Nonrigid 3D motion estimation at high temporal resolution from prospectively undersampled k‐space data using low‐rank MR‐MOTUS

Huttinga, Niek R. F.; Bruijnen, Tom; Berg, C.A.T. Van den; Sbrizzi, Alessandro

doi:10.1002/mrm.28562

Cited by 23 publications

(60 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many of the early MoCo methods rely on a high-resolution static reference image [15]. Recent approaches [7,16] rely on motion-resolved XD-GRASP reconstructions.…”

Section: Brief Background On Motion-resolved and Motion-compensated R...mentioning

confidence: 99%

“…Many of the approaches require a high-resolution reference image. The recovered images are then registered to the reference image to obtain the motion fields [15]. Another approach is to estimate the motion-maps between the phase images reconstructed by XD-GRASP; the different motion phases are registered together and averaged to obtain a MoCo volume [7].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic imaging using Motion-Compensated SmooThness Regularization on Manifolds (MoCo-SToRM)

Zou,

Torres,

Fain

et al. 2021

Preprint

View full text Add to dashboard Cite

We introduce an unsupervised motion-compensated reconstruction scheme for high-resolution free-breathing pulmonary MRI. We model the image frames in the time series as the deformed version of the 3D template image volume. We assume the deformation maps to be points on a smooth manifold in high-dimensional space. Specifically, we model the deformation map at each time instant as the output of a CNN-based generator that has the same weight for all time-frames, driven by a low-dimensional latent vector. The time series of latent vectors account for the dynamics in the dataset, including respiratory motion and bulk motion. The template image volume, the parameters of the generator, and the latent vectors are learned directly from the k-t space data in an unsupervised fashion. Our experimental results show improved reconstructions compared to state-of-the-art methods, especially in the context of bulk motion during the scans.

show abstract

“…Many of the early MoCo methods rely on a high-resolution static reference image [15]. Recent approaches [7,16] rely on motion-resolved XD-GRASP reconstructions.…”

Section: Brief Background On Motion-resolved and Motion-compensated R...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Dynamic imaging using Motion-Compensated SmooThness Regularization on Manifolds (MoCo-SToRM)

Zou,

Torres,

Fain

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…[3] to acquired k-space data. We refer the reader to our previous works [17], [18] for an extensive discussion on the assumptions underlying the signal model Eq. [3].…”

Section: A Background Mr-motus 1) Forward Signal Modelmentioning

confidence: 99%

“…In this work, we focus on the aforementioned technical challenge in MRgRT and propose real-time low-rank MR-MOTUS [17], [18] for real-time estimation of non-rigid 3D respiratory motion-fields directly from prospectively undersampled 3D kspace data, with a total latency (acquisition and reconstruction) below 200 ms. The MR-MOTUS signal model relates motionfields and a reference image to k-space data, which allows to reconstruct motion-fields directly from k-space data, given a reference image [17].…”

mentioning

confidence: 99%

“…By operating in k-space, the method's latency is not directly impaired by the typical MR-acquisition latency. In [18], MR-MOTUS was extended to reconstruct 3D+t motion-fields in the order of minutes, using an explicit low-rank factorization in static spatial motion components and dynamic temporal motion components. Here, the same lowrank factorization is assumed, and the framework is extended to perform real-time computations by observing that a lowrank factorization allows to split the reconstruction in two phases: (I) an offline preparation phase that separates static spatial motion components from dynamic temporal motion components, and (II) an online phase in which the pre-trained static motion components are fixed and dynamic motion components are estimated with real-time acquisitions and realtime model-based reconstructions.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Real-time non-rigid 3D respiratory motion estimation for MR-guided radiotherapy using MR-MOTUS

Huttinga,

Bruijnen,

Berg

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The MR-Linac is a combination of an MRscanner and radiotherapy linear accelerator (Linac) which holds the promise to increase the precision of radiotherapy treatments with MR-guided radiotherapy by monitoring motion during radiotherapy with MRI, and adjusting the radiotherapy plan accordingly. Optimal MR-guidance for respiratory motion during radiotherapy requires MRbased 3D motion estimation with a latency of 200-500 ms. Currently this is still challenging since typical methods rely on MR-images, and are therefore limited to the 3D MRimaging latency. In this work, we present a method to perform non-rigid 3D respiratory motion estimation with 170 ms latency, including both acquisition and reconstruction. The proposed method called real-time low-rank MR-MOTUS reconstructs motion-fields directly from k-space data, and leverages an explicit low-rank decomposition of motionfields to split the large scale 3D+t motion-field reconstruction problem posed in our previous work into two parts: (I) a medium-scale offline preparation phase and (II) a smallscale online inference phase which exploits the results of the offline phase for real-time computations. The method was validated on free-breathing data of five volunteers, acquired with a 1.5T Elekta Unity MR-Linac. Results show that the reconstructed 3D motion-fields are anatomically plausible, highly correlated with a self-navigation motion surrogate (R = 0.975 ± 0.0110), and can be reconstructed with a total latency of 170 ms that is sufficient for real-time MR-guided abdominal radiotherapy.

