Deep Generative Adversarial Neural Networks for Compressive Sensing MRI

Generative adversarial network based regularized image reconstruction for PET

Positron emission tomography (PET) is an ill-posed inverse problem and suffers high noise due to limited number of detected events. Prior information can be used to improve the quality of reconstructed PET images. Deep neural networks have also been applied to regularized image reconstruction. One method is to use a pretrained denoising neural network to represent the PET image and to perform a constrained maximum likelihood estimation. In this work, we propose to use a generative adversarial network (GAN) to further improve the network performance. We also modify the objective function to include a data-matching term on the network input. Experimental studies using computer-based Monte Carlo simulations and real patient datasets demonstrate that the proposed method leads to noticeable improvements over the kernel-based and U-net-based regularization methods in terms of lesion contrast recovery versus background noise trade-offs.

Subsampled brain MRI reconstruction by generative adversarial neural networks

A main challenge in magnetic resonance imaging (MRI) is speeding up scan time. Beyond improving patient experience and reducing operational costs, faster scans are essential for time-sensitive imaging, such as fetal, cardiac, or functional MRI, where temporal resolution is important and target movement is unavoidable, yet must be reduced. Current MRI acquisition methods speed up scan time at the expense of lower spatial resolution and costlier hardware. We introduce a practical, software-only framework, based on deep learning, for accelerating MRI acquisition, while maintaining anatomically meaningful imaging. This is accomplished by MRI subsampling followed by estimating the missing k-space samples via generative adversarial neural networks. A generator-discriminator interplay enables the introduction of an adversarial cost in addition to fidelity and image-quality losses used for optimizing the reconstruction. Promising reconstruction results are obtained from feasible sampling patterns of up to a fivefold acceleration of diverse brain MRIs, from a large publicly available dataset of healthy adult scans as well as multimodal acquisitions of multiple sclerosis patients and dynamic contrast-enhanced MRI (DCE-MRI) sequences of stroke and tumor patients. Clinical usability of the reconstructed MRI scans is assessed by performing either lesion or healthy tissue segmentation and comparing the results to those obtained by using the original, fully sampled images. Reconstruction quality and usability of the DCE-MRI sequences is demonstrated by calculating the pharmacokinetic (PK) parameters. The proposed MRI reconstruction approach is shown to outperform state-of-the-art methods for all datasets tested in terms of the peak signal-to-noise ratio (PSNR), the structural similarity index (SSIM), as well as either the mean squared error (MSE) with respect to the PK parameters, calculated for the fully sampled DCE-MRI sequences, or the segmentation compatibility, measured in terms of Dice scores and Hausdorff distance. The code is available on GitHub.

Fidelity imposed network edit (FINE) for solving ill-posed image reconstruction

Deep learning (DL) is increasingly used to solve ill-posed inverse problems in medical imaging, such as reconstruction from noisy and/or incomplete data, as DL offers advantages over conventional methods that rely on explicit image features and hand engineered priors. However, supervised DL-based methods may achieve poor performance when the test data deviates from the training data, for example, when it has pathologies not encountered in the training data. Furthermore, DL-based image reconstructions do not always incorporate the underlying forward physical model, which may improve performance. Therefore, in this work we introduce a novel approach, called fidelity imposed network edit (FINE), which modifies the weights of a pre-trained reconstruction network for each case in the testing dataset. This is achieved by minimizing an unsupervised fidelity loss function that is based on the forward physical model. FINE is applied to two important inverse problems in neuroimaging: quantitative susceptibility mapping (QSM) and under-sampled image reconstruction in MRI. Our experiments demonstrate that FINE can improve reconstruction accuracy.

MRI reconstruction using deep Bayesian estimation

PURPOSE

To develop a deep learning-based Bayesian estimation for MRI reconstruction.

METHODS

We modeled the MRI reconstruction problem with Bayes's theorem, following the recently proposed PixelCNN++ method. The image reconstruction from incomplete k-space measurement was obtained by maximizing the posterior possibility. A generative network was utilized as the image prior, which was computationally tractable, and the k-space data fidelity was enforced by using an equality constraint. The stochastic backpropagation was utilized to calculate the descent gradient in the process of maximum a posterior, and a projected subgradient method was used to impose the equality constraint. In contrast to the other deep learning reconstruction methods, the proposed one used the likelihood of prior as the training loss and the objective function in reconstruction to improve the image quality.

