Varying undersampling directions for accelerating multiple acquisition magnetic resonance imaging.

In this study, we propose a new sampling strategy for efficiently accelerating multiple acquisition MRI. The new sampling strategy is to obtain data along different phase-encoding directions across multiple acquisitions. The proposed sampling strategy was evaluated in multicontrast MR imaging (T1, T2, proton density) and multiple phase-cycled (PC) balanced steady-state free precession (bSSFP) imaging by using convolutional neural networks with central and random sampling patterns. In vivo MRI acquisitions as well as a public database were used to test the concept. Based on both visual inspection and quantitative analysis, the proposed sampling strategy showed better performance than sampling along the same phase-encoding direction in both multicontrast MR imaging and multiple PC-bSSFP imaging, regardless of sampling pattern (central, random) or datasets (public, retrospective and prospective in vivo). For the prospective in vivo applications, acceleration was performed by sampling along different phase-encoding directions at the time of acquisition with a conventional rectangular field of view, which demonstrated the advantage of the proposed sampling strategy in the real environment. Preliminary trials on compressed sensing (CS) also demonstrated improvement of CS with the proposed idea. Sampling along different phase-encoding directions across multiple acquisitions is advantageous for accelerating multiacquisition MRI, irrespective of sampling pattern or datasets, with further improvement through transfer learning.

Artificial neural network for Slice Encoding for Metal Artifact Correction (SEMAC) MRI.

PURPOSE: To develop new artificial neural networks (ANNs) to accelerate slice encoding for metal artifact correction (SEMAC) MRI. METHODS: Eight titanium phantoms and 77 patients after brain tumor surgery involving metallic neuro-plating instruments were scanned using SEMAC at a 3T Skyra scanner. For the phantoms, proton-density, T1-, and T2-weighted images were acquired for developing both multilayer perceptron (MLP) and convolutional neural network (CNN). For the patients, T2-weighted images were acquired for developing CNN. All networks were trained with the SEMAC factor 4 or 6 as input and the factor 12 as label, yielding an acceleration factor of 3 or 2. Performance of the CNN model was compared against parallel imaging and compressed sensing on the phantom datasets. Two extra T1-weighted in vivo sets were acquired to investigate generalizability of the models to different contrasts. RESULTS: Both multilayer perceptron and CNN provided artifact-suppressed images better than the input images and comparable to the label images visually and quantitatively, a trend observable regardless of input SEMAC factor and image type (P < .01). CNN suppressed the artifacts better than multilayer perceptron, parallel imaging, and compressed sensing (P < .01). Tests on the patient datasets demonstrated clear metal artifact suppression visually and quantitatively (P < .01). Tests on T1 datasets also demonstrated clear visual metal artifact suppression. CONCLUSION: Our study introduced a new effective way of artificial neural networks to accelerate SEMAC MRI while maintaining the comparable quality of metal artifact suppression. Application on the preliminary patient datasets proved the feasibility in clinical usage, which warrants further investigation.

Artifact-suppressed optimal three-dimensional T1 - and T2 *-weighted dual-echo imaging.

PURPOSE: To develop a new artifact-suppressed optimal three-dimensional (3D) T1 - and T2 *-weighted dual-echo imaging. METHODS: We optimized flip angles for 3D T1 - and T2 *-weighted imaging by conventional dual-echo in vivo experiments and computer simulations, and then implemented a dual-echo sequence with an echo-specific k-space reordering scheme to satisfy the optimal flip angles for both T1 and T2 * contrast. We also proposed two strategies to suppress ringing artifacts induced by the abrupt flip angle jumps in the proposed dual echo sequence: (i) implementing smooth transition regions and (ii) discarding the k-space regions of the abrupt flip angle jumps as dummy phase-encoding steps. RESULTS: The optimal flip angles measured from experiments were different between T1 - and T2 *-weighted contrast, in agreement with simulations. The echo-specific k-space reordered dual-echo sequence showed optimal T1 and T2 * contrast simultaneously, but also showed ringing artifacts because of high flip-angle changes between k-space regions. The two proposed strategies effectively suppressed the ringing artifacts. CONCLUSION: The proposed 3D dual-echo sequence provided optimal T1 and T2 * contrast simultaneously with no artifacts and thus is potentially applicable to routine clinical applications for simultaneous high resolution T1 - and T2 *-weighted imaging. Magn Reson Med 76:1504-1511, 2016. © 2015 International Society for Magnetic Resonance in Medicine.

