Deep Learning Reconstruction Enables Prospectively Accelerated Clinical Knee MRI
Background MRI is a powerful diagnostic tool with a long acquisition time. Recently, deep learning (DL) methods have provided accelerated high-quality image reconstructions from undersampled data, but it is unclear if DL image reconstruction can be reliably translated to everyday clinical practice. Purpose To determine the diagnostic equivalence of prospectively accelerated DL-reconstructed knee MRI compared with conventional accelerated MRI for evaluating internal derangement of the knee in a clinical setting. Materials and Methods A DL reconstruction model was trained with images from 298 clinical 3-T knee examinations. In a prospective analysis, patients clinically referred for knee MRI underwent a conventional accelerated knee MRI protocol at 3 T followed by an accelerated DL protocol between January 2020 and February 2021. The equivalence of the DL reconstruction of the images relative to the conventional images for the detection of an abnormality was assessed in terms of interchangeability. Each examination was reviewed by six musculoskeletal radiologists. Analyses pertaining to the detection of meniscal or ligament tears and bone marrow or cartilage abnormalities were based on four-point ordinal scores for the likelihood of an abnormality. Additionally, the protocols were compared with use of four-point ordinal scores for each aspect of image quality: overall image quality, presence of artifacts, sharpness, and signal-to-noise ratio. Results A total of 170 participants (mean age ± SD, 45 years ± 16; 76 men) were evaluated. The DL-reconstructed images were determined to be of diagnostic equivalence with the conventional images for detection of abnormalities. The overall image quality score, averaged over six readers, was significantly better (P < .001) for the DL than for the conventional images. Conclusion In a clinical setting, deep learning reconstruction enabled a nearly twofold reduction in scan time for a knee MRI and was diagnostically equivalent with the conventional protocol. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Roemer in this issue.
Deep Learning Denoising of Low-Dose Computed Tomography Chest Images: A Quantitative and Qualitative Image Analysis
PURPOSE/OBJECTIVE:To assess deep learning denoised (DLD) computed tomography (CT) chest images at various low doses by both quantitative and qualitative perceptual image analysis. METHODS:Simulated noise was inserted into sinogram data from 32 chest CTs acquired at 100 mAs, generating anatomically registered images at 40, 20, 10, and 5 mAs. A DLD model was developed, with 23 scans selected for training, 5 for validation, and 4 for test.Quantitative analysis of perceptual image quality was assessed with Structural SIMilarity Index (SSIM) and Fréchet Inception Distance (FID). Four thoracic radiologists graded overall diagnostic image quality, image artifact, visibility of small structures, and lesion conspicuity. Noise-simulated and denoised image series were evaluated in comparison with one another, and in comparison with standard 100 mAs acquisition at the 4 mAs levels. Statistical tests were conducted at the 2-sided 5% significance level, with multiple comparison correction. RESULTS:At the same mAs levels, SSIM and FID between noise-simulated and reconstructed DLD images indicated that images were closer to a perfect match with increasing mAs (closer to 1 for SSIM, and 0 for FID).In comparing noise-simulated and DLD images to standard-dose 100-mAs images, DLD improved SSIM and FID. Deep learning denoising improved SSIM of 40-, 20-, 10-, and 5-mAs simulations in comparison with standard-dose 100-mAs images, with change in SSIM from 0.91 to 0.94, 0.87 to 0.93, 0.67 to 0.87, and 0.54 to 0.84, respectively. Deep learning denoising improved FID of 40-, 20-, 10-, and 5-mAs simulations in comparison with standard-dose 100-mAs images, with change in FID from 20 to 13, 46 to 21, 104 to 41, and 148 to 69, respectively.Qualitative image analysis showed no significant difference in lesion conspicuity between DLD images at any mAs in comparison with 100-mAs images. Deep learning denoising images at 10 and 5 mAs were rated lower for overall diagnostic image quality (P < 0.001), and at 5 mAs lower for overall image artifact and visibility of small structures (P = 0.002), in comparison with 100 mAs. CONCLUSIONS:Deep learning denoising resulted in quantitative improvements in image quality. Qualitative assessment demonstrated DLD images at or less than 10 mAs to be rated inferior to standard-dose images.