RESULTS

The proposed method showed an improved performance in preserving image details and reducing aliasing artifacts, compared with GRAPPA, ℓ 1 -ESPRiT, model-based deep learning architecture for inverse problems (MODL), and variational network (VN), last two were state-of-the-art deep learning reconstruction methods. The proposed method generally achieved more than 3 dB peak signal-to-noise ratio improvement for compressed sensing and parallel imaging reconstructions compared with the other methods.

CONCLUSIONS

The Bayesian estimation significantly improved the reconstruction performance, compared with the conventional ℓ 1 -sparsity prior in compressed sensing reconstruction tasks. More importantly, the proposed reconstruction framework can be generalized for most MRI reconstruction scenarios.

DIAGNOSTIC IMAGE QUALITY ASSESSMENT AND CLASSIFICATION IN MEDICAL IMAGING: OPPORTUNITIES AND CHALLENGES

Magnetic Resonance Imaging (MRI) suffers from several artifacts, the most common of which are motion artifacts. These artifacts often yield images that are of non-diagnostic quality. To detect such artifacts, images are prospectively evaluated by experts for their diagnostic quality, which necessitates patient-revisits and rescans whenever non-diagnostic quality scans are encountered. This motivates the need to develop an automated framework capable of accessing medical image quality and detecting diagnostic and non-diagnostic images. In this paper, we explore several convolutional neural network-based frameworks for medical image quality assessment and investigate several challenges therein.

Utility of deep learning super-resolution in the context of osteoarthritis MRI biomarkers

BACKGROUND

Super-resolution is an emerging method for enhancing MRI resolution; however, its impact on image quality is still unknown.

PURPOSE

To evaluate MRI super-resolution using quantitative and qualitative metrics of cartilage morphometry, osteophyte detection, and global image blurring.

STUDY TYPE

Retrospective.

POPULATION

In all, 176 MRI studies of subjects at varying stages of osteoarthritis.

FIELD STRENGTH/SEQUENCE

Original-resolution 3D double-echo steady-state (DESS) and DESS with 3× thicker slices retrospectively enhanced using super-resolution and tricubic interpolation (TCI) at 3T.

ASSESSMENT

A quantitative comparison of femoral cartilage morphometry was performed for the original-resolution DESS, the super-resolution, and the TCI scans in 17 subjects. A reader study by three musculoskeletal radiologists assessed cartilage image quality, overall image sharpness, and osteophytes incidence in all three sets of scans. A referenceless blurring metric evaluated blurring in all three image dimensions for the three sets of scans.

STATISTICAL TESTS

Mann-Whitney U-tests compared Dice coefficients (DC) of segmentation accuracy for the DESS, super-resolution, and TCI images, along with the image quality readings and blurring metrics. Sensitivity, specificity, and diagnostic odds ratio (DOR) with 95% confidence intervals compared osteophyte detection for the super-resolution and TCI images, with the original-resolution as a reference.

RESULTS

DC for the original-resolution (90.2 ± 1.7%) and super-resolution (89.6 ± 2.0%) were significantly higher (P < 0.001) than TCI (86.3 ± 5.6%). Segmentation overlap of super-resolution with the original-resolution (DC = 97.6 ± 0.7%) was significantly higher (P < 0.0001) than TCI overlap (DC = 95.0 ± 1.1%). Cartilage image quality for sharpness and contrast levels, and the through-plane quantitative blur factor for super-resolution images, was significantly (P < 0.001) better than TCI. Super-resolution osteophyte detection sensitivity of 80% (76-82%), specificity of 93% (92-94%), and DOR of 32 (22-46) was significantly higher (P < 0.001) than TCI sensitivity of 73% (69-76%), specificity of 90% (89-91%), and DOR of 17 (13-22).

DATA CONCLUSION

Super-resolution appears to consistently outperform naïve interpolation and may improve image quality without biasing quantitative biomarkers.

LEVEL OF EVIDENCE

2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:768-779.

