24,883 research outputs found
Autocalibrating and calibrationless parallel magnetic resonance imaging as a bilinear inverse problem
No full textModern reconstruction methods for magnetic resonance imaging (MRI) exploit the spatially varying sensitivity profiles of receive-coil arrays as additional source of information. This allows to reduce the number of time-consuming Fourier-encoding steps by undersampling. The receive sensitivities are a priori unknown and influenced by geometry and electric properties of the (moving) subject. For optimal results, they need to be estimated jointly with the image from the same undersampled measurement data. Formulated as an inverse problem, this leads to a bilinear reconstruction problem related to multi-channel blind deconvolution. In this work, we will discuss some recently developed approaches for the solution of this problem
A Multi-GPU Programming Library for Real-Time Applications
No full textWe present MGPU, a C++ programming library targeted at single-node multi-GPU systems. Such systems combine disproportionate floating point performance with high data locality and are thus well suited to implement real-time algorithms. We describe the library design, programming interface and implementation details in light of this specific problem domain. The core concepts of this work are a novel kind of container abstraction and MPI-like communication methods for intra-system communication. We further demonstrate how MGPU is used as a framework for porting existing GPU libraries to multi-device architectures. Putting our library to the test, we accelerate an iterative non-linear image reconstruction algorithm for real-time magnetic resonance imaging using multiple GPUs. We achieve a speed-up of about 1.7 using 2 GPUs and reach a final speed-up of 2.1 with 4 GPUs. These promising results lead us to conclude that multi-GPU systems are a viable solution for real-time MRI reconstruction as well as signal-processing applications in general
Bayesian MRI reconstruction with joint uncertainty estimation using diffusion models
No full textPurpose
We introduce a framework that enables efficient sampling from learned probability distributions for MRI reconstruction.
Method
Samples are drawn from the posterior distribution given the measured k-space using the Markov chain Monte Carlo (MCMC) method, different from conventional deep learning-based MRI reconstruction techniques. In addition to the maximum a posteriori estimate for the image, which can be obtained by maximizing the log-likelihood indirectly or directly, the minimum mean square error estimate and uncertainty maps can also be computed from those drawn samples. The data-driven Markov chains are constructed with the score-based generative model learned from a given image database and are independent of the forward operator that is used to model the k-space measurement.
Results
We numerically investigate the framework from these perspectives: (1) the interpretation of the uncertainty of the image reconstructed from undersampled k-space; (2) the effect of the number of noise scales used to train the generative models; (3) using a burn-in phase in MCMC sampling to reduce computation; (4) the comparison to conventional
-wavelet regularized reconstruction; (5) the transferability of learned information; and (6) the comparison to fastMRI challenge.
Conclusion
A framework is described that connects the diffusion process and advanced generative models with Markov chains. We demonstrate its flexibility in terms of contrasts and sampling patterns using advanced generative priors and the benefits of also quantifying the uncertainty for every pixel
Inverse reconstruction method for segmented multi-shot diffusion-weighted MRI with multiple coils.
Full text linkEach k-space segment in multishot diffusion-weighted MRI is affected by a different spatially varying phase which is caused by unavoidable motions and amplified by the diffusion-encoding gradients. A proper image reconstruction therefore requires phase maps for each segment. Such maps are commonly derived from two-dimensional navigators at relatively low resolution but do not offer robust solutions. For example, phase variations in diffusion-weighted MRI of the brain are often characterized by high spatial frequencies. To overcome this problem, an inverse reconstruction method for segmented multishot diffusion-weighted MRI is described that takes advantage of the full k-space data acquired from multiple receiver coils. First, the individual coil sensitivities are determined from the non-diffusion-weighted acquisitions by regularized nonlinear inversion. These coil sensitivities are then used to estimate accurate motion-associated phase maps for each segment by iterative linear inversion. Finally, the coil sensitivities and phase maps serve to reconstruct artifact-free images of the object by iterative linear inversion, taking advantage of the data of all segments. The efficiency of the new method is demonstrated for segmented diffusion-weighted stimulated echo acquisition mode MRI of the human brain. Magn Reson Med 62:1342-1348, 2009. (C) 2009 Wiley-Liss, Inc
Nonlinear inverse reconstruction for real-time MRI of the human heart using undersampled radial FLASH.
