瓦里安Eclipse计划系统中基于云计算的ARC技术的剂量计算时间的探讨

摘要

目的 探讨如何缩短基于云计算的ARC技术的剂量计算时间.方法 随机选择15例鼻咽癌患者,计划方式选择ARC技术,射野选择四个弧度,计算网格大小选择2.5 mm.对于每一例患者剂量计算时,先选择调用带有Eclipse的2台工作站,记录剂量计算时间,再分别选择调用4,5,6,7,8台工作站,并记录剂量计算时间.结果 剂量计算选择调用6台工作站时计算时间最短,选择7台工作站和8台工作站计算时间不但没有缩短,反而略有增加.选择调用6台工作站比2台工作站计算时间平均缩短约19′06″.采用Mann-Whitney U检验,6台工作站与2,4,5,7,8台工作站相比较,Z=-4.667,-2.634,-1.348,-0.415,-1.390,P=0.000,0.008,0.178,0.678,0.165.结论 调用6台工作站时所用的计算时间短于2台和4台,具有统计学差异;调用6台工作站时所用的计算时间短于5、7和8台,但不具有统计学差异.局域网中增加一定数量用于计算调用的工作站,能够一定程度的缩短剂量计算的时间,提高肿瘤放疗计划设计的效率.%Objective To explore how to shorten dose calculation time of cloud computing based on ARC technique. Methods Fifteen patients with nasopharyngeal carcinoma were randomly selected. ARC technique was chosen to do the plan and four arc fields were used. Two point five mm was selected as calculation grid size. To do dose calculation for every patient, two workstations with Eclipse was selected firstly and dose calculation time was recorded. Then 4, 5, 6, 7 and 8 workstations were separately selected, and dose calculation time was also recorded. Results The calculation time was the shortest when choosing 6 workstations for dose calculation. The calculation time of 7 workstations and 8 workstations was not shortened, but increased slightly. The average computation time of calling 6 workstations was about 19.1 minutes, which was shorter than that of 2 workstations. At the same time, when Mann-Whitney U test was used to conduct comparing analysis on 6 workstations with 2, 4, 5, 7, 8 workstations, the Z score was separately -4.667,-2.634, -1.348, -0.415 and -1.390, and P value was separately 0.000, 0.008, 0.178, 0.678 and 0.165. Conclusion The computation time of calling 6 workstations was shorter than that of 2 and 4 stations, which had statistical difference. The computation time of calling 6 workstations was shorter than that of 5, 7 and 8 stations, but there was no statistical difference. To increase the number of workstations used to calculation call in a LAN, the dose time can be reduced to a certain extent, and the efficiency of tumor radiotherapy plan designing can be improved.