Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Department of Radiation Oncology, Peking University Cancer Hospital & Institute, Beijing 100142, China.
State Key Laboratory of Nuclear Physics and Technology, Peking University, Beijing 100871, China.
Phys Med. 2020 Mar;71:82-87. doi: 10.1016/j.ejmp.2020.02.016. Epub 2020 Feb 22.
To evaluate the impact of deformation magnitude and image modality on deformable-image-registration (DIR) accuracy using Halcyon megavoltage cone beam CT images (MVCBCT).
Planning CT images of an anthropomorphic Head phantom were aligned rigidly with MVCBCT and re-sampled to achieve the same resolution, denoted as pCT. MVCBCT was warped with twenty simulated pre-known virtual deformation fields (T, i = 1-20) with increasing deformation magnitudes, yielding warped CBCT (wCBCT). The pCT and MVCBCT were registered to wCBCT respectively (Multi-modality and Uni-modality DIR), generating deformation vector fields V and V' (i = 1-20). V and V' were compared with T respectively to assess the DIR accuracy geometrically. In addition, V, T, and V' were applied to pCT, generating deformed CT (dCT), ground-truth CT (G) and deformed CT' (dCT') respectively. The Hounsfield Unit (HU) on these virtual CT images were also compared.
The mean errors of vector displacement increased with the deformation magnitude. For deformation magnitudes between 2.82 mm and 7.71 mm, the errors of uni-modality DIR were 1.16 mm ~ 1.73 mm smaller than that of multi-modality (p = 0.0001, Wilcoxon signed rank test). DIR could reduce the maximum signed and absolute HU deviations from 70.8 HU to 11.4 HU and 208 HU to 46.2 HU respectively.
As deformation magnitude increases, DIR accuracy continues to deteriorate and uni-modality DIR consistently outperformed multi-modality DIR. DIR-based adaptive radiotherapy utilizing the noisy MVCBCT images is only conditionally applicable with caution.
使用 Halcyon 兆伏锥形束 CT 图像（MVCBCT）评估变形幅度和图像模态对可变形图像配准（DIR）精度的影响。
将人体头部模型的计划 CT 图像与 MVCBCT 刚性配准，并重新采样以达到相同的分辨率，记为 pCT。使用二十个模拟的已知虚拟变形场（T，i=1-20）对 MVCBCT 进行变形，变形幅度逐渐增大，得到变形后的锥形束 CT（wCBCT）。分别将 pCT 和 MVCBCT 配准到 wCBCT 上（多模态和单模态 DIR），生成变形矢量场 V 和 V'（i=1-20）。分别将 V 和 V'与 T 进行比较，以评估 DIR 的几何精度。此外，将 V、T 和 V'应用于 pCT，生成变形 CT（dCT）、真实 CT（G）和变形 CT'（dCT'）。还比较了这些虚拟 CT 图像上的 CT 值。
矢量位移的平均误差随变形幅度的增大而增大。在 2.82mm 到 7.71mm 的变形幅度之间，单模态 DIR 的误差比多模态 DIR 小 1.16mm 到 1.73mm（p=0.0001，Wilcoxon 符号秩检验）。DIR 可将最大符号和绝对值 HU 偏差从 70.8HU 降低到 11.4HU 和从 208HU 降低到 46.2HU。
随着变形幅度的增大，DIR 的精度持续恶化，单模态 DIR 始终优于多模态 DIR。利用噪声较大的 MVCBCT 图像进行基于 DIR 的自适应放疗仅在谨慎条件下适用。
