Search

23Na MRI can be used to quantify in-vivo tissue sodium concentration (TSC), but the low 23Na signal leads to long scan times and/or noisy or low-resolution images. Reconstruction algorithms such as CS have been proposed to mitigate low SNR; although, these can result in unnatural images, suboptimal denoising and long processing times. Recently, ML has been used to denoise 1H MRI acquisitions; however, this approach typically requires large volumes of high-quality training data, which is not readily available for 23Na MRI. Here, we train a denoising CNN using 1H data, which we subsequently demonstrate on prospective 23Na images of the calf. 1893 1H transverse slices of the knee were used to train denoising CNNs for different levels of noise. Low SNR images were generated by adding gaussian noise to the high-quality 1H kspace data before reconstruction to create paired training data. For prospective testing, 23Na images of the calf were acquired in 10 volunteers with 150 averages, which were used as a reference throughout the study. From this data, lower-average images were reconstructed using a NUFFT as well as CS, with the NUFFT images subsequently denoised using the trained CNN. CNNs were successfully applied to 23Na images reconstructed with 50, 40 and 30 averages. SNR was significantly higher in CNN images compared to NUFFT, CS and reference images. Edge sharpness was equivalent for all images. For image quality ranking, CNN images ranked equally best with reference images and significantly better than NUFFT and CS images. Muscle and skin TSC quantification from CNN images were equivalent to those from CS images, with <0.9 mM bias compared to reference values. Denoising CNNs trained on 1H data can be successfully applied to 23Na images of the calf; thus, allowing scan time to be reduced from 10 minutes to 2 minutes with little impact on image quality or TSC quantification accuracy., Comment: 18 pages, 4 figures, 5 tables. Submitted to Magnetic Resonance Imaging

Computational fluid dynamics (CFD) can be used for evaluation of hemodynamics. However, its routine use is limited by labor-intensive manual segmentation, CFD mesh creation, and time-consuming simulation. This study aims to train a deep learning model to both generate patient-specific volume-meshes of the pulmonary artery from 3D cardiac MRI data and directly estimate CFD flow fields. This study used 135 3D cardiac MRIs from both a public and private dataset. The pulmonary arteries in the MRIs were manually segmented and converted into volume-meshes. CFD simulations were performed on ground truth meshes and interpolated onto point-point correspondent meshes to create the ground truth dataset. The dataset was split 85/10/15 for training, validation and testing. Image2Flow, a hybrid image and graph convolutional neural network, was trained to transform a pulmonary artery template to patient-specific anatomy and CFD values. Image2Flow was evaluated in terms of segmentation and accuracy of CFD predicted was assessed using node-wise comparisons. Centerline comparisons of Image2Flow and CFD simulations performed using machine learning segmentation were also performed. Image2Flow achieved excellent segmentation accuracy with a median Dice score of 0.9 (IQR: 0.86-0.92). The median node-wise normalized absolute error for pressure and velocity magnitude was 11.98% (IQR: 9.44-17.90%) and 8.06% (IQR: 7.54-10.41), respectively. Centerline analysis showed no significant difference between the Image2Flow and conventional CFD simulated on machine learning-generated volume-meshes. This proof-of-concept study has shown it is possible to simultaneously perform patient specific volume-mesh based segmentation and pressure and flow field estimation. Image2Flow completes segmentation and CFD in ~205ms, which ~7000 times faster than manual methods, making it more feasible in a clinical environment., Comment: 22 pages, 7 figures, 3 tables