show abstract

Volumetric MRI with sparse sampling for MR‐guided 3D motion tracking via sparse prior‐augmented implicit neural representation learning

Liu,

Shen,

Johansson

et al. 2023

Medical Physics

View full text Add to dashboard Cite

BackgroundVolumetric reconstruction of magnetic resonance imaging (MRI) from sparse samples is desirable for 3D motion tracking and promises to improve magnetic resonance (MR)‐guided radiation treatment precision. Data‐driven sparse MRI reconstruction, however, requires large‐scale training datasets for prior learning, which is time‐consuming and challenging to acquire in clinical settings.PurposeTo investigate volumetric reconstruction of MRI from sparse samples of two orthogonal slices aided by sparse priors of two static 3D MRI through implicit neural representation (NeRP) learning, in support of 3D motion tracking during MR‐guided radiotherapy.MethodsA multi‐layer perceptron network was trained to parameterize the NeRP model of a patient‐specific MRI dataset, where the network takes 4D data coordinates of voxel locations and motion states as inputs and outputs corresponding voxel intensities. By first training the network to learn the NeRP of two static 3D MRI with different breathing motion states, prior information of patient breathing motion was embedded into network weights through optimization. The prior information was then augmented from two motion states to 31 motion states by querying the optimized network at interpolated and extrapolated motion state coordinates. Starting from the prior‐augmented NeRP model as an initialization point, we further trained the network to fit sparse samples of two orthogonal MRI slices and the final volumetric reconstruction was obtained by querying the trained network at 3D spatial locations. We evaluated the proposed method using 5‐min volumetric MRI time series with 340 ms temporal resolution for seven abdominal patients with hepatocellular carcinoma, acquired using golden‐angle radial MRI sequence and reconstructed through retrospective sorting. Two volumetric MRI with inhale and exhale states respectively were selected from the first 30 s of the time series for prior embedding and augmentation. The remaining 4.5‐min time series was used for volumetric reconstruction evaluation, where we retrospectively subsampled each MRI to two orthogonal slices and compared model‐reconstructed images to ground truth images in terms of image quality and the capability of supporting 3D target motion tracking.ResultsAcross the seven patients evaluated, the peak signal‐to‐noise‐ratio between model‐reconstructed and ground truth MR images was 38.02 ± 2.60 dB and the structure similarity index measure was 0.98 ± 0.01. Throughout the 4.5‐min time period, gross tumor volume (GTV) motion estimated by deforming a reference state MRI to model‐reconstructed and ground truth MRI showed good consistency. The 95‐percentile Hausdorff distance between GTV contours was 2.41 ± 0.77 mm, which is less than the voxel dimension. The mean GTV centroid position difference between ground truth and model estimation was less than 1 mm in all three orthogonal directions.ConclusionA prior‐augmented NeRP model has been developed to reconstruct volumetric MRI from sparse samples of orthogonal cine slices. Only one exhale and one inhale 3D MRI were needed to train the model to learn prior information of patient breathing motion for sparse image reconstruction. The proposed model has the potential of supporting 3D motion tracking during MR‐guided radiotherapy for improved treatment precision and promises a major simplification of the workflow by eliminating the need for large‐scale training datasets.

show abstract

Nonrigid 3D motion estimation at high temporal resolution from prospectively undersampled k‐space data using low‐rank MR‐MOTUS

Abstract: This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

Cited by 23 publications

References 62 publications

Dynamic imaging using Motion-Compensated SmooThness Regularization on Manifolds (MoCo-SToRM)

Dynamic imaging using Motion-Compensated SmooThness Regularization on Manifolds (MoCo-SToRM)

Real-time non-rigid 3D respiratory motion estimation for MR-guided radiotherapy using MR-MOTUS

Volumetric MRI with sparse sampling for MR‐guided 3D motion tracking via sparse prior‐augmented implicit neural representation learning

Contact Info

Product

Resources

About