ADMM-CSNet: A Deep Learning Approach for Image Compressive Sensing

Compressive sensing (CS) is an effective technique for reconstructing image from a small amount of sampled data. It has been widely applied in medical imaging, remote sensing, image compression, etc. In this paper, we propose two versions of a novel deep learning architecture, dubbed as ADMM-CSNet, by combining the traditional model-based CS method and data-driven deep learning method for image reconstruction from sparsely sampled measurements. We first consider a generalized CS model for image reconstruction with undetermined regularizations in undetermined transform domains, and then two efficient solvers using Alternating Direction Method of Multipliers (ADMM) algorithm for optimizing the model are proposed. We further unroll and generalize the ADMM algorithm to be two deep architectures, in which all parameters of the CS model and the ADMM algorithm are discriminatively learned by end-to-end training. For both applications of fast CS complex-valued MR imaging and CS imaging of real-valued natural images, the proposed ADMM-CSNet achieved favorable reconstruction accuracy in fast computational speed compared with the traditional and the other deep learning methods.

A deep error correction network for compressed sensing MRI

BACKGROUND

CS-MRI (compressed sensing for magnetic resonance imaging) exploits image sparsity properties to reconstruct MRI from very few Fourier k-space measurements. Due to imperfect modelings in the inverse imaging, state-of-the-art CS-MRI methods tend to leave structural reconstruction errors. Compensating such errors in the reconstruction could help further improve the reconstruction quality.

RESULTS

In this work, we propose a DECN (deep error correction network) for CS-MRI. The DECN model consists of three parts, which we refer to as modules: a guide, or template, module, an error correction module, and a data fidelity module. Existing CS-MRI algorithms can serve as the template module for guiding the reconstruction. Using this template as a guide, the error correction module learns a CNN (convolutional neural network) to map the k-space data in a way that adjusts for the reconstruction error of the template image. We propose a deep error correction network. Our experimental results show the proposed DECN CS-MRI reconstruction framework can considerably improve upon existing inversion algorithms by supplementing with an error-correcting CNN.

CONCLUSIONS

In the proposed a deep error correction framework, any off-the-shelf CS-MRI algorithm can be used as template generation. Then a deep neural network is used to compensate reconstruction errors. The promising experimental results validate the effectiveness and utility of the proposed framework.

A deep adversarial variational autoencoder model for dimensionality reduction in single-cell RNA sequencing analysis

BACKGROUND

Single-cell RNA sequencing (scRNA-seq) is an emerging technology that can assess the function of an individual cell and cell-to-cell variability at the single cell level in an unbiased manner. Dimensionality reduction is an essential first step in downstream analysis of the scRNA-seq data. However, the scRNA-seq data are challenging for traditional methods due to their high dimensional measurements as well as an abundance of dropout events (that is, zero expression measurements).

RESULTS

To overcome these difficulties, we propose DR-A (Dimensionality Reduction with Adversarial variational autoencoder), a data-driven approach to fulfill the task of dimensionality reduction. DR-A leverages a novel adversarial variational autoencoder-based framework, a variant of generative adversarial networks. DR-A is well-suited for unsupervised learning tasks for the scRNA-seq data, where labels for cell types are costly and often impossible to acquire. Compared with existing methods, DR-A is able to provide a more accurate low dimensional representation of the scRNA-seq data. We illustrate this by utilizing DR-A for clustering of scRNA-seq data.

CONCLUSIONS

Our results indicate that DR-A significantly enhances clustering performance over state-of-the-art methods.

QSMGAN: Improved Quantitative Susceptibility Mapping using 3D Generative Adversarial Networks with increased receptive field

Quantitative susceptibility mapping (QSM) is a powerful MRI technique that has shown great potential in quantifying tissue susceptibility in numerous neurological disorders. However, the intrinsic ill-posed dipole inversion problem greatly affects the accuracy of the susceptibility map. We propose QSMGAN: a 3D deep convolutional neural network approach based on a 3D U-Net architecture with increased receptive field of the input phase compared to the output and further refined the network using the WGAN with gradient penalty training strategy. Our method generates accurate QSM maps from single orientation phase maps efficiently and performs significantly better than traditional non-learning-based dipole inversion algorithms. The generalization capability was verified by applying the algorithm to an unseen pathology--brain tumor patients with radiation-induced cerebral microbleeds.

Reconstruction of undersampled 3D non-Cartesian image-based navigators for coronary MRA using an unrolled deep learning model

PURPOSE

To rapidly reconstruct undersampled 3D non-Cartesian image-based navigators (iNAVs) using an unrolled deep learning (DL) model, enabling nonrigid motion correction in coronary magnetic resonance angiography (CMRA).