Purpose: To develop and assess a deep learning (DL) pipeline to learn dynamic MR image reconstruction from publicly available natural videos (Inter4K). Materials and Methods: Learning was performed for a range of DL architectures (VarNet, 3D UNet, FastDVDNet) and corresponding sampling patterns (Cartesian, radial, spiral) either from true multi-coil cardiac MR data (N=692) or from pseudo-MR data simulated from Inter4K natural videos (N=692). Real-time undersampled dynamic MR images were reconstructed using DL networks trained with cardiac data and natural videos, and compressed sensing (CS). Differences were assessed in simulations (N=104 datasets) in terms of MSE, PSNR, and SSIM and prospectively for cardiac (short axis, four chambers, N=20) and speech (N=10) data in terms of subjective image quality ranking, SNR and Edge sharpness. Friedman Chi Square tests with post-hoc Nemenyi analysis were performed to assess statistical significance. Results: For all simulation metrics, DL networks trained with cardiac data outperformed DL networks trained with natural videos, which outperformed CS (p<0.05). However, in prospective experiments DL reconstructions using both training datasets were ranked similarly (and higher than CS) and presented no statistical differences in SNR and Edge Sharpness for most conditions. Additionally, high SSIM was measured between the DL methods with cardiac data and natural videos (SSIM>0.85). Conclusion: The developed pipeline enabled learning dynamic MR reconstruction from natural videos preserving DL reconstruction advantages such as high quality fast and ultra-fast reconstructions while overcoming some limitations (data scarcity or sharing). The natural video dataset, code and pre-trained networks are made readily available on github. Key Words: real-time; dynamic MRI; deep learning; image reconstruction; machine learning

Three-dimensional (3D) imaging is popular in medical applications, however, anisotropic 3D volumes with thick, low-spatial-resolution slices are often acquired to reduce scan times. Deep learning (DL) offers a solution to recover high-resolution features through super-resolution reconstruction (SRR). Unfortunately, paired training data is unavailable in many 3D medical applications and therefore we propose a novel unpaired approach; CLADE (Cycle Loss Augmented Degradation Enhancement). CLADE uses a modified CycleGAN architecture with a cycle-consistent gradient mapping loss, to learn SRR of the low-resolution dimension, from disjoint patches of the high-resolution plane within the anisotropic 3D volume data itself. We show the feasibility of CLADE in abdominal MRI and abdominal CT and demonstrate significant improvements in CLADE image quality over low-resolution volumes and state-of-the-art self-supervised SRR; SMORE (Synthetic Multi-Orientation Resolution Enhancement). Quantitative PIQUE (qualitative perception-based image quality evaluator) scores and quantitative edge sharpness (ES - calculated as the maximum gradient of pixel intensities over a border of interest), showed superior performance for CLADE in both MRI and CT. Qualitatively CLADE had the best overall image quality and highest perceptual ES over the low-resolution volumes and SMORE. This paper demonstrates the potential of using CLADE for super-resolution reconstruction of anisotropic 3D medical imaging data without the need for paired 3D training data.

Purpose: To develop and evaluate an end-to-end deep learning pipeline for segmentation and analysis of cardiac magnetic resonance images to provide core-lab processing for a multi-centre registry of Fontan patients. Materials and Methods: This retrospective study used training (n = 175), validation (n = 25) and testing (n = 50) cardiac magnetic resonance image exams collected from 13 institutions in the UK, US and Canada. The data was used to train and evaluate a pipeline containing three deep-learning models. The pipeline's performance was assessed on the Dice and IoU score between the automated and reference standard manual segmentation. Cardiac function values were calculated from both the automated and manual segmentation and evaluated using Bland-Altman analysis and paired t-tests. The overall pipeline was further evaluated qualitatively on 475 unseen patient exams. Results: For the 50 testing dataset, the pipeline achieved a median Dice score of 0.91 (0.89-0.94) for end-diastolic volume, 0.86 (0.82-0.89) for end-systolic volume, and 0.74 (0.70-0.77) for myocardial mass. The deep learning-derived end-diastolic volume, end-systolic volume, myocardial mass, stroke volume and ejection fraction had no statistical difference compared to the same values derived from manual segmentation with p values all greater than 0.05. For the 475 unseen patient exams, the pipeline achieved 68% adequate segmentation in both systole and diastole, 26% needed minor adjustments in either systole or diastole, 5% needed major adjustments, and the cropping model only failed in 0.4%. Conclusion: Deep learning pipeline can provide standardised 'core-lab' segmentation for Fontan patients. This pipeline can now be applied to the >4500 cardiac magnetic resonance exams currently in the FORCE registry as well as any new patients that are recruited., Comment: 17 pages, 6 figures

