Zhang Zhehao, Hao Yao, Jin Xiyao, Yang Deshan, Kamilov Ulugbek S, Hugo Geoffrey D
Department of Radiation Oncology, Washington University School of Medicine in St. Louis, St. Louis, MO, United States of America.
Department of Radiation Oncology, Duke University School of Medicine, Durham, NC, United States of America.
Biomed Phys Eng Express. 2024 Dec 23;11(1):015030. doi: 10.1088/2057-1976/ad97c1.
. Previous work has that deep learning (DL)-enhanced 4D cone beam computed tomography (4D-CBCT) images improve motion modeling and subsequent motion-compensated (MoCo) reconstruction for 4D-CBCT. However, building the motion model at treatment time via conventional deformable image registration (DIR) methods is not temporally feasible. This work aims to improve the efficiency of 4D-CBCT MoCo reconstruction using DL-based registration for the rapid generation of a motion model prior to treatment.An artifact-reduction DL model was first used to improve the initial 4D-CBCT reconstruction by reducing streaking artifacts. Based on the artifact-reduced phase images, a groupwise DIR employing DL was used to estimate the inter-phase motion model. Two DL DIR models using different learning strategies were employed: (1) a patient-specific one-shot DIR model which was trained from scratch only using the images to be registered, and (2) a population DIR model which was pre-trained using collected 4D-CT images from 35 patients. The registration accuracy of two DL DIR models was assessed and compared to a conventional groupwise DIR approach implemented in the Elastix toolbox using the publicly available DIR-Lab dataset, a Monte Carlo simulation dataset from the SPARE challenge, and two clinical cases.The patient-specific DIR model and the population DIR model demonstrated registration accuracy comparable to the conventional state-of-the-art methods on the DIR-Lab dataset. No significant difference in image quality was observed between the final MoCo reconstructions using the patient-specific model and population model for motion modeling, compared to using the conventional approach. The average runtime (hh:mm:ss) of the entire MoCo reconstruction on SPARE dataset was reduced from 01:37:26 using conventional DIR method to 00:10:59 using patient-specific model and 00:01:05 using the pre-trained population model.DL-based registration methods can improve the efficiency in generating motion models for 4D-CBCT without compromising the performance of final MoCo reconstruction.
先前的研究表明，深度学习（DL）增强的4D锥束计算机断层扫描（4D-CBCT）图像可改善4D-CBCT的运动建模及后续运动补偿（MoCo）重建。然而，在治疗时通过传统的可变形图像配准（DIR）方法构建运动模型在时间上是不可行的。这项工作旨在通过基于DL的配准提高4D-CBCT MoCo重建的效率，以便在治疗前快速生成运动模型。首先使用一个减少伪影的DL模型，通过减少条纹伪影来改善初始4D-CBCT重建。基于减少伪影后的相位图像，采用基于DL的分组DIR来估计相间运动模型。使用了两种采用不同学习策略的DL DIR模型：（1）一种针对特定患者的一次性DIR模型，该模型仅使用待配准图像从头开始训练；（2）一种群体DIR模型，该模型使用从35名患者收集的4D-CT图像进行预训练。使用公开可用的DIR-Lab数据集、来自SPARE挑战的蒙特卡罗模拟数据集以及两个临床病例，评估了两种DL DIR模型的配准精度，并与在Elastix工具箱中实现的传统分组DIR方法进行了比较。针对特定患者的DIR模型和群体DIR模型在DIR-Lab数据集上的配准精度与传统的最先进方法相当。与使用传统方法相比，使用针对特定患者的模型和群体模型进行运动建模的最终MoCo重建在图像质量上没有观察到显著差异。在SPARE数据集上，整个MoCo重建的平均运行时间（时:分:秒）从使用传统DIR方法的01:37:26减少到使用针对特定患者模型的00:10:59和使用预训练群体模型的00:01:05。基于DL的配准方法可以提高为4D-CBCT生成运动模型的效率，而不会影响最终MoCo重建的性能。