METHODS

An end-to-end unrolled network is trained to reconstruct beat-to-beat 3D iNAVs acquired during a CMRA sequence. The unrolled model incorporates a nonuniform FFT operator in TensorFlow to perform the data-consistency operation, and the regularization term is learned by a convolutional neural network (CNN) based on the proximal gradient descent algorithm. The training set includes 6,000 3D iNAVs acquired from 7 different subjects and 11 scans using a variable-density (VD) cones trajectory. For testing, 3D iNAVs from 4 additional subjects are reconstructed using the unrolled model. To validate reconstruction accuracy, global and localized motion estimates from DL model-based 3D iNAVs are compared with those extracted from 3D iNAVs reconstructed with l 1 -ESPIRiT. Then, the high-resolution coronary MRA images motion corrected with autofocusing using the l 1 -ESPIRiT and DL model-based 3D iNAVs are assessed for differences.

RESULTS

3D iNAVs reconstructed using the DL model-based approach and conventional l 1 -ESPIRiT generate similar global and localized motion estimates and provide equivalent coronary image quality. Reconstruction with the unrolled network completes in a fraction of the time compared to CPU and GPU implementations of l 1 -ESPIRiT (20× and 3× speed increases, respectively).

CONCLUSIONS

We have developed a deep neural network architecture to reconstruct undersampled 3D non-Cartesian VD cones iNAVs. Our approach decreases reconstruction time for 3D iNAVs, while preserving the accuracy of nonrigid motion information offered by them for correction.

fastMRI: A Publicly Available Raw k-Space and DICOM Dataset of Knee Images for Accelerated MR Image Reconstruction Using Machine Learning

A publicly available dataset containing k-space data as well as Digital Imaging and Communications in Medicine image data of knee images for accelerated MR image reconstruction using machine learning is presented.

Compressive Sensing Spectroscopy Using a Residual Convolutional Neural Network

Compressive sensing (CS) spectroscopy is well known for developing a compact spectrometer which consists of two parts: compressively measuring an input spectrum and recovering the spectrum using reconstruction techniques. Our goal here is to propose a novel residual convolutional neural network (ResCNN) for reconstructing the spectrum from the compressed measurements. The proposed ResCNN comprises learnable layers and a residual connection between the input and the output of these learnable layers. The ResCNN is trained using both synthetic and measured spectral datasets. The results demonstrate that ResCNN shows better spectral recovery performance in terms of average root mean squared errors (RMSEs) and peak signal to noise ratios (PSNRs) than existing approaches such as the sparse recovery methods and the spectral recovery using CNN. Unlike sparse recovery methods, ResCNN does not require a priori knowledge of a sparsifying basis nor prior information on the spectral features of the dataset. Moreover, ResCNN produces stable reconstructions under noisy conditions. Finally, ResCNN is converged faster than CNN.

Deep Magnetic Resonance Image Reconstruction: Inverse Problems Meet Neural Networks

Image reconstruction from undersampled k-space data has been playing an important role in fast MRI. Recently, deep learning has demonstrated tremendous success in various fields and has also shown potential in significantly accelerating MRI reconstruction with fewer measurements. This article provides an overview of the deep learning-based image reconstruction methods for MRI. Two types of deep learning-based approaches are reviewed: those based on unrolled algorithms and those which are not. The main structure of both approaches are explained, respectively. Several signal processing issues for maximizing the potential of deep reconstruction in fast MRI are discussed. The discussion may facilitate further development of the networks and the analysis of performance from a theoretical point of view.

Compressed Sensing: From Research to Clinical Practice with Deep Neural Networks

Compressed sensing (CS) reconstruction methods leverage sparse structure in underlying signals to recover high-resolution images from highly undersampled measurements. When applied to magnetic resonance imaging (MRI), CS has the potential to dramatically shorten MRI scan times, increase diagnostic value, and improve overall patient experience. However, CS has several shortcomings which limit its clinical translation such as: 1) artifacts arising from inaccurate sparse modelling assumptions, 2) extensive parameter tuning required for each clinical application, and 3) clinically infeasible reconstruction times. Recently, CS has been extended to incorporate deep neural networks as a way of learning complex image priors from historical exam data. Commonly referred to as unrolled neural networks, these techniques have proven to be a compelling and practical approach to address the challenges of sparse CS. In this tutorial, we will review the classical compressed sensing formulation and outline steps needed to transform this formulation into a deep learning-based reconstruction framework. Supplementary open source code in Python will be used to demonstrate this approach with open databases. Further, we will discuss considerations in applying unrolled neural networks in the clinical setting.