Background: Computational fluid dynamics (CFD) is increasingly used to assess blood flow conditions in patients with congenital heart disease (CHD). This requires patient-specific anatomy, usually obtained from segmented 3D cardiovascular magnetic resonance (CMR) images. However, segmentation is time-consuming and needs expert input. This study aims to develop and validate a machine learning (ML) method for segmentation of the aorta and pulmonary arteries (PAs) for CFD studies. Methods: 90 CHD patients were retrospectively selected for this study. 3D CMR images were manually segmented to obtain ground-truth (GT) background, aorta and PA labels. These were used to train and optimize a U-Net model. Segmentation performance was primarily evaluated using Dice score. CFD simulations were set up from GT and ML segmentations using a semi-automatic meshing and simulation pipeline. Pressure and velocity fields were computed, and a mean average percentage error (MAPE) was calculated for each vessel pair. A secondary observer (SO) segmented the test dataset to assess inter-observer variability. Friedman tests were used to compare segmentation metrics and flow field errors. Results: The model's Dice score (ML vs GT) was 0.945 for the aorta and 0.885 for the PAs. Differences with the inter-observer Dice score (SO vs GT) and ML vs SO Dice scores were not statistically significant for either aorta or PAs. The ML vs GT MAPEs for pressure and velocity in the aorta were 10.1% and 4.1% respectively, and for the PAs 14.6% and 6.3%, respectively. Inter-observer (SO vs GT) and ML vs SO pressure and velocity MAPEs were of a similar magnitude to ML vs GT. Conclusions: The proposed method can successfully segment the great vessels for CFD, with errors similar to inter-observer variability. This reduces the time and effort needed for CFD analysis, making it more attractive for routine clinical use., Comment: 27 pages, 10 figures

Computational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high computational resources, specialist experience needed for designing simulation set-ups, and long processing times. The aim of this study was to explore the use of machine learning (ML) to replicate conventional aortic CFD with automatic and fast regression models. Data used to train/test the model consisted of 3,000 CFD simulations performed on synthetically generated 3D aortic shapes. These subjects were generated from a statistical shape model (SSM) built on real patient-specific aortas (N=67). Inference performed on 200 test shapes resulted in average errors of 6.01% +/-3.12 SD and 3.99% +/-0.93 SD for pressure and velocity, respectively. Our ML-based models performed CFD in +/-0.075 seconds (4,000x faster than the solver). This proof-of-concept study shows that results from conventional vascular CFD can be reproduced using ML at a much faster rate, in an automatic process, and with high accuracy., Comment: 22 pages, 19 figures

Purpose: Real-time monitoring of cardiac output (CO) requires low latency reconstruction and segmentation of real-time phase contrast MR (PCMR), which has previously been difficult to perform. Here we propose a deep learning framework for 'Flow Reconstruction and Segmentation for low latency Cardiac Output monitoring' (FReSCO). Methods: Deep artifact suppression and segmentation U-Nets were independently trained. Breath hold spiral PCMR data (n=516) was synthetically undersampled using a variable density spiral sampling pattern and gridded to create aliased data for training of the artifact suppression U-net. A subset of the data (n=96) was segmented and used to train the segmentation U-net. Real-time spiral PCMR was prospectively acquired and then reconstructed and segmented using the trained models (FReSCO) at low latency at the scanner in 10 healthy subjects during rest, exercise and recovery periods. CO obtained via FReSCO was compared to a reference rest CO and rest and exercise Compressed Sensing (CS) CO. Results: FReSCO was demonstrated prospectively at the scanner. Beat-to-beat heartrate, stroke volume and CO could be visualized with a mean latency of 622ms. No significant differences were noted when compared to reference at rest (Bias = -0.21+-0.50 L/min, p=0.246) or CS at peak exercise (Bias=0.12+-0.48 L/min, p=0.458). Conclusion: FReSCO was successfully demonstrated for real-time monitoring of CO during exercise and could provide a convenient tool for assessment of the hemodynamic response to a range of stressors., Comment: 10 pages, 5 Figures, 4 Supporting Information Figures