Full text linkA previously proposed nonlinear inverse reconstruction for auto-calibrated parallel imaging simultaneously estimates coil sensitivities and image content. This work exploits this property for real-time MRI, where coil sensitivities need to be dynamically adapted to the conditions generated by moving objects. The development comprises (i) an extension of the nonlinear inverse algorithm to non-Cartesian k-space encodings, (ii) its implementation on a graphical processing unit to reduce reconstruction times, and (iii) the use of a convolution-based iteration, which considerably simplifies the graphical processing unit implementation compared to a gridding technique. The method is validated for real-time MRI of the human heart at 3 T using radio frequency-spoiled radial FLASH (pulse repetition time/echo time = 2.0/1.3 ms, flip angle 8 degrees). The results demonstrate artifact-free reconstructions from only 65-85 spokes, with 256 oversampled data points. Acquisition times of 130-170 ms resulted in 29-38 frames per second for sliding window reconstructions (factor 5). While offline reconstructions required 1-2 sec, real-time applications with modified parameters and slightly lower image quality were achieved within 90 ms per graphical processing unit. Magn Reson Med 63:1456-1462, 2010. (C) 2010 Wiley-Liss, Inc
Method and device for reconstructing a sequence of magnetic resonance images
No full textA method for reconstructing a sequence of magnetic resonance (MR) images of an object under investigation, includes the steps of (a) providing a series of sets of image raw data including an image content of the MR images to be reconstructed, the image raw data being collected with the use of at least one radiofrequency receiver coil of a magnetic resonance imaging (MRI) device, wherein each set of image raw data includes a plurality of data samples being generated with a gradient-echo sequence, in particular a FLASH sequence, that spatially encodes an MRI signal received with the at least one radiofrequency receiver coil using a non-Cartesian k-space trajectory, each set of image raw data includes a set of homogeneously distributed lines in k-space with equivalent spatial frequency content, the lines of each set of image raw data cross the center of k-space and cover a continuous range of spatial frequencies, and the positions of the lines of each set of image raw data differ in successive sets of image raw data, and (b) subjecting the sets of image raw data to a regularized nonlinear inverse reconstruction process to provide the sequence of MR images, wherein each of the MR images is created by a simultaneous estimation of a sensitivity of the at least one receiver coil and the image content and in dependency on a difference between a current estimation of the sensitivity of the at least one receiver coil and the image content and a preceding estimation of the sensitivity of the at least one receiver coil and the image content
Assessment of deep learning segmentation for real-time free-breathing cardiac magnetic resonance imaging at rest and under exercise stress
No full textAbstract In recent years, a variety of deep learning networks for cardiac MRI (CMR) segmentation have been developed and analyzed. However, nearly all of them are focused on cine CMR under breathold. In this work, accuracy of deep learning methods is assessed for volumetric analysis (via segmentation) of the left ventricle in real-time free-breathing CMR at rest and under exercise stress. Data from healthy volunteers (n = 15) for cine and real-time free-breathing CMR at rest and under exercise stress were analyzed retrospectively. Exercise stress was performed using an ergometer in the supine position. Segmentations of two deep learning methods, a commercially available technique (comDL) and an openly available network (nnU-Net), were compared to a reference model created via the manual correction of segmentations obtained with comDL. Segmentations of left ventricular endocardium (LV), left ventricular myocardium (MYO), and right ventricle (RV) are compared for both end-systolic and end-diastolic phases and analyzed with Dice’s coefficient. The volumetric analysis includes the cardiac function parameters LV end-diastolic volume (EDV), LV end-systolic volume (ESV), and LV ejection fraction (EF), evaluated with respect to both absolute and relative differences. For cine CMR, nnU-Net and comDL achieve Dice’s coefficients above 0.95 for LV and 0.9 for MYO, and RV. For real-time CMR, the accuracy of nnU-Net exceeds that of comDL overall. For real-time CMR at rest, nnU-Net achieves Dice’s coefficients of 0.94 for LV, 0.89 for MYO, and 0.90 for RV and the mean absolute differences between nnU-Net and the reference are 2.9 mL for EDV, 3.5 mL for ESV, and 2.6% for EF. For real-time CMR under exercise stress, nnU-Net achieves Dice’s coefficients of 0.92 for LV, 0.85 for MYO, and 0.83 for RV and the mean absolute differences between nnU-Net and reference are 11.4 mL for EDV, 2.9 mL for ESV, and 3.6% for EF. Deep learning methods designed or trained for cine CMR segmentation can perform well on real-time CMR. For real-time free-breathing CMR at rest, the performance of deep learning methods is comparable to inter-observer variability in cine CMR and is usable for fully automatic segmentation. For real-time CMR under exercise stress, the performance of nnU-Net could promise a higher degree of automation in the future
- …
