Jang Seong Soon, An Na-Young, Kim Hee Kyung, Yang Youngjun, Huh Gil Ja, Jeong Joon Won
Department of Radiation Oncology, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
Department of Radiation Oncology, Daejeon St. Mary's Hospital, Daejeon, Republic of Korea.
Thorac Cancer. 2025 Aug;16(15):e70147. doi: 10.1111/1759-7714.70147.
To evaluate the target localization accuracy of image registration methods in cone beam computed tomography (CBCT)-based image guidance (IG) for lung stereotactic body radiation therapy (SBRT) and the associations with tumor characteristics such as size, mobility, and location.
Four methods involving different matching axes and regions were used to register the planning CT and 3D CBCT images of 36 lung tumors treated with SBRT. The registration axes were divided into 3D translational axes and 6D axes (translational and rotational axes), and the regions were divided into wide rectangular (WR) volume-of-interest (VOI) and tumor-focused (TF) VOI, consisting of the internal target volume (ITV) plus a 1 cm margin.
Compared with the WR registrations, the TF registrations yielded higher localization accuracies for all registration pairs, with differences of 6.3%-9.1% in the percentage of inclusion of the registered CBCT gross tumor volume (GTV) into the ITV and 1.3-1.8 mm in the 3D distance between the ITV and registered CBCT GTV centroids. The TF3D and TF6D registrations yielded similar accuracy metrics of 91.9% and 1.4 mm, respectively, whereas the WR6D registration exhibited improved accuracies compared with the WR3D registration. The localization accuracies with the TF registrations decreased with increasing tumor mobility index, expressed as the ITV/GTV50% ratio, with significant differences between the tumor groups at a cutoff of 1.7.
The localization accuracies of our image registration methods may serve as clinically useful references for selecting the most suitable registration in CBCT-based IG for lung SBRT.
评估基于锥形束计算机断层扫描(CBCT)的图像引导(IG)在肺部立体定向体部放射治疗(SBRT)中图像配准方法的靶区定位准确性,以及与肿瘤大小、活动度和位置等特征的相关性。
使用四种涉及不同匹配轴和区域的方法,对36例接受SBRT治疗的肺部肿瘤的计划CT图像和三维CBCT图像进行配准。配准轴分为三维平移轴和六维轴(平移和旋转轴),区域分为宽矩形(WR)感兴趣体积(VOI)和肿瘤聚焦(TF)VOI,后者由内部靶区体积(ITV)加1 cm边界组成。
与WR配准相比,TF配准在所有配准对中均产生更高的定位准确性,配准后的CBCT大体肿瘤体积(GTV)纳入ITV的百分比差异为6.3%-9.1%,ITV与配准后的CBCT GTV质心之间的三维距离差异为1.3-1.8 mm。TF3D和TF6D配准分别产生了相似的准确性指标,即91.9%和1.4 mm,而WR6D配准与WR3D配准相比准确性有所提高。TF配准的定位准确性随着肿瘤活动度指数(以ITV/GTV50%比值表示)的增加而降低,在临界值为1.7时肿瘤组之间存在显著差异。
我们的图像配准方法的定位准确性可为在基于CBCT的肺部SBRT图像引导中选择最合适的配准提供临床有用的参考。