提高磁共振成像速度是磁共振领域待解决的问题，采用多线圈扫描是一种常用加速方式．但在加速因子大于4时，传统压缩感知磁共振（CS-MRI）重建算法得到的图像质量不高．为此，本文提出一种基于物理模型的ISTAVS-Net多线圈MRI图像重建方法．将ISTA算法与VS-Net网络拆分思想相结合，提出ISTAVS算法，并展开成ISTAVS-Net网络．将每步迭代与网络模块结合，比黑盒U-Net网络的可解释性更强．网络中引入残差机制，增加了网络的非线性表达能力和稳定性. 稀疏变换、收缩阈值以及正则化参数在训练中自动学习，提高了重建的灵活性．采用Globus膝关节数据集的测试结果表明不同加速因子下ISTAVS-Net网络效果均优于传统的 L1-ESPIRiT和ISTA迭代算法，图像质量和性能指标比U-Net、ISTA-Net+和VS-Net网络提升明显，在高加速因子下对组织细节恢复能力更强．该网络鲁棒性强，更适合对临床扫描数据进行快速高质量重建，可拓宽MRI应用范围．

New theoretical and practical concepts are presented for considerably enhancing the performance of magnetic resonance imaging (MRI) by means of arrays of multiple receiver coils. Sensitivity encoding (SENSE) is based on the fact that receiver sensitivity generally has an encoding effect complementary to Fourier preparation by linear field gradients. Thus, by using multiple receiver coils in parallel scan time in Fourier imaging can be considerably reduced. The problem of image reconstruction from sensitivity encoded data is formulated in a general fashion and solved for arbitrary coil configurations and k-space sampling patterns. Special attention is given to the currently most practical case, namely, sampling a common Cartesian grid with reduced density. For this case the feasibility of the proposed methods was verified both in vitro and in vivo. Scan time was reduced to one-half using a two-coil array in brain imaging. With an array of five coils double-oblique heart images were obtained in one-third of conventional scan time. Magn Reson Med 42:952-962, 1999.Copyright 1999 Wiley-Liss, Inc.

In this study, a novel partially parallel acquisition (PPA) method is presented which can be used to accelerate image acquisition using an RF coil array for spatial encoding. This technique, GeneRalized Autocalibrating Partially Parallel Acquisitions (GRAPPA) is an extension of both the PILS and VD-AUTO-SMASH reconstruction techniques. As in those previous methods, a detailed, highly accurate RF field map is not needed prior to reconstruction in GRAPPA. This information is obtained from several k-space lines which are acquired in addition to the normal image acquisition. As in PILS, the GRAPPA reconstruction algorithm provides unaliased images from each component coil prior to image combination. This results in even higher SNR and better image quality since the steps of image reconstruction and image combination are performed in separate steps. After introducing the GRAPPA technique, primary focus is given to issues related to the practical implementation of GRAPPA, including the reconstruction algorithm as well as analysis of SNR in the resulting images. Finally, in vivo GRAPPA images are shown which demonstrate the utility of the technique.Copyright 2002 Wiley-Liss, Inc.

In this work an iterative reconstruction method based on generalized autocalibrating partially parallel acquisitions (GRAPPA) reconstruction is introduced. In the new method the reconstructed lines are used to reestimate and refine the weights from all the acquired data by applying the GRAPPA procedure iteratively with regularization. Both phantom and in vivo MRI experiments demonstrated that, compared to GRAPPA, the iterative approach reduces parallel imaging artifacts and permits high-quality image reconstruction with a relatively small number of calibration lines and slight changes of GRAPPA weights.

A new approach for utilizing conjugate k-space symmetry for improved parallel MRI performance is presented. By generating virtual coils containing conjugate symmetric k-space signals from actual coils, additional image- and coil-phase information can be incorporated into the reconstruction process for parallel acquisition techniques. In that way the reconstruction conditions are improved, resulting in less noise enhancement. In particular in combination with generalized autocalibrating partially parallel acquisitions (GRAPPA), the virtual coil concept represents a practical approach since no explicit spatial phase information is required. In addition, the influence of phase variations originating from the complex receiver coils as well as from the background is investigated. It is shown that there exist background phase distributions yielding an optimized pMRI reconstruction.

本文提出一种基于虚拟共轭线圈（Virtual Coil Concept，VCC）技术和k空间插值鲁棒人工神经网络（Robust Artificial-neural-networks for k-space Interpolation，RAKI）的图像重建方法，用于磁共振多层同时激发成像（Simultaneous Multi-Slice imaging，SMS），该方法能够有效提升重建图像的质量，被命名为VIRGINIA（VIRtual conjuGate coIls Neural-networks InterpolAtion）.为了得到更高质量的SMS图像，本文提出的VIRGINIA方法利用磁共振线圈数据的复数共轭对称性质扩展了SMS所获取的多通道数据，并将扩展后的数据用于RAKI网络的训练，利用训练后的网络实现高质量的SMS图像重建.本文将VIRGINIA方法和其他SMS图像重建方法（RAKI和Slice-GRAPPA方法）进行了对比，并采用结构相似指数（Structural Similarity Index，SSIM）、峰值信噪比（Peak Signal-to-Noise Ratio，PSNR）和均方根误差（Root Mean Square Error，RMSE）对不同方法的重建图像进行了量化对比分析.结果显示，在相同的SMS加速倍数下，使用VIRGINIA方法进行重建的图像质量均好于RAKI方法，且远好于传统Slice-GRAPPA方法.