CT Super-Resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE)

In this paper, we present a semi-supervised deep learning approach to accurately recover high-resolution (HR) CT images from low-resolution (LR) counterparts. Specifically, with the generative adversarial network (GAN) as the building block, we enforce the cycle-consistency in terms of the Wasserstein distance to establish a nonlinear end-to-end mapping from noisy LR input images to denoised and deblurred HR outputs. We also include the joint constraints in the loss function to facilitate structural preservation. In this process, we incorporate deep convolutional neural network (CNN), residual learning, and network in network techniques for feature extraction and restoration. In contrast to the current trend of increasing network depth and complexity to boost the imaging performance, we apply a parallel 1×1 CNN to compress the output of the hidden layer and optimize the number of layers and the number of filters for each convolutional layer. The quantitative and qualitative evaluative results demonstrate that our proposed model is accurate, efficient and robust for super-resolution (SR) image restoration from noisy LR input images. In particular, we validate our composite SR networks on three large-scale CT datasets, and obtain promising results as compared to the other state-of-the-art methods.

Minimal Linear Networks for Magnetic Resonance Image Reconstruction

Modern sequences for Magnetic Resonance Imaging (MRI) trade off scan time with computational challenges, resulting in ill-posed inverse problems and the requirement to account for more elaborated signal models. Various deep learning techniques have shown potential for image reconstruction from reduced data, outperforming compressed sensing, dictionary learning and other advanced techniques based on regularization, by characterization of the image manifold. In this work we suggest a framework for reducing a “neural” network to the bare minimum required by the MR physics, reducing the network depth and removing all non-linearities. The networks performed well both on benchmark simulated data and on arterial spin labeling perfusion imaging, showing clear images while preserving sensitivity to the minute signal changes. The results indicate that the deep learning framework plays a major role in MR image reconstruction, and suggest a concrete approach for probing into the contribution of additional elements.

Next generation research applications for hybrid PET/MR and PET/CT imaging using deep learning

INTRODUCTION

Recently there have been significant advances in the field of machine learning and artificial intelligence (AI) centered around imaging-based applications such as computer vision. In particular, the tremendous power of deep learning algorithms, primarily based on convolutional neural network strategies, is becoming increasingly apparent and has already had direct impact on the fields of radiology and nuclear medicine. While most early applications of computer vision to radiological imaging have focused on classification of images into disease categories, it is also possible to use these methods to improve image quality. Hybrid imaging approaches, such as PET/MRI and PET/CT, are ideal for applying these methods.

METHODS

This review will give an overview of the application of AI to improve image quality for PET imaging directly and how the additional use of anatomic information from CT and MRI can lead to further benefits. For PET, these performance gains can be used to shorten imaging scan times, with improvement in patient comfort and motion artifacts, or to push towards lower radiotracer doses. It also opens the possibilities for dual tracer studies, more frequent follow-up examinations, and new imaging indications. How to assess quality and the potential effects of bias in training and testing sets will be discussed.

CONCLUSION

Harnessing the power of these new technologies to extract maximal information from hybrid PET imaging will open up new vistas for both research and clinical applications with associated benefits in patient care.

Accelerated dynamic contrast enhanced MRI based on region of interest compressed sensing

Magnetic Resonance Imaging (MRI) provides excellent soft tissue contrast with one significant limitation of slow data acquisition. Dynamic Contrast Enhanced MRI (DCE-MRI) is one of the widely employed techniques to estimate tumor tissue physiological parameters using contrast agents. DCE-MRI data acquisition and reconstruction requires high spatiotemporal resolution, especially during the post-contrast phase. The region of Interest Compressed Sensing (ROICS) is based on Compressed Sensing (CS) framework and works on the hypothesis that limiting CS to an ROI can achieve superior CS performance. In this work, ROICS has been demonstrated on breast DCE-MRI data at chosen acceleration factors and the results are compared with conventional CS implementation. Normalized Root Mean Square Error (NRMSE) was calculated to compare ROICS with CS quantitatively. CS and ROICS reconstructed images were used to compare K^trans and v_e values derived using standard Tofts Model (TM). This also validated the superior performance of ROICS over conventional CS. ROICS generated Concentration Time Curves (CTC's) at chosen acceleration factors follow similar trend as the ground truth data as compared to CS. Both qualitative and quantitative analyses show that ROICS outperforms CS particularly at acceleration factors of 5× and above.

