Center for Proton Therapy, Paul Scherrer Institut, Villigen PSI, Switzerland.
Department of Computer Science, ETH Zürich, Zürich, Switzerland.
Med Phys. 2024 Nov;51(11):8302-8316. doi: 10.1002/mp.17338. Epub 2024 Aug 13.
The use of magnetic resonance (MR) imaging for proton therapy treatment planning is gaining attention as a highly effective method for guidance. At the core of this approach is the generation of computed tomography (CT) images from MR scans. However, the critical issue in this process is accurately aligning the MR and CT images, a task that becomes particularly challenging in frequently moving body areas, such as the head-and-neck. Misalignments in these images can result in blurred synthetic CT (sCT) images, adversely affecting the precision and effectiveness of the treatment planning.
This study introduces a novel network that cohesively unifies image generation and registration processes to enhance the quality and anatomical fidelity of sCTs derived from better-aligned MR images.
The approach synergizes a generation network (G) with a deformable registration network (R), optimizing them jointly in MR-to-CT synthesis. This goal is achieved by alternately minimizing the discrepancies between the generated/registered CT images and their corresponding reference CT counterparts. The generation network employs a UNet architecture, while the registration network leverages an implicit neural representation (INR) of the displacement vector fields (DVFs). We validated this method on a dataset comprising 60 head-and-neck patients, reserving 12 cases for holdout testing.
Compared to the baseline Pix2Pix method with MAE 124.95 30.74 HU, the proposed technique demonstrated 80.98 7.55 HU. The unified translation-registration network produced sharper and more anatomically congruent outputs, showing superior efficacy in converting MR images to sCTs. Additionally, from a dosimetric perspective, the plan recalculated on the resulting sCTs resulted in a remarkably reduced discrepancy to the reference proton plans.
This study conclusively demonstrates that a holistic MR-based CT synthesis approach, integrating both image-to-image translation and deformable registration, significantly improves the precision and quality of sCT generation, particularly for the challenging body area with varied anatomic changes between corresponding MR and CT.
磁共振（MR）成像在质子治疗计划中作为一种非常有效的指导方法受到关注。该方法的核心是从 MR 扫描中生成计算机断层扫描（CT）图像。然而，在这个过程中，关键问题是准确地对齐 MR 和 CT 图像，在头颈部等经常移动的身体区域，这一任务变得特别具有挑战性。这些图像的配准不良会导致合成 CT（sCT）图像模糊，从而影响治疗计划的准确性和有效性。
本研究介绍了一种新颖的网络，该网络将图像生成和配准过程统一起来，以提高从配准较好的 MR 图像中得出的 sCT 的质量和解剖保真度。
该方法通过交替最小化生成/配准的 CT 图像与其相应参考 CT 图像之间的差异，协同使用生成网络（G）和可变形配准网络（R）来实现 MR 到 CT 合成中的优化。生成网络采用 UNet 架构，配准网络利用位移矢量场（DVFs）的隐式神经表示（INR）。我们在包含 60 例头颈部患者的数据集上验证了该方法，其中 12 例用于保留测试。
与基线 Pix2Pix 方法相比，MAE 为 124.95 30.74 HU，该方法的 MAE 为 80.98 7.55 HU。所提出的统一翻译-配准网络产生的输出更加清晰和解剖一致，在将 MR 图像转换为 sCT 方面显示出更好的效果。此外，从剂量学的角度来看，在生成的 sCT 上重新计算计划，与参考质子计划的差异显著减小。
本研究明确表明，基于 MR 的 CT 综合方法，将图像到图像的转换和可变形配准集成在一起，显著提高了 sCT 生成的精度和质量，特别是对于对应 MR 和 CT 之间解剖变化较大的挑战性身体区域。
