Quality of density deformations in patient anatomy during synthetic CT guided online adaptive radiation therapy for pelvic cancer

盆腔癌合成CT引导在线自适应放射治疗过程中患者解剖结构密度形变的质量

阅读:1

Abstract

BACKGROUND: Online adaptive radiotherapy (oART) with cone-beam computed tomography (CBCT) enables real-time plan adaptation to daily anatomical changes in the patient. Synthetic CTs (sCTs), generated by deforming planning CTs (pCTs) to CBCTs, are crucial for precise dose calculations. However, variations such as gaseous volumes in the pelvis may impact alignment accuracy. MATERIALS AND METHODS: This retrospective study analyzed 90 treatment fractions for nine patients with cancer in the pelvic region, treated with CBCT-guided oART on the Ethos system. SCTs were generated and compared to corresponding CBCTs using similarity metrics, focusing on bony structures, external body contours, and gaseous volumes. Symmetric differences and Dice Similarity Coefficients (DSC) were calculated to evaluate the registration accuracy. RESULTS: Bony structure deformations were minor (average volume change 0.2%, DSC 0.89). External body contours on sCTs aligned closely with CBCT (mean deviation 1.0 ± 0.4%), while gaseous volumes exhibited lower alignment (DSC < 0.5 in many fractions). Gaseous volume discrepancies were independent of the pCT and showed time-dependent variability, emphasizing the challenges in adapting density maps. Only a small portion of the gaseous volume differences overlapped with the planning target volume (PTV), suggesting limited direct impact on dose calculation. CONCLUSIONS: CBCT-based sCT adaptation effectively aligns external contours but does not accurately handle gaseous volumes, which may vary in great extent. Minimizing the time between CBCT acquisition and treatment is crucial to maintain alignment accuracy. Advanced imaging solutions enabling direct dose calculation without involving sCT generation could further enhance dose reconstruction but may not resolve the time-sensitive nature of gaseous volume changes.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。