SANTIS: Sampling-Augmented Neural neTwork with Incoherent Structure for MR image reconstruction

PURPOSE

To develop and evaluate a novel deep learning-based reconstruction framework called SANTIS (Sampling-Augmented Neural neTwork with Incoherent Structure) for efficient MR image reconstruction with improved robustness against sampling pattern discrepancy.

METHODS

With a combination of data cycle-consistent adversarial network, end-to-end convolutional neural network mapping, and data fidelity enforcement for reconstructing undersampled MR data, SANTIS additionally utilizes a sampling-augmented training strategy by extensively varying undersampling patterns during training, so that the network is capable of learning various aliasing structures and thereby removing undersampling artifacts more effectively and robustly. The performance of SANTIS was demonstrated for accelerated knee imaging and liver imaging using a Cartesian trajectory and a golden-angle radial trajectory, respectively. Quantitative metrics were used to assess its performance against different references. The feasibility of SANTIS in reconstructing dynamic contrast-enhanced images was also demonstrated using transfer learning.

RESULTS

Compared to conventional reconstruction that exploits image sparsity, SANTIS achieved consistently improved reconstruction performance (lower errors and greater image sharpness). Compared to standard learning-based methods without sampling augmentation (e.g., training with a fixed undersampling pattern), SANTIS provides comparable reconstruction performance, but significantly improved robustness, against sampling pattern discrepancy. SANTIS also achieved encouraging results for reconstructing liver images acquired at different contrast phases.

CONCLUSION

By extensively varying undersampling patterns, the sampling-augmented training strategy in SANTIS can remove undersampling artifacts more robustly. The novel concept behind SANTIS can particularly be useful for improving the robustness of deep learning-based image reconstruction against discrepancy between training and inference, an important, but currently less explored, topic.

Optimized fast GPU implementation of robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction

BACKGROUND

Robust Artificial-neural-networks for k-space Interpolation (RAKI) is a recently proposed deep-learning-based reconstruction algorithm for parallel imaging. Its main premise is to perform k-space interpolation using convolutional neural networks (CNNs) trained on subject-specific autocalibration signal (ACS) data. Since training is performed individually for each subject, the reconstruction time is longer than approaches that pre-train on databases. In this study, we sought to reduce the computational time of RAKI.

METHODS

RAKI was implemented using CPU multi-processing and process pooling to maximize the utility of GPU resources. We also proposed an alternative CNN architecture that interpolates all output channels jointly for specific skipped k-space lines. This new architecture was compared to the original CNN architecture in RAKI, as well as to GRAPPA in phantom, brain and knee MRI datasets, both qualitatively and quantitatively.

RESULTS

The optimized GPU implementations were approximately 2-to-5-fold faster than a simple GPU implementation. The new CNN architecture further improved the computational time by 4-to-5-fold compared to the optimized GPU implementation using the original RAKI CNN architecture. It also provided significant improvement over GRAPPA both visually and quantitatively, although it performed slightly worse than the original RAKI CNN architecture.

CONCLUSIONS

The proposed implementations of RAKI bring the computational time towards clinically acceptable ranges. The new CNN architecture yields faster training, albeit at a slight performance loss, which may be acceptable for faster visualization in some settings.

Artificial intelligence in pediatric and adult congenital cardiac MRI: an unmet clinical need

Cardiac MRI (CMR) allows non-invasive, non-ionizing assessment of cardiac function and anatomy in patients with congenital heart disease (CHD). The utility of CMR as a non-invasive imaging tool for evaluation of CHD have been growing exponentially over the past decade. The algorithms based on artificial intelligence (AI), and in particular, deep learning, have rapidly become a methodology of choice for analyzing CMR. A wide range of applications for AI have been developed to tackle challenges in various aspects of CMR, and significant advances have also been made from image acquisition to image analysis and diagnosis. We include an overview of AI definitions, different architectures, and details on well-known methods. This paper reviews the major deep learning concepts used for analyses of patients with CHD. In the end, we have summarized a list of open challenges and concerns to be considered for future studies.