Purpose: We implemented the Machine Learning (ML) aided k-t SENSE reconstruction to enable high resolution quantitative real-time phase contrast MR (PCMR). Methods: A residual U-net and our U-net M were used to generate the high resolution x-f space estimate for k-t SENSE regularisation prior. The networks were judged on their ability to generalise to real undersampled data. The in-vivo validation was done on 20 real-time 18x prospectively undersmapled GASperturbed PCMR data. The ML aided k-t SENSE reconstruction results were compared against the free-breathing Cartesian retrospectively gated sequence and the compressed sensing (CS) reconstruction of the same data. Results: In general, the ML aided k-t SENSE generated flow curves that were visually sharper than those produced using CS. In two exceptional cases, U-net M predictions exhibited blurring which propagated to the extracted velocity curves. However, there were no statistical differences in the measured peak velocities and stroke volumes between the tested methods. The ML aided k-t SENSE was estimated to be ~3.6x faster in processing than CS. Conclusion: The ML aided k-t SENSE reconstruction enables artefact suppression on a par with CS with no significant differences in quantitative measures. The timing results suggest the on-line implementation could deliver a substantial increase in clinical throughput., Comment: 26 pages, 6 figures, 6 videos

Magnetic Resonance Imaging (MRI) plays a vital role in diagnosis, management and monitoring of many diseases. However, it is an inherently slow imaging technique. Over the last 20 years, parallel imaging, temporal encoding and compressed sensing have enabled substantial speed-ups in the acquisition of MRI data, by accurately recovering missing lines of k-space data. However, clinical uptake of vastly accelerated acquisitions has been limited, in particular in compressed sensing, due to the time-consuming nature of the reconstructions and unnatural looking images. Following the success of machine learning in a wide range of imaging tasks, there has been a recent explosion in the use of machine learning in the field of MRI image reconstruction. A wide range of approaches have been proposed, which can be applied in k-space and/or image-space. Promising results have been demonstrated from a range of methods, enabling natural looking images and rapid computation. In this review article we summarize the current machine learning approaches used in MRI reconstruction, discuss their drawbacks, clinical applications, and current trends., Comment: 34 pages, 3 figures, 1 table. review article

Background: Three-dimensional, whole heart, balanced steady state free precession (WH-bSSFP) sequences provide delineation of intra-cardiac and vascular anatomy. However, they have long acquisition times. Here, we propose significant speed ups using a deep learning single volume super resolution reconstruction, to recover high resolution features from rapidly acquired low resolution WH-bSSFP images. Methods: A 3D residual U-Net was trained using synthetic data, created from a library of high-resolution WH-bSSFP images by simulating 0.5 slice resolution and 0.5 phase resolution. The trained network was validated with synthetic test data, as well as prospective low-resolution data. Results: Synthetic low-resolution data had significantly better image quality after super-resolution reconstruction. Qualitative image scores showed super-resolved images had better edge sharpness, fewer residual artefacts and less image distortion than low-resolution images, with similar scores to high-resolution data. Quantitative image scores showed super-resolved images had significantly better edge sharpness than low-resolution or high-resolution images, with significantly better signal-to-noise ratio than high-resolution data. Vessel diameters measurements showed over-estimation in the low-resolution measurements, compared to the high-resolution data. No significant differences and no bias was found in the super-resolution measurements. Conclusion: This paper demonstrates the potential of using a residual U-Net for super-resolution reconstruction of rapidly acquired low-resolution whole heart bSSFP data within a clinical setting. The resulting network can be applied very quickly, making these techniques particularly appealing within busy clinical workflow. Thus, we believe that this technique may help speed up whole heart CMR in clinical practice.

