基于R语言BP神经网络瓦里安NovalisTx直线加速器MLC系统故障预测模型研究

doi:10.3760/cma.j.issn.1004-4221.2018.05.012

摘要
图/表
参考文献()
相关文章 (13)
点击分布统计
下载分布统计

全文: PDF (0 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS) 背景资料

摘要

目的构建并研究瓦里安NovalisTx直线加速器MLC系统故障预测BP神经网络模型。方法取加速器临床使用18个月MLC系统故障统计数据为研究对象,以加速器使用总时间、月治疗患者数量、日均开机工作时间、RapidArc计划数量及加速器保养后时间间隔为输入故障因素,以故障频次预测为输出结果,采用R语言AMORE包构建MLC系统故障预测BP神经网络模型并对其进行仿真验证。结果模型采用3层网络实现输入输出转换,其输入层5个节点、隐层13个节点、输出层1个节点;输入层至隐层、隐层至输出层分别选用tansig、purelin传递函数;模型设定最大训练学习次数150次,实际使用111次,设定误差3%,实际误差2.7%,表明其收敛较好。该模型对18个月临床故障数据仿真验证结果表明预测数据与实际数据较为接近。结论基于R语言BP神经网络故障预测模型实现了MLC系统故障因素与故障频次间映射关系描述,可为设备故障规律了解和备件库存管理提供参考。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	邓永锦
	肖振华
	欧阳斌
	王振宇
	黄伯天
	黄镜先
	包勇

关键词 ：神经网络, 多叶准直器, 故障预测, R语言

Abstract：

Objective To construct and investigate the multi-leaf collimator (MLC) fault prediction model of Varian NovalisTx medical linear accelerator based on BP neural network. Methods The MLC fault data applied in clinical trial for 18 months were collected and analyzed. The total use time of accelerator, the quantity of patients per month, average daily working hours of accelerator, volume of RapidArc plans and time interval between accelerator maintenance were used as the input factors and the prediction of MLC fault frequency was considered as the output result. The BP neural network model of MLC fault prediction was realized by AMORE package of R language and the simulation results were validated. Results The model contained 3 layers of network to realize the input-output switch. There were 5 nodes in the input layer,13 nodes in the hide layer and 1 node in the output layer, respectively. The transfer function from the input layer to the hide layer selected the tansig function and purelin function was used from the hide layer to the output layer. The maximum time of training was pre-set as 150 in the designed model. Actually, 111 times of training were performed. The pre-set error was 3% and the actual error was 2.7%, which indicated good convergence. The simulation results of MLC fault applied in clinical trial for 18 months were similar to the actual data. Conclusions The BP neural network model realized by R language of MLC fault prediction can describe the mapping relationship between fault factors and fault frequency, which provides references for the understanding of accelerator fault and management of spare parts inventory.

Key words： Neural network Multi-leaf collimator Fault prediction R language

收稿日期: 2017-08-07

通讯作者: 包勇,Email:baoyong@mail.sysu.edu.cn

引用本文:

邓永锦,肖振华,欧阳斌等. 基于R语言BP神经网络瓦里安NovalisTx直线加速器MLC系统故障预测模型研究[J]. 中华放射肿瘤学杂志, 2018, 27(5): 495-499.

Deng Yongjin,Xiao Zhenhua,Ouyang Bin et al. Research on multi-leaf collimator fault prediction model of Varian Novalis Tx medical linear accelerator based on BP Neural Network realized by R language[J]. Chinese Journal of Radiation Oncology, 2018, 27(5): 495-499.

链接本文:

http://journal12.magtechjournal.com/Jweb_fszlx/CN/10.3760/cma.j.issn.1004-4221.2018.05.012 或 http://journal12.magtechjournal.com/Jweb_fszlx/CN/Y2018/V27/I5/495

