Estimating the 4D respiratory lung motion by spatiotemporal registration and super‐resolution image reconstruction

Abstract

Purpose:One of the main challenges in lung cancer radiation therapy is how to reduce the treatment margin but accommodate the geometric uncertainty of moving tumor. 4D-CT is able to provide the full range of motion information for the lung and tumor. However, accurate estimation of lung motion with respect to the respiratory phase is difficult due to various challenges in image registration, e.g., motion artifacts and large interslice thickness in 4D-CT. Meanwhile, the temporal coherence across respiration phases is usually not guaranteed in the conventional registration methods which consider each phase image in 4D-CT independently. To address these challenges, the authors present a unified approach to estimate the respiratory lung motion with two iterative steps. Methods: First, the authors propose a novel spatiotemporal registration algorithm to align all phase images of 4D-CT (in low-resolution) to a high-resolution group-mean image in the common space. The temporal coherence of registration is maintained by a set of temporal fibers that delineate temporal correspondences across different respiratory phases. Second, a super-resolution technique is utilized to build the high-resolution group-mean image with more anatomical details than any individual phase image, thus largely alleviating the registration uncertainty especially in correspondence detection. In particular, the authors use the concept of sparse representation to keep the group-mean image as sharp as possible. Results: The performance of our 4D motion estimation method has been extensively evaluated on both the simulated datasets and real lung 4D-CT datasets. In all experiments, our method achieves more accurate and consistent results in lung motion estimation than all other state-of-the-art approaches under comparison. Conclusions:The authors have proposed a novel spatiotemporal registration method to estimate the lung motion in 4D-CT. Promising results have been obtained, which indicates the high applicability of our method in clinical lung cancer radiation therapy.

LROC Investigation of Three Strategies for Reducing the Impact of Respiratory Motion on the Detection of Solitary Pulmonary Nodules in SPECT

“…Its application is clinically dependent on a dual-modality imaging system such as SPECT/CT (or PET/CT) where the motion of the SPN could be estimated in respiratory gated CT slices, and applied to correct the SPECT reconstructions [16;17;37]. Currently two separate imaging sessions are necessary with SPECT/CT in order to acquire both the lesion motion and SPECT data.…”

Lung motion estimation using dynamic point shifting: An innovative model based on a robust point matching algorithm

“…We use the target registration errors 39 of the registration results to evaluate the respiratory motion estimation of the lung. The registration results from the methods of Castillo et al, 3 Wu et al, 13 Metz et al, 5 Heinrich et al, 7 and our method on 300 landmark points between the MI phase and ME phase are shown in Table II Table II, the mean and standard deviation of the target registration errors for the 3000 landmark points of ten cases in the DIR-lab dataset are 1.21 and 1.04 mm, respectively. The results of our method are the best for most of the cases with the smallest mean and standard deviation of the target registration errors.…”

“…In Table III, the results obtained by Metz et al 5 are only available for the POPI-model dataset and for the first five cases of the Dir-lab dataset, and the results obtained by Wu et al 13 are available for the entire DIR-lab dataset. For fair comparison, the results of our method will be compared to the results of Wu et al obtained without super-resolution.…”

“…Compared to an intensity-based image registration algorithm, a featurebased image registration algorithm avoids the need to solve a large system of equations and can handle an image registration problem with a large deformation. Wu et al 13 improved the corresponding precision of points by adding image information consisting of image intensity, intensity gradient magnitude, and zero-order geometric moments of each segmented structure, but the positions of the points are fixed, and the registration precision cannot be further improved.…”

Tensor-based dictionary learning for dynamic tomographic reconstruction

“…The correlation across different phases must be taken into account [1–4][19]. For example, a spatio-temporal dictionary has been trained to encode the whole dataset which is superior to the spatial dictionary [3][4] because the temporal coherence across respiration phases is usually not guaranteed otherwise [20]. Generally speaking, the dimensionality of this dynamic dictionary is 3-D or higher dimensions, and the tensor notion has been used to deal with multidimensional data as a powerful tool [21]–[24].…”