Image Synthesis in Multi-Contrast MRI With Conditional Generative Adversarial Networks

Acquiring images of the same anatomy with multiple different contrasts increases the diversity of diagnostic information available in an MR exam. Yet, the scan time limitations may prohibit the acquisition of certain contrasts, and some contrasts may be corrupted by noise and artifacts. In such cases, the ability to synthesize unacquired or corrupted contrasts can improve diagnostic utility. For multi-contrast synthesis, the current methods learn a nonlinear intensity transformation between the source and target images, either via nonlinear regression or deterministic neural networks. These methods can, in turn, suffer from the loss of structural details in synthesized images. Here, in this paper, we propose a new approach for multi-contrast MRI synthesis based on conditional generative adversarial networks. The proposed approach preserves intermediate-to-high frequency details via an adversarial loss, and it offers enhanced synthesis performance via pixel-wise and perceptual losses for registered multi-contrast images and a cycle-consistency loss for unregistered images. Information from neighboring cross-sections are utilized to further improve synthesis quality. Demonstrations on T₁- and T₂- weighted images from healthy subjects and patients clearly indicate the superior performance of the proposed approach compared to the previous state-of-the-art methods. Our synthesis approach can help improve the quality and versatility of the multi-contrast MRI exams without the need for prolonged or repeated examinations.

Applications of Deep Learning to Neuro-Imaging Techniques

Many clinical applications based on deep learning and pertaining to radiology have been proposed and studied in radiology for classification, risk assessment, segmentation tasks, diagnosis, prognosis, and even prediction of therapy responses. There are many other innovative applications of AI in various technical aspects of medical imaging, particularly applied to the acquisition of images, ranging from removing image artifacts, normalizing/harmonizing images, improving image quality, lowering radiation and contrast dose, and shortening the duration of imaging studies. This article will address this topic and will seek to present an overview of deep learning applied to neuroimaging techniques.

Undersampled MR image reconstruction using an enhanced recursive residual network

When using aggressive undersampling, it is difficult to recover the high quality image with reliably fine features. In this paper, we propose an enhanced recursive residual network (ERRN) that improves the basic recursive residual network with a high-frequency feature guidance, an error-correction unit and dense connections. The feature guidance is designed to predict the underlying anatomy based on image a priori learned from the label data, playing a complementary role to the residual learning. The ERRN is adapted for two important applications: compressed sensing (CS) MRI and super resolution (SR) MRI, while an application-specific error-correction unit is added into the framework, i.e. data consistency for CS-MRI and back projection for SR-MRI due to their different sampling schemes. Our proposed network was evaluated using a real-valued brain dataset, a complex-valued knee dataset, pathological brain data and in vivo rat brain data with different undersampling masks and rates. Experimental results demonstrated that ERRN presented superior reconstructions at all cases with distinctly restored structural features and highest image quality metrics compared to both the state-of-the-art convolutional neural networks and the conventional optimization-based methods, particularly for the undersampling rate over 5-fold. Thus, an excellent framework design can endow the network with a flexible architecture, fewer parameters, outstanding performances for various undersampling schemes, and reduced overfitting in generalization, which will facilitate real-time reconstruction on MRI scanners.

Deep learning application engine (DLAE): Development and integration of deep learning algorithms in medical imaging

Herein we introduce a deep learning (DL) application engine (DLAE) system concept, present potential uses of it, and describe pathways for its integration in clinical workflows. An open-source software application was developed to provide a code-free approach to DL for medical imaging applications. DLAE supports several DL techniques used in medical imaging, including convolutional neural networks, fully convolutional networks, generative adversarial networks, and bounding box detectors. Several example applications using clinical images were developed and tested to demonstrate the capabilities of DLAE. Additionally, a model deployment example was demonstrated in which DLAE was used to integrate two trained models into a commercial clinical software package.

Parallel imaging and convolutional neural network combined fast MR image reconstruction: Applications in low-latency accelerated real-time imaging

PURPOSE

To develop and evaluate a parallel imaging and convolutional neural network combined image reconstruction framework for low-latency and high-quality accelerated real-time MR imaging.

METHODS

Conventional Parallel Imaging reconstruction resolved as gradient descent steps was compacted as network layers and interleaved with convolutional layers in a general convolutional neural network. All parameters of the network were determined during the offline training process, and applied to unseen data once learned. The proposed network was first evaluated for real-time cardiac imaging at 1.5 T and real-time abdominal imaging at 0.35 T, using threefold to fivefold retrospective undersampling for cardiac imaging and threefold retrospective undersampling for abdominal imaging. Then, prospective undersampling with fourfold acceleration was performed on cardiac imaging to compare the proposed method with standard clinically available GRAPPA method and the state-of-the-art L1-ESPIRiT method.

