头颈部肿瘤射野摆位误差的差异性研究

The Variability of Field Placement' Errors in Patients with Head and Neck Cancer

目的:研究采用电子射野系统评价头颈部肿瘤患者摆位误差时的观察者间变异和观察者自身变异。方法:两队研究小组,分别由四名医师和四名技师组成,两组分别对6名头颈部肿瘤的患者,采用电子射野影像仪(electronic portalimaging device,EPID)拍摄验证片(electronic portal images,EPIs),在EPIs上勾画骨性标志,以放疗计划生成的数字重建图像(digitally reconstructed radiographs,DRRs)做为参考图像,定量分析不同观察研究人员之间和研究人员自身采用EPID确定头颈部肿瘤患者的射野摆位误差(field placement errors,FPEs)的差异性。结果:在前/后位射野图像上,不同医师之间、医师自身及技师自身对摆位误差的判断无明显差异,但在技师之间出现了明显的自身差异性,医师组和技师组在前/后射野图像上的均方根(root-mean-square,RMS)分别为2.52±0.46和3.43±0.43,两者具有明显差异;在侧位野图像上,医师自身对摆位误差的判断有较好的稳定性,但部分不同医师之间在腹背、头足方向上部分患者中出现差异,而不同的技师之间存在明显差异性,医师组和技师组在侧位射野图像上的RMS分别为2.72±0.16和2.62±0.22,两者无明显差异。结论:医师和技师组在采用电子射野系统对头颈部摆位误差进行判断时存在人员之间的误差,应对所有人员进行统一训练才能减少射野摆位误差,从而提高IMRT治疗效果。

Objective： To investigate the variability of inter-observer and intra-observer when evaluating field position errors in patients with head and neck cancer. Methods： Two observergroups consisted of four radiation oncologist and four radiation technologists respectively. For six patients with head and neck cancer, electronic portal images （EPIs）were obtained by an electronic portal imaging device （EPID）, the bony landmarks were outlined on EPIs. These digitally reconstructed radiographs images from radiotherapy planning were regarded as the reference images. Quantitatively assessment was performed to analysed the variability of inter-observer and intra-observer when evaluating field position errors in patients with head and neck cancer using EPID. Results： On anterior/posterior portal fields, there are no statistically significant for intra-oncologist, intra-technologists variation and inter-fraction-patient position variation for radiation oncologist, the part of inter-fraction-patient position variation were statistically significant for radiation technologists （P〈0.05）. Averaged RMS deviation for the two directional components was statistically significant, 2.52±0.46 and 3.43±0.43 respectively.On lateral portal fields, there was a high degree of conformity for all intra-observer variability in assessing FPEs. There was statistically significant for a part of inter-oncologist variation in both anterior-posterior and superior-inferior direction. There was significant difference for inter-technologists; Averaged RMS deviation for the two directional components wasn＇t statistically significant, 2.72±0.16 and 2.62± 0.22 respectively. Conclusions： There was a high degree of variability between different groups when evaluating field position errors in patients with head and neck cancer using EPID. This results showed that all technologists must be trained to reduce field position errors to improve the treatment efficiency of IMRT.