Learning‐based motion artifact removal networks for quantitative R2∗ mapping

Xu, Xiaojian; Kothapalli, Satya V. V. N.; Liu, Jiaming; Kahali, Sayan; Gan, Weijie; Yablonskiy, Dmitriy A.; Kamilov, Ulugbek S.

doi:10.1002/mrm.29188

Cited by 15 publications

(24 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We applied the DL-based R * 2 estimation method LEARN-BIO [60] to the reconstructed mGRE images from baseline methods to compute the corresponding R * 2 maps as comparisons to the ones from our end-to-end training. LEARN-BIO has the same architecture as E ϕ of CoRRECT, except that it is not jointly trained with R θ .…”

Section: Baseline Methods For R * 2 Estimationmentioning

confidence: 99%

“…It has also been applied to help magnetic resonance fingerprinting (MRF) [56] with a better and more efficient generation of qMRI maps such as T 1 and T 2 [57,58]. The work [59,60] has explored the potential of self-supervised learning for training qMRI estimation networks directly on MRI images using biophysical models without ground truth qMRI maps. When the measurement operator A is available, it can be combined with biophysical models to enforce data consistency relative to the subsampled measurements, leading to a model-based qMRI mapping.…”

Section: Deep Qmri Map Estimationsmentioning

confidence: 99%

“…We implement this process by replacing sections of k-space lines of the ground-truth MR images x with those from their moved versions to synthesize motion artifacts. To generate a wide-range of artifacts, we control the number of movements, the duration of each movement, and the amplitude of each movements as random numbers following the configuration in [60].…”

Section: Accounting For Object Motionmentioning

confidence: 99%

See 2 more Smart Citations

CoRRECT: A Deep Unfolding Framework for Motion-Corrected Quantitative R2* Mapping

Xu¹,

Gan²,

Kothapalli³

et al. 2022

Preprint

View full text Add to dashboard Cite

Quantitative MRI (qMRI) refers to a class of MRI methods for quantifying the spatial distribution of biological tissue parameters. Traditional qMRI methods usually deal separately with artifacts arising from accelerated data acquisition, involuntary physical motion, and magnetic-field inhomogeneities, leading to suboptimal end-to-end performance. This paper presents CoRRECT, a unified deep unfolding (DU) framework for qMRI consisting of a model-based end-to-end neural network, a method for motion-artifact reduction, and a self-supervised learning scheme. The network is trained to produce R2* maps whose k-space data matches the real data by also accounting for motion and field inhomogeneities. When deployed, CoRRECT only uses the k-space data without any pre-computed parameters for motion or inhomogeneity correction. Our results on experimentally collected multi-Gradient-Recalled Echo (mGRE) MRI data show that CoRRECT recovers motion and inhomogeneity artifact-free R2* maps in highly accelerated acquisition settings. This work opens the door to DU methods that can integrate physical measurement models, biophysical signal models, and learned prior models for high-quality qMRI.

show abstract

Section: Baseline Methods For R * 2 Estimationmentioning

confidence: 99%

Section: Deep Qmri Map Estimationsmentioning

confidence: 99%

Section: Accounting For Object Motionmentioning

confidence: 99%

See 1 more Smart Citation

CoRRECT: A Deep Unfolding Framework for Motion-Corrected Quantitative R2* Mapping

Xu¹,

Gan²,

Kothapalli³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The latter idea of motion detection and undersampled reconstruction has also been demonstrated purely deep learningbased for cardiac ECG mistriggering artefacts [19], enabling end-to-end training with relevant downstream tasks. In the context of T 2 * quantification, Xu et al [25] demonstrate that image-based MoCo of GRE data can be performed endto-end with deep learning based T 2 * mapping. Yet, to the best of our knowledge no data-consistent MoCo method for T2*-weighted GRE data exists so far.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Detection of Motion-Affected k-Space Lines for T2*-Weighted MRI

