Domain transformation learning for MR image reconstruction from dual domain input.

Medical images are acquired through diverse imaging systems, with each system employing specific image reconstruction techniques to transform sensor data into images. In MRI, sensor data (i.e., k-space data) is encoded in the frequency domain, and fully sampled k-space data is transformed into an image using the inverse Fourier Transform. However, in efforts to reduce acquisition time, k-space is often subsampled, necessitating a sophisticated image reconstruction method beyond a simple transform. The proposed approach addresses this challenge by training a model to learn domain transform, generating the final image directly from undersampled k-space input. Significantly, to improve the stability of reconstruction from randomly subsampled k-space data, folded images are incorporated as supplementary inputs in the dual-input ETER-net. Moreover, modifications are made to the formation of inputs for the bi-RNN stages to accommodate non-fixed k-space trajectories. Experimental validation, encompassing both regular and irregular sampling trajectories, validates the method's effectiveness. The results demonstrated superior performance, measured by PSNR, SSIM, and VIF, across acceleration factors of 4 and 8. In summary, the dual-input ETER-net emerges as an effective both regular and irregular sampling trajectories, and accommodating diverse acceleration factors.

An End-to-End Recurrent Neural Network for Radial MR Image Reconstruction.

Recent advances in deep learning have contributed greatly to the field of parallel MR imaging, where a reduced amount of k-space data are acquired to accelerate imaging time. In our previous work, we have proposed a deep learning method to reconstruct MR images directly from k-space data acquired with Cartesian trajectories. However, MRI utilizes various non-Cartesian trajectories, such as radial trajectories, with various numbers of multi-channel RF coils according to the purpose of an MRI scan. Thus, it is important for a reconstruction network to efficiently unfold aliasing artifacts due to undersampling and to combine multi-channel k-space data into single-channel data. In this work, a neural network named 'ETER-net' is utilized to reconstruct an MR image directly from k-space data acquired with Cartesian and non-Cartesian trajectories and multi-channel RF coils. In the proposed image reconstruction network, the domain transform network converts k-space data into a rough image, which is then refined in the following network to reconstruct a final image. We also analyze loss functions including adversarial and perceptual losses to improve the network performance. For experiments, we acquired k-space data at a 3T MRI scanner with Cartesian and radial trajectories to show the learning mechanism of the direct mapping relationship between the k-space and the corresponding image by the proposed network and to demonstrate the practical applications. According to our experiments, the proposed method showed satisfactory performance in reconstructing images from undersampled single- or multi-channel k-space data with reduced image artifacts. In conclusion, the proposed method is a deep-learning-based MR reconstruction network, which can be used as a unified solution for parallel MRI, where k-space data are acquired with various scanning trajectories.

Unsupervised anomaly detection in MR images using multicontrast information.

PURPOSE: Anomaly detection in magnetic resonance imaging (MRI) is to distinguish the relevant biomarkers of diseases from those of normal tissues. In this paper, an unsupervised algorithm is proposed for pixel-level anomaly detection in multicontrast MRI. METHODS: A deep neural network is developed, which uses only normal MR images as training data. The network has the two stages of feature generation and density estimation. For feature generation, relevant features are extracted from multicontrast MR images by performing contrast translation and dimension reduction. For density estimation, the distributions of the extracted features are estimated by using Gaussian mixture model (GMM). The two processes are trained to estimate normative distributions well presenting large normal datasets. In test phases, the proposed method can detect anomalies by measuring log-likelihood that a test sample belongs to the estimated normative distributions. RESULTS: The proposed method and its variants were applied to detect glioblastoma and ischemic stroke lesion. Comparison studies with six previous anomaly detection algorithms demonstrated that the proposed method achieved relevant improvements in quantitative and qualitative evaluations. Ablation studies by removing each module from the proposed framework validated the effectiveness of each proposed module. CONCLUSION: The proposed deep learning framework is an effective tool to detect anomalies in multicontrast MRI. The unsupervised approaches would have great potentials in detecting various lesions where annotated lesion data collection is limited.

A k-space-to-image reconstruction network for MRI using recurrent neural network.