To develop an improved k-space reconstruction method using scan-specific deep learning that is trained on autocalibration signal (ACS) data.Robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction trains convolutional neural networks on ACS data. This enables nonlinear estimation of missing k-space lines from acquired k-space data with improved noise resilience, as opposed to conventional linear k-space interpolation-based methods, such as GRAPPA, which are based on linear convolutional kernels.The training algorithm is implemented using a mean square error loss function over the target points in the ACS region, using a gradient descent algorithm. The neural network contains 3 layers of convolutional operators, with 2 of these including nonlinear activation functions. The noise performance and reconstruction quality of the RAKI method was compared with GRAPPA in phantom, as well as in neurological and cardiac in vivo data sets.Phantom imaging shows that the proposed RAKI method outperforms GRAPPA at high (≥4) acceleration rates, both visually and quantitatively. Quantitative cardiac imaging shows improved noise resilience at high acceleration rates (rate 4:23% and rate 5:48%) over GRAPPA. The same trend of improved noise resilience is also observed in high-resolution brain imaging at high acceleration rates.The RAKI method offers a training database-free deep learning approach for MRI reconstruction, with the potential to improve many existing reconstruction approaches, and is compatible with conventional data acquisition protocols.© 2018 International Society for Magnetic Resonance in Medicine.

Deep learning has become a popular tool for medical image analysis, but the limited availability of training data remains a major challenge, particularly in the medical field where data acquisition can be costly and subject to privacy regulations. Data augmentation techniques offer a solution by artificially increasing the number of training samples, but these techniques often produce limited and unconvincing results. To address this issue, a growing number of studies have proposed the use of deep generative models to generate more realistic and diverse data that conform to the true distribution of the data. In this review, we focus on three types of deep generative models for medical image augmentation: variational autoencoders, generative adversarial networks, and diffusion models. We provide an overview of the current state of the art in each of these models and discuss their potential for use in different downstream tasks in medical imaging, including classification, segmentation, and cross-modal translation. We also evaluate the strengths and limitations of each model and suggest directions for future research in this field. Our goal is to provide a comprehensive review about the use of deep generative models for medical image augmentation and to highlight the potential of these models for improving the performance of deep learning algorithms in medical image analysis.

A magnetic resonance imaging (MRI) exam typically consists of the acquisition of multiple MR pulse sequences, which are required for a reliable diagnosis. With the rise of generative deep learning models, approaches for the synthesis of MR images are developed to either synthesize additional MR contrasts, generate synthetic data, or augment existing data for AI training. While current generative approaches allow only the synthesis of specific sets of MR contrasts, we developed a method to generate synthetic MR images with adjustable image contrast. Therefore, we trained a generative adversarial network (GAN) with a separate auxiliary classifier (AC) network to generate synthetic MR knee images conditioned on various acquisition parameters (repetition time, echo time, and image orientation). The AC determined the repetition time with a mean absolute error (MAE) of 239.6 ms, the echo time with an MAE of 1.6 ms, and the image orientation with an accuracy of 100%. Therefore, it can properly condition the generator network during training. Moreover, in a visual Turing test, two experts mislabeled 40.5% of real and synthetic MR images, demonstrating that the image quality of the generated synthetic and real MR images is comparable. This work can support radiologists and technologists during the parameterization of MR sequences by previewing the yielded MR contrast, can serve as a valuable tool for radiology training, and can be used for customized data generation to support AI training.

时空编码（SPEN）磁共振成像（MRI）是一种超快MRI技术，通过该技术采集获得的原始图像空间分辨率较低，需要基于序列物理原理进行超分辨率重建以提高其原始图像的分辨率，而现有的基于深度学习SPEN超分辨率重建算法在提取图像像素长距离依赖关系上的能力有限.为了解决此问题，本文提出了一种基于Transformer的SPEN MRI超分辨率重建算法.该算法采用编码器-解码器结构，并引入Transformer模块以提取特征图的局部上下文信息和长距离依赖关系.实验结果表明，本文所提的重建算法可以在不增加额外采样点的情况下从SPEN低分辨率图像中重建出高空间分辨率、无混叠伪影的超分辨率图像.与现有的超分辨率算法相比，本文提出的算法在临床前以及临床数据集上都取得了更好的重建效果.

We describe methods for simultaneously acquiring and subsequently combining data from a multitude of closely positioned NMR receiving coils. The approach is conceptually similar to phased array radar and ultrasound and hence we call our techniques the "NMR phased array." The NMR phased array offers the signal-to-noise ratio (SNR) and resolution of a small surface coil over fields-of-view (FOV) normally associated with body imaging with no increase in imaging time. The NMR phased array can be applied to both imaging and spectroscopy for all pulse sequences. The problematic interactions among nearby surface coils is eliminated (a) by overlapping adjacent coils to give zero mutual inductance, hence zero interaction, and (b) by attaching low input impedance preamplifiers to all coils, thus eliminating interference among next nearest and more distant neighbors. We derive an algorithm for combining the data from the phased array elements to yield an image with optimum SNR. Other techniques which are easier to implement at the cost of lower SNR are explored. Phased array imaging is demonstrated with high resolution (512 x 512, 48-cm FOV, and 32-cm FOV) spin-echo images of the thoracic and lumbar spine. Data were acquired from four-element linear spine arrays, the first made of 12-cm square coils and the second made of 8-cm square coils. When compared with images from a single 15 x 30-cm rectangular coil and identical imaging parameters, the phased array yields a 2X and 3X higher SNR at the depth of the spine (approximately 7 cm).