Eichhorn¹,

Hammernik²,

Spieker³

et al. 2023

Preprint

View full text Add to dashboard Cite

T2*-weighted gradient echo MR imaging is strongly impacted by subject head motion due to motion-related changes in B0 inhomogeneities. Within the oxygenation-sensitive mqBOLD protocol, even mild motion during the acquisition of the T2*-weighted data propagates into errors in derived quantitative parameter maps. In order to correct these images without the need of repeated measurements, we propose to learn a classification of motion-affected k-space lines. To test this, we perform realistic motion simulations including motion-induced field inhomogeneity changes for supervised training. To detect the presence of motion in each phase encoding line, we train a convolutional neural network, leveraging the multi-echo information of the T2*-weighted images. The proposed network accurately detects motion-affected k-space lines for simulated displacements of ≥0.5 mm (accuracy on test set: 92.5 %). Finally, we show example reconstructions where we include these classification labels as weights in the data consistency term of an iterative reconstruction procedure, opening up exciting opportunities of k-space line detection in combination with more powerful reconstruction methods.

show abstract

“…From this perspective, using a deep learning CNN instead of ANN provides a clear advantage in dealing with noisy data 16 and also reducing computation time. Beyond noise compression, CNNs were also demonstrated to be capable of handling motion correction, producing both high‐quality complex qGRE images and quantitative R2 * maps 18 with significantly reduced motion artifacts.…”

Section: Introductionmentioning

confidence: 99%

Deep learning‐based Accelerated and Noise‐Suppressed Estimation (DANSE) of quantitative Gradient‐Recalled Echo (qGRE) magnetic resonance imaging metrics associated with human brain neuronal structure and hemodynamic properties

Kahali

Kothapalli

et al. 2022

NMR in Biomedicine

Self Cite

View full text Add to dashboard Cite

The purpose of the current study was to introduce a Deep learning‐based Accelerated and Noise‐Suppressed Estimation (DANSE) method for reconstructing quantitative maps of biological tissue cellular‐specific, R2t*, and hemodynamic‐specific, R2’, metrics of quantitative gradient‐recalled echo (qGRE) MRI. The DANSE method adapts a supervised learning paradigm to train a convolutional neural network for robust estimation of R2t* and R2’ maps with significantly reduced sensitivity to noise and the adverse effects of macroscopic (B0) magnetic field inhomogeneities directly from the gradient‐recalled echo (GRE) magnitude images. The R2t* and R2’ maps for training were generated by means of a voxel‐by‐voxel fitting of a previously developed biophysical quantitative qGRE model accounting for tissue, hemodynamic, and B0‐inhomogeneities contributions to multigradient‐echo GRE signal using a nonlinear least squares (NLLS) algorithm. We show that the DANSE model efficiently estimates the aforementioned qGRE maps and preserves all the features of the NLLS approach with significant improvements including noise suppression and computation speed (from many hours to seconds). The noise‐suppression feature of DANSE is especially prominent for data with low signal‐to‐noise ratio (SNR ~ 50–100), where DANSE‐generated R2t* and R2’ maps had up to three times smaller errors than that of the NLLS method. The DANSE method enables fast reconstruction of qGRE maps with significantly reduced sensitivity to noise and magnetic field inhomogeneities. The DANSE method does not require any information about field inhomogeneities during application. It exploits spatial and gradient echo time‐dependent patterns in the GRE data and previously gained knowledge from the biophysical model, thus producing high quality qGRE maps, even in environments with high noise levels. These features along with fast computational speed can lead to broad qGRE clinical and research applications.

show abstract

Learning‐based motion artifact removal networks for quantitative R2∗ mapping

Cited by 15 publications

References 54 publications

CoRRECT: A Deep Unfolding Framework for Motion-Corrected Quantitative R2* Mapping

CoRRECT: A Deep Unfolding Framework for Motion-Corrected Quantitative R2* Mapping

Deep Learning-Based Detection of Motion-Affected k-Space Lines for T2*-Weighted MRI

Deep learning‐based Accelerated and Noise‐Suppressed Estimation (DANSE) of quantitative Gradient‐Recalled Echo (qGRE) magnetic resonance imaging metrics associated with human brain neuronal structure and hemodynamic properties

Contact Info

Product

Resources

About