Physics-Driven Deep Learning for Computational Magnetic Resonance Imaging: Combining physics and machine learning for improved medical imaging
Physics-driven deep learning methods have emerged as a powerful tool for computational magnetic resonance imaging (MRI) problems, pushing reconstruction performance to new limits. This article provides an overview of the recent developments in incorporating physics information into learning-based MRI reconstruction. We consider inverse problems with both linear and nonlinear forward models for computational MRI and review the classical approaches for solving these. We then focus on physics-driven deep learning approaches, covering physics-driven loss functions, plug-and-play (PnP) methods, generative models, and unrolled networks. We highlight domain-specific challenges, such as real- and complex-valued building blocks of neural networks, and translational applications in MRI with linear and nonlinear forward models. Finally, we discuss common issues and open challenges, and we draw connections to the importance of physics-driven learning when combined with other downstream tasks in the medical imaging pipeline.
Exploring the Acceleration Limits of Deep Learning Variational Network-based Two-dimensional Brain MRI
PURPOSE/UNASSIGNED:To explore the limits of deep learning-based brain MRI reconstruction and identify useful acceleration ranges for general-purpose imaging and potential screening. MATERIALS AND METHODS/UNASSIGNED:potential use in a screening protocol. A Monte Carlo procedure was developed to estimate reconstruction error with only undersampled data. The model was evaluated on both in-domain and out-of-domain data. The 95% CIs were calculated using the percentile bootstrap method. RESULTS/UNASSIGNED:Radiologists rated 100% of 69 volumes as having sufficient image quality for general-purpose imaging at up to 4× acceleration and 65 of 69 volumes (94%) as having sufficient image quality for screening at up to 14× acceleration. The Monte Carlo procedure estimated ground truth peak signal-to-noise ratio and mean squared error with coefficients of determination greater than 0.5 at 2× to 20× acceleration levels. Out-of-distribution experiments demonstrated the model's ability to produce images substantially distinct from the training set, even at 100× acceleration. CONCLUSION/UNASSIGNED:© RSNA, 2022.
New-Generation Low-Field Magnetic Resonance Imaging of Hip Arthroplasty Implants Using Slice Encoding for Metal Artifact Correction: First In Vitro Experience at 0.55 T and Comparison With 1.5 T
OBJECTIVES/OBJECTIVE:Despite significant progress, artifact-free visualization of the bone and soft tissues around hip arthroplasty implants remains an unmet clinical need. New-generation low-field magnetic resonance imaging (MRI) systems now include slice encoding for metal artifact correction (SEMAC), which may result in smaller metallic artifacts and better image quality than standard-of-care 1.5 T MRI. This study aims to assess the feasibility of SEMAC on a new-generation 0.55 T system, optimize the pulse protocol parameters, and compare the results with those of a standard-of-care 1.5 T MRI. MATERIALS AND METHODS/METHODS:Titanium (Ti) and cobalt-chromium total hip arthroplasty implants embedded in a tissue-mimicking American Society for Testing and Materials gel phantom were evaluated using turbo spin echo, view angle tilting (VAT), and combined VAT and SEMAC (VAT + SEMAC) pulse sequences. To refine an MRI protocol at 0.55 T, the type of metal artifact reduction techniques and the effect of various pulse sequence parameters on metal artifacts were assessed through qualitative ranking of the images by 3 expert readers while taking measured spatial resolution, signal-to-noise ratios, and acquisition times into consideration. Signal-to-noise ratio efficiency and artifact size of the optimized 0.55 T protocols were compared with the 1.5 T standard and compressed-sensing SEMAC sequences. RESULTS:Overall, the VAT + SEMAC sequence with at least 6 SEMAC encoding steps for Ti and 9 for cobalt-chromium implants was ranked higher than other sequences for metal reduction (P < 0.05). Additional SEMAC encoding partitions did not result in further metal artifact reductions. Permitting minimal residual artifacts, low magnetic susceptibility Ti constructs may be sufficiently imaged with optimized turbo spin echo sequences obviating the need for SEMAC. In cross-platform comparison, 0.55 T acquisitions using the optimized protocols are associated with 45% to 64% smaller artifacts than 1.5 T VAT + SEMAC and VAT + compressed-sensing/SEMAC protocols at the expense of a 17% to 28% reduction in signal-to-noise ratio efficiency. B1-related artifacts are invariably smaller at 0.55 T than 1.5 T; however, artifacts related to B0 distortion, although frequently smaller, may appear as signal pileups at 0.55 T. CONCLUSIONS:Our results suggest that new-generation low-field SEMAC MRI reduces metal artifacts around hip arthroplasty implants to better advantage than current 1.5 T MRI standard of care. While the appearance of B0-related artifacts changes, reduction in B1-related artifacts plays a major role in the overall benefit of 0.55 T.