Reconstruction of multicontrast MR images through deep learning.

PURPOSE: Magnetic resonance (MR) imaging with a long scan time can lead to degraded images due to patient motion, patient discomfort, and increased costs. For these reasons, the role of rapid MR imaging is important. In this study, we propose the joint reconstruction of multicontrast brain MR images from down-sampled data to accelerate the data acquisition process using a novel deep-learning network. METHODS: Twenty-one healthy volunteers (female/male = 7/14, age = 26 ± 4 yr, range 22-35 yr) and 16 postoperative patients (female/male = 7/9, age = 49 ± 9 yr, range 37-62 yr) were scanned on a 3T whole-body scanner for prospective and retrospective studies, respectively, using both T1-weighted spin-echo (SE) and T2-weighted fast spin-echo (FSE) sequences. We proposed a network which we term "X-net" to reconstruct both T1- and T2-weighted images from down-sampled images as well as a network termed "Y-net" which reconstructs T2-weighted images from highly down-sampled T2-weighted images and fully sampled T1-weighted images. Both X-net and Y-net are composed of two concatenated subnetworks. We investigate optimal sampling patterns, the optimal patch size for augmentation, and the optimal acceleration factors for network training. An additional Y-net combined with a generative adversarial network (GAN) was also implemented and tested to investigate the effects of the GAN on the Y-net performance. Single- and joint-reconstruction parallel-imaging and compressed-sensing algorithms along with a conventional U-net were also tested and compared with the proposed networks. For this comparison, the structural similarity (SSIM), normalized mean square error (NMSE), and Fréchet inception distance (FID) were calculated between the outputs of the networks and fully sampled images. The statistical significance of the performance was evaluated by assessing the interclass correlation and in paired t-tests. RESULTS: The outputs from the two concatenated subnetworks were closer to the fully sampled images compared to those from one subnetwork, with this result showing statistical significance. Uniform down-sampling led to a statically significant improvement in the image quality compared to random or central down-sampling patterns. In addition, the proposed networks provided higher SSIM and NMSE values than U-net, compressed-sensing, and parallel-imaging algorithms, all at statistically significant levels. The GAN-based Y-net showed a better FID and more realistic images compared to a non-GAN-based Y-net. The performance capabilities of the networks were similar between normal subjects and patients. CONCLUSIONS: The proposed X-net and Y-net effectively reconstructed full images from down-sampled images, outperforming the conventional parallel-imaging, compressed-sensing and U-net methods and providing more realistic images in combination with a GAN. The developed networks potentially enable us to accelerate multicontrast anatomical MR imaging in routine clinical studies including T1-and T2-weighted imaging.

Mapping cerebral perfusion from time-resolved contrast-enhanced MR angiographic data.

A common method to acquire both perfusion and angiographic information is to have separate MRI scans for each information. In this study, we propose to achieve the goal by deriving perfusion parameters, specifically cerebral blood volume (CBV) and Tmax, from time-resolved contrast-enhanced magnetic resonance angiography (CE-MRA). Both CE-MRA and DSC-MRI were performed on seven subjects with a diagnosed ischemic stroke. Concentration functions from CE-MRA were modeled using a modified gamma-variate function to appreciate the full first-pass transition of the tracer bolus. Perfusion parameters were calculated using concentration function derived from each imaging method and were compared to each other both visually and quantitatively by means of correlation studies. CBV and Tmax maps generally showed good agreement between the two methods. This study proved the concept of using time-resolved CE-MRA as both vascular imaging and tissue perfusion mapping while using a single injection of contrast agent, potentially reducing cost and improving patient safety and comfort.

Simultaneous Variable-Slab Dual-Echo TOF MR Angiography and Susceptibility-Weighted Imaging.