PURPOSE: Reconstructing the images from undersampled k-space data are an ill-posed inverse problem. As a solution to this problem, we propose a method to reconstruct magnetic resonance (MR) images directly from k-space data using a recurrent neural network. METHODS: A novel neural network architecture named "ETER-net" is developed as a unified solution to reconstruct MR images from undersampled k-space data, where two bi-RNNs and convolutional neural network (CNN) are utilized to perform domain transformation and de-aliasing. To demonstrate the practicality of the proposed method, we conducted model optimization, cross-validation, and network pruning using in-house data from a 3T MRI scanner and public dataset called "FastMRI." RESULTS: The experimental results showed that the proposed method could be utilized for accurate image reconstruction from undersampled k-space data. The size of the proposed model was optimized and cross-validation was performed to show the robustness of the proposed method. For in-house dataset (R = 4), the proposed method provided nMSE = 1.09% and SSIM = 0.938. For "FastMRI" dataset, the proposed method provided nMSE = 1.05 % and SSIM = 0.931 for R = 4, and nMSE = 3.12 % and SSIM = 0.884 for R = 8. The performance of the pruned model trained the loss function including with L2 regularization was consistent for a pruning ratio of up to 70%. CONCLUSIONS: The proposed method is an end-to-end MR image reconstruction method based on recurrent neural networks. It performs direct mapping of the input k-space data and the reconstructed images, operating as a unified solution that is applicable to various scanning trajectories.

Self-gated cardiac cine imaging using phase information.

PURPOSE: To obtain multiphase cardiac cine images with high resolution, a novel self-gating method for both cardiac and respiratory motions is proposed. METHODS: The proposed method uses the phase of projection data obtained from a separate axial slice to measure cardiac and respiratory motion, after the acquisition of every k-space line in the image plane. Cardiac motion is estimated from the phase of the projection data passing through the aorta, which is amplified by superior-inferior directional bipolar gradients, whereas respiratory motion is estimated from the phase of the left-right directional projection data of the abdomen. To verify the proposed self-gating method, a simulation and in vivo steady state free precession cardiac imaging were performed. RESULTS: The proposed method provides high resolution multiphase cardiac cine images. Using the proposed self-gating method, the phase variation of the projection data offers information about cardiac and respiratory motions that is equivalent to external gating devices. CONCLUSION: The proposed method can capture time-resolved cardiac and respiratory motion from the phase information of the projection data. Because the projection data is obtained from a separate gating slice, the self-gating signals are not affected by imaging planes. Magn Reson Med 77:1216-1222, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Multi-slice imAGe generation using intra-slice paraLLel imaging and Inter-slice shifting (MAGGULLI).

For acceleration of imaging time, multi-band imaging techniques (e.g. CAIPIRINHA) use the sensitivity differences of the multi-channel RF coils in the slice selection direction. To more effectively utilize the RF coil characteristics than the conventional multi-band imaging techniques, we propose a new imaging technique, called multi-slice image generation using intra-slice parallel imaging and inter-slice shifting (MAGGULLI). The proposed technique used an inter-slice shifting gradient in slice selection direction to make multi-slice images shift in the frequency encoding direction. Thus, aliasing caused by sub-sampling in the phase encoding direction is orthogonal to that by multi-band imaging with the inter-slice shifting, both of which are resolved by using the sensitivity information of the RF coil. Phantom and in vivo imaging experiments for the acceleration factors up to 10 demonstrate that the quality of the images reconstructed by MAGGULLI are better than that of CAIPIRINHA for high acceleration factors in the qualitative and quantitative analysis.

Isotropic diffusion weighting for measurement of a high-resolution apparent diffusion coefficient map using a single radial scan in MRI.

This work proposes an isotropic diffusion weighting method for a high-resolution diffusion-weighted image and for a high-resolution apparent diffusion coefficient (ADC) map using a single radial scan in MRI. By using a conventional radial imaging technique, a high-resolution diffusion-weighted (DW) image can be obtained at the cost of a long imaging time. To reduce the imaging time, the proposed method acquires a DW image by altering the diffusion gradient directions for each radial spoke. The acquisition order and directions of the diffusion gradients for an accurate DW image and an ADC map are also proposed by modifying the golden angle ratio in 3D space. In addition, an individual-direction diffusion-weighted (id-DW) image can also be obtained by a diffusion gradient direction, which is one of the multiple directions used in isotropic diffusion weighting. Computer simulations and experiment results show that the proposed method is more accurate and faster than the conventional radial diffusion-weighted imaging. This study suggests that the proposed isotropic diffusion-weighted imaging can be used to obtain a DW image and a high-resolution ADC map accurately in a single radial scan, while reducing the artifacts caused by the diffusion anisotropy, compared to the diffusion-weighted echo-planar-imaging.