Deep Learning Reconstruction Enables Highly Accelerated Biparametric MR Imaging of the Prostate
BACKGROUND:Early diagnosis and treatment of prostate cancer (PCa) can be curative; however, prostate-specific antigen is a suboptimal screening test for clinically significant PCa. While prostate magnetic resonance imaging (MRI) has demonstrated value for the diagnosis of PCa, the acquisition time is too long for a first-line screening modality. PURPOSE/OBJECTIVE:To accelerate prostate MRI exams, utilizing a variational network (VN) for image reconstruction. STUDY TYPE/METHODS:Retrospective. SUBJECTS/METHODS:One hundred and thirteen subjects (train/val/test: 70/13/30) undergoing prostate MRI. FIELD STRENGTH/SEQUENCE/UNASSIGNED:3.0â€‰T; a T2 turbo spin echo (TSE) T2-weighted image (T2WI) sequence in axial and coronal planes, and axial echo-planar diffusion-weighted imaging (DWI). ASSESSMENT/RESULTS:, and apparent diffusion coefficient map-according to the Prostate Imaging Reporting and Data System (PI-RADS v2.1), for both VN and standard reconstructions. Accuracy of PI-RADS â‰¥3 for clinically significant cancer was computed. Projected scan time of the retrospectively under-sampled biparametric exam was also computed. STATISTICAL TESTS/UNASSIGNED:One-sided Wilcoxon signed-rank test was used for comparison of image quality. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for lesion detection and grading. Generalized estimating equation with cluster effect was used to compare differences between standard and VN bp-MRI. A P-value of <0.05 was considered statistically significant. RESULTS:(Reader 1: 3.20â€‰Â±â€‰0.70 (Standard), 3.40â€‰Â±â€‰0.75 (VN) PÂ =Â 0.98; Reader 2: 2.85â€‰Â±â€‰0.81 (Standard), 3.00â€‰Â±â€‰0.79 (VN) PÂ =Â 0.93; Reader 3: 4.45â€‰Â±â€‰0.72 (Standard), 4.05â€‰Â±â€‰0.69 (VN) PÂ =Â 0.02; Reader 4: 4.50â€‰Â±â€‰0.69 (Standard), 4.45â€‰Â±â€‰0.76 (VN) PÂ =Â 0.50). In the lesion evaluation study, there was no significant difference in the number of PI-RADS â‰¥3 lesions identified on standard vs. VN bp-MRI (PÂ =Â 0.92, 0.59, 0.87) with similar sensitivity and specificity for clinically significant cancer. The average scan time of the standard clinical biparametric exam was 11.8Â minutes, and this was projected to be 3.2Â minutes for the accelerated exam. DATA CONCLUSION/UNASSIGNED:Diagnostic accelerated biparametric prostate MRI exams can be performed using deep learning methods in <4 minutes, potentially enabling rapid screening prostate MRI. LEVEL OF EVIDENCE/METHODS:3 TECHNICAL EFFICACY: Stage 5.
Estimation of the capillary level input function for dynamic contrast-enhanced MRI of the breast using a deep learning approach
PURPOSE/OBJECTIVE:To develop a deep learning approach to estimate the local capillary-level input function (CIF) for pharmacokinetic model analysis of DCE-MRI. METHODS:A deep convolutional network was trained with numerically simulated data to estimate the CIF. The trained network was tested using simulated lesion data and used to estimate voxel-wise CIF for pharmacokinetic model analysis of breast DCE-MRI data using an abbreviated protocol from women with malignant (n = 25) and benign (n = 28) lesions. The estimated parameters were used to build a logistic regression model to detect the malignancy. RESULT/RESULTS:The pharmacokinetic parameters estimated using the network-predicted CIF from our breast DCE data showed significant differences between the malignant and benign groups for all parameters. Testing the diagnostic performance with the estimated parameters, the conventional approach with arterial input function (AIF) showed an area under the curve (AUC) between 0.76 and 0.87, and the proposed approach with CIF demonstrated similar performance with an AUC between 0.79 and 0.81. CONCLUSION/CONCLUSIONS:This study shows the feasibility of estimating voxel-wise CIF using a deep neural network. The proposed approach could eliminate the need to measure AIF manually without compromising the diagnostic performance to detect