum by the ML modules is merely 0.0052s. ESSM-ML provides the theoretical and algorithmic foundation for realizing IGRT-BNCT.

ESSM-MLIGRT-BNCTMachine learningDatabaseComputational efficiency

展开英文信息

评论