Multi-dimensional non-Cartesian MRI encoding using the precomputed interpolator can encounter the curse of dimensionality, in which the interpolator size exceeds the available memory on the parallel accelerators. Here we reformulate the multi-dimensional non-uniform fast Fourier transform (NUFFT) to a tensor form. The exponentially growing size of the fully precomputed interpolator can be reduced by tensor analysis. We propose a tree-like, mixed-radix tensor method which flexibly reduces the storage of the NUFFT. A parallel tensor product algorithm is proposed and tested with in vivo cardiac MRI data. Cross-architecture comparisons show that up to 88.1% and 62.4% memory savings are seen in 3D and 2D CINE MRI, respectively, subject only to a negligible loss of accuracy compared to the double-precision CPU version., Comment: The submitted paper in its current form is inappropriate

Background: Real-time (RT) assessment of ventricular volumes and function enables data acquisition during free-breathing. However, in children the requirement for high spatiotemporal resolution requires accelerated imaging techniques. In this study, we implemented a novel RT bSSFP spiral sequence reconstructed using Compressed Sensing (CS) and validated it against the breath-hold (BH) reference standard for assessment of ventricular volumes in children with heart disease. Methods: Data was acquired in 60 children. Qualitative image scoring and evaluation of ventricular volumes was performed by 3 clinical cardiac MR specialists. 30 cases were reassessed for intra-observer variability, and the other 30 cases for inter-observer variability. Results: Spiral RT images were of good quality, however qualitative scores reflected more residual artefact than standard BH images and slightly lower edge definition. Quantification of Left Ventricular (LV) and Right Ventricular (RV) metrics showed excellent correlation between the techniques with narrow limits of agreement. However, we observed small but statistically significant overestimation of LV end-diastolic volume, underestimation of LV end-systolic volume, as well as a small overestimation of RV stroke volume and ejection fraction using the RT imaging technique. No difference in inter-observer or intra-observer variability were observed between the BH and RT sequences. Conclusions: Real-time bSSFP imaging using spiral trajectories combined with a compressed sensing reconstruction is feasible. The main benefit is that it can be acquired during free breathing. However, another important secondary benefit is that a whole ventricular stack can be acquired in ~20 seconds, as opposed to ~6 minutes for standard BH imaging. Thus, this technique holds the potential to significantly shorten MR scan times in children.

PURPOSE: Real-time assessment of ventricular volumes requires high acceleration factors. Residual convolutional neural networks (CNN) have shown potential for removing artifacts caused by data undersampling. In this study we investigated the effect of different radial sampling patterns on the accuracy of a CNN. We also acquired actual real-time undersampled radial data in patients with congenital heart disease (CHD), and compare CNN reconstruction to Compressed Sensing (CS). METHODS: A 3D (2D plus time) CNN architecture was developed, and trained using 2276 gold-standard paired 3D data sets, with 14x radial undersampling. Four sampling schemes were tested, using 169 previously unseen 3D 'synthetic' test data sets. Actual real-time tiny Golden Angle (tGA) radial SSFP data was acquired in 10 new patients (122 3D data sets), and reconstructed using the 3D CNN as well as a CS algorithm; GRASP. RESULTS: Sampling pattern was shown to be important for image quality, and accurate visualisation of cardiac structures. For actual real-time data, overall reconstruction time with CNN (including creation of aliased images) was shown to be more than 5x faster than GRASP. Additionally, CNN image quality and accuracy of biventricular volumes was observed to be superior to GRASP for the same raw data. CONCLUSION: This paper has demonstrated the potential for the use of a 3D CNN for deep de-aliasing of real-time radial data, within the clinical setting. Clinical measures of ventricular volumes using real-time data with CNN reconstruction are not statistically significantly different from the gold-standard, cardiac gated, BH techniques.