In this paper, we propose a new 3-D dual-echo method for simultaneous multislab time-of-flight MR angiography (TOF MRA) and single-slab susceptibility-weighted imaging (SWI). The previous echo-specific k-space reordering scheme for compatible dual-echo arteriovenography (CODEA) was advanced to applying excitation RF pulses for multiple thin slabs and a single thick slab to the first (TOF MRA) and second (SWI) echoes, respectively. Single-slab CODEA and multislab CODEA (fixed-slab CODEAs) were additionally acquired as comparison reference to the proposed variable-slab CODEA. Parallel imaging was also tested for feasibility of accelerating the proposed method. TOF MRA and SWI from the proposed variable-slab CODEA were visually and quantitatively comparable to multislab TOF MRA and single-slab SWI, respectively, separately acquired from the fixed-slab CODEAs. The parallel imaging reduced the scan time from 10.3 to 5.6 min. Furthermore, the proposed variable-slab approach improved the vessel continuities at slab boundaries of TOF MRA for CODEA as well as for the conventional single echo method. The proposed variable-slab CODEA provided multislab TOF MRA and single-slab SWI simultaneously in a clinically reasonable scan time of ~5 min with minimal impact on image qualities, while suppressing slab boundary artifacts in TOF MRA.

Improving resolution of MR images with an adversarial network incorporating images with different contrast.

PURPOSE: The routine MRI scan protocol consists of multiple pulse sequences that acquire images of varying contrast. Since high frequency contents such as edges are not significantly affected by image contrast, down-sampled images in one contrast may be improved by high resolution (HR) images acquired in another contrast, reducing the total scan time. In this study, we propose a new deep learning framework that uses HR MR images in one contrast to generate HR MR images from highly down-sampled MR images in another contrast. MATERIALS AND METHODS: The proposed convolutional neural network (CNN) framework consists of two CNNs: (a) a reconstruction CNN for generating HR images from the down-sampled images using HR images acquired with a different MRI sequence and (b) a discriminator CNN for improving the perceptual quality of the generated HR images. The proposed method was evaluated using a public brain tumor database and in vivo datasets. The performance of the proposed method was assessed in tumor and no-tumor cases separately, with perceptual image quality being judged by a radiologist. To overcome the challenge of training the network with a small number of available in vivo datasets, the network was pretrained using the public database and then fine-tuned using the small number of in vivo datasets. The performance of the proposed method was also compared to that of several compressed sensing (CS) algorithms. RESULTS: Incorporating HR images of another contrast improved the quantitative assessments of the generated HR image in reference to ground truth. Also, incorporating a discriminator CNN yielded perceptually higher image quality. These results were verified in regions of normal tissue as well as tumors for various MRI sequences from pseudo k-space data generated from the public database. The combination of pretraining with the public database and fine-tuning with the small number of real k-space datasets enhanced the performance of CNNs in in vivo application compared to training CNNs from scratch. The proposed method outperformed the compressed sensing methods. CONCLUSIONS: The proposed method can be a good strategy for accelerating routine MRI scanning.

Mapping blood flow directionality in the human brain.

Diffusion properties of tissue are often expressed on the basis of directional variance, i.e., diffusion tensor imaging. In comparison, common perfusion-weighted imaging such as arterial spin labeling yields perfusion in a scalar quantity. The purpose of this study was to test the feasibility of mapping cerebral blood flow directionality using alternate ascending/descending directional navigation (ALADDIN), a recently-developed arterial spin labeling technique with sensitivity to blood flow directions. ALADDIN was applied along 3 orthogonal directions to assess directional blood flow in a vector form and also along 6 equally-spaced directions to extract blood flow tensor matrix (P) based on a blood flow ellipsoid model. Tensor elements (eigenvalues, eigenvectors, etc) were calculated to investigate characteristics of the blood flow tensor, in comparison with time-of-flight MR angiogram. While the directions of the main eigenvectors were heterogeneous throughout the brain, regional clusters of blood flow directionality were reproducible across subjects. The technique could show heterogeneous blood flow directionality within and around brain tumor, which was different from that of the contralateral normal side. The proposed method is deemed to provide information of blood flow directionality, which has not been demonstrated before. The results warrant further studies to assess changes in the directionality map as a function of scan parameters, to understand the signal sources, to investigate the possibility of mapping local blood perfusion directionality, and to evaluate its usefulness for clinical diagnosis.