RESULTS

Both retrospective and prospective evaluations confirmed that the proposed network was able to images with a lower noise level and reduced aliasing artifacts in comparison with the single-coil based and L1-ESPIRiT reconstructions for cardiac imaging at 1.5 T, and the GRAPPA and L1-ESPIRiT reconstructions for abdominal imaging at 0.35 T. Using the proposed method, each frame can be reconstructed in less than 100 ms, suggesting its clinical compatibility.

CONCLUSION

The proposed Parallel Imaging and convolutional neural network combined reconstruction framework is a promising technique that allows low-latency and high-quality real-time MR imaging.

Deep residual network for off-resonance artifact correction with application to pediatric body MRA with 3D cones

PURPOSE

To enable rapid imaging with a scan time-efficient 3D cones trajectory with a deep-learning off-resonance artifact correction technique.

METHODS

A residual convolutional neural network to correct off-resonance artifacts (Off-ResNet) was trained with a prospective study of pediatric MRA exams. Each exam acquired a short readout scan (1.18 ms ± 0.38) and a long readout scan (3.35 ms ± 0.74) at 3 T. Short readout scans, with longer scan times but negligible off-resonance blurring, were used as reference images and augmented with additional off-resonance for supervised training examples. Long readout scans, with greater off-resonance artifacts but shorter scan time, were corrected by autofocus and Off-ResNet and compared with short readout scans by normalized RMS error, structural similarity index, and peak SNR. Scans were also compared by scoring on 8 anatomical features by two radiologists, using analysis of variance with post hoc Tukey's test and two one-sided t-tests. Reader agreement was determined with intraclass correlation.

RESULTS

The total scan time for long readout scans was on average 59.3% shorter than short readout scans. Images from Off-ResNet had superior normalized RMS error, structural similarity index, and peak SNR compared with uncorrected images across ±1 kHz off-resonance (P < .01). The proposed method had superior normalized RMS error over -677 Hz to +1 kHz and superior structural similarity index and peak SNR over ±1 kHz compared with autofocus (P < .01). Radiologic scoring demonstrated that long readout scans corrected with Off-ResNet were noninferior to short readout scans (P < .05).

CONCLUSION

The proposed method can correct off-resonance artifacts from rapid long-readout 3D cones scans to a noninferior image quality compared with diagnostically standard short readout scans.

MANTIS: Model-Augmented Neural neTwork with Incoherent k-space Sampling for efficient MR parameter mapping

PURPOSE

To develop and evaluate a novel deep learning-based image reconstruction approach called MANTIS (Model-Augmented Neural neTwork with Incoherent k-space Sampling) for efficient MR parameter mapping.

METHODS

MANTIS combines end-to-end convolutional neural network (CNN) mapping, incoherent k-space undersampling, and a physical model as a synergistic framework. The CNN mapping directly converts a series of undersampled images straight into MR parameter maps using supervised training. Signal model fidelity is enforced by adding a pathway between the undersampled k-space and estimated parameter maps to ensure that the parameter maps produced synthesized k-space consistent with the acquired undersampling measurements. The MANTIS framework was evaluated on the T₂ mapping of the knee at different acceleration rates and was compared with 2 other CNN mapping methods and conventional sparsity-based iterative reconstruction approaches. Global quantitative assessment and regional T₂ analysis for the cartilage and meniscus were performed to demonstrate the reconstruction performance of MANTIS.

RESULTS

MANTIS achieved high-quality T₂ mapping at both moderate (R = 5) and high (R = 8) acceleration rates. Compared to conventional reconstruction approaches that exploited image sparsity, MANTIS yielded lower errors (normalized root mean square error of 6.1% for R = 5 and 7.1% for R = 8) and higher similarity (structural similarity index of 86.2% at R = 5 and 82.1% at R = 8) to the reference in the T₂ estimation. MANTIS also achieved superior performance compared to direct CNN mapping and a 2-step CNN method.

CONCLUSION

The MANTIS framework, with a combination of end-to-end CNN mapping, signal model-augmented data consistency, and incoherent k-space sampling, is a promising approach for efficient and robust estimation of quantitative MR parameters.