[1]Rosenberg I.Radiation Oncology physics:a handbook for teachers and students[J].Br J Cancer,2008,98(5):1020.DOI:10.1038/sj.bjc.6604224
[2]Varian Medical Systems, Inc., Oncology Systems (Palo Alto, California). Varian HD 120 MLC[J].Biomed Saf Stand,2011,41(20):159.DOI:10.1097/01.BMSAS.0000407640.03039.53.
[3]肖振华,王振宇,文碧秀.医用电子直线加速器硬件系统的研究进展和临床应用[J].中国现代医生,2013,51(27):19-21,24.
Xiao ZH,Wang ZY,Wen BX.Hardware system of medical electronic linear accelerator:advance and its clinical application[J].China Mod Doct,2013,51(27):19-21,24.
[4]邓永锦,肖振华,王成涛,等.Synergy医用电子直线加速器运行1年故障原因和频率分析[J].中国医药导报,2017,14(6):73-76.
Deng YJ,Xiao ZH,Wang CT,et al. Analysis of causes and frequency on malfunction of Synergy medical linear accelerator running in one year[J].China Med Her,2017,14(6):73-76.
[5]谭庭强,李黎,黄仁炳.23EX直线加速器7年故障统计分析经验[J].中华放射肿瘤学杂志,2016,25(12):1341-1344.DOI:10.3760/cma.j.issn.1004?4221.2016.12.015.
Tan TQ,Li L,Huang RB.Experience in statistical analysis of failure of a 23EX linear accelerator within 7 years[J].Chin J Radiat Oncol,2016,25(12):1341-1344.DOI:10.3760/cma.j.issn.1004?4221.2016.12.015.
[6]Mobley RK.An introduction to predictive maintenance[M].2nd ed. Woburn,MA:Elsevier,2002.
[7]何勇枢,陈赣.基于BP神经网络模型的故障预测分析[J].微计算机信息,2006,22(16):220-222.DOI:10.3969/j.issn.1008-0570.2006.16.083.
He YS,Chen G.Fault forecast analysis based on BP neural network model[J].Control Autom,2006,22(16):220-222.DOI:10.3969/j.issn.1008-0570.2006.16.083.
[8]刘春.遗传算法优化BP神经网络的网络流量预测[J].信息安全与技术,2014,5(6):82-86.DOI:10.3969/j.issn.1674-9456.2014.06.028.
Liu C.Network traffic prediction model based on genetic algorithm optimizing BP neural network[J].Inf Secur Technol,2014,5(6):82-86.DOI:10.3969/j.issn.1674-9456.2014.06.028.
[9]Able C,Hampton C,Baydush A,et al. SU-E-T-205:MLC predictive maintenance using statistical process control analysis[J].Med Phys,2012,39(6):3750.DOI:10.1118/1.4735265.
[10]曾文韬,孔建益,胡晓莉,等.基于BP神经网络混凝土泵车臂架故障预测研究[J].机械设计与制造,2015(4):44-47.DOI:10.3969/j.issn.1001-3997.2015.04.012.
Zeng WT,Kong JY,Hu XL,et al. Research on concrete pump truck arm frame fault prediction based on BP neural network[J].Mach Des Manufact,2015(4):44-47.DOI:10.3969/j.issn.1001-3997.2015.04.012.
[11]Guedes J,Kikuchi N.Preprocessing and postprocessing for materials based on the homogenization method with adaptive finite element methods[J].Compu. MethodsAppl Mech Eng,1990,83(2):143-198.DOI:10.1016/0045-7825(90)90148-F.
[12]Haykin S.Neural networks:a comprehensive foundation[M].3rd ed. Upper Saddle River,NJ,USA:Prentice Hall International,Inc.,2007.
[13]Kabacoff R.R in action[M].Shelter Island,NY:Manning,2011.
[14]王广善,杨波,庞廷田,等.VMAT模式下MLC叶片运动速度对到位误差影响[J].中华放射肿瘤学杂志,2017,26(2):182-186.DOI:10.3760/cma.j.issn.1004-4221.2017.02.013.
Wang GS,Yang B,Pang TT,et al. Effect of leaf movement speed on setup error of multileaf collimator in volumetric modulated arc therapy mode[J].Chin J Radiat Oncol,2017,26(2):182-186.DOI:10.3760/cma.j.issn.1004-4221.2017.02.013.
[15]Able CM,Baydush AH,Nguyen C,et al. A model for preemptive maintenance of medical linear accelerators—predictive maintenance[J].Radiat Oncol,2016,11:36.