Department of Radiotherapy, University Medical Center Utrecht, Utrecht, Netherlands. Center for Image Sciences, University Medical Center Utrecht, Utrecht, Netherlands. Image Science Institute, University Medical Center Utrecht, Utrecht, Netherlands. The authors equally contributed.
Phys Med Biol. 2018 Sep 10;63(18):185001. doi: 10.1088/1361-6560/aada6d.
To enable magnetic resonance (MR)-only radiotherapy and facilitate modelling of radiation attenuation in humans, synthetic CT (sCT) images need to be generated. Considering the application of MR-guided radiotherapy and online adaptive replanning, sCT generation should occur within minutes. This work aims at assessing whether an existing deep learning network can rapidly generate sCT images for accurate MR-based dose calculations in the entire pelvis. A study was conducted on data of 91 patients with prostate (59), rectal (18) and cervical (14) cancer who underwent external beam radiotherapy acquiring both CT and MRI for patients' simulation. Dixon reconstructed water, fat and in-phase images obtained from a conventional dual gradient-recalled echo sequence were used to generate sCT images. A conditional generative adversarial network (cGAN) was trained in a paired fashion on 2D transverse slices of 32 prostate cancer patients. The trained network was tested on the remaining patients to generate sCT images. For 30 patients in the test set, dose recalculations of the clinical plan were performed on sCT images. Dose distributions were evaluated comparing voxel-based dose differences, gamma and dose-volume histogram (DVH) analysis. The sCT generation required 5.6 s and 21 s for a single patient volume on a GPU and CPU, respectively. On average, sCT images resulted in a higher dose to the target of maximum 0.3%. The average gamma pass rates using the 3%, 3 mm and 2%, 2 mm criteria were above 97 and 91%, respectively, for all volumes of interests considered. All DVH points calculated on sCT differed less than ±2.5% from the corresponding points on CT. Results suggest that accurate MR-based dose calculation using sCT images generated with a cGAN trained on prostate cancer patients is feasible for the entire pelvis. The sCT generation was sufficiently fast for integration in an MR-guided radiotherapy workflow.
为了实现仅基于磁共振（MR）的放射治疗并促进人体辐射衰减建模，需要生成合成 CT（sCT）图像。考虑到 MR 引导放射治疗和在线自适应调整计划的应用，sCT 生成应在数分钟内完成。本研究旨在评估现有的深度学习网络是否能够快速生成 sCT 图像，以便在整个骨盆中进行基于 MR 的精确剂量计算。对 91 例前列腺癌（59 例）、直肠癌（18 例）和宫颈癌（14 例）患者进行了研究，这些患者在接受外照射放射治疗时均同时采集了 CT 和 MRI 用于患者模拟。使用从常规双梯度回波序列获得的狄克逊重建水、脂肪和同相位图像来生成 sCT 图像。在 32 例前列腺癌患者的 2D 横断面上以配对方式训练条件生成对抗网络（cGAN）。在测试集中的其余患者上测试训练好的网络以生成 sCT 图像。对测试集中的 30 例患者，在 sCT 图像上对临床计划进行剂量重新计算。通过基于体素的剂量差异、伽马和剂量体积直方图（DVH）分析评估剂量分布。单个患者体积在 GPU 和 CPU 上的 sCT 生成分别需要 5.6 秒和 21 秒。平均而言，sCT 图像导致靶区的剂量最高增加 0.3%。使用 3%、3mm 和 2%、2mm 标准的平均伽马通过率在所有考虑的感兴趣体积中均超过 97%和 91%。在 sCT 上计算的所有 DVH 点与 CT 上相应点的差异均小于±2.5%。结果表明，使用针对前列腺癌患者进行训练的 cGAN 生成的 sCT 图像进行基于 MR 的精确剂量计算对于整个骨盆是可行的。sCT 生成速度足够快，可以集成到 MR 引导的放射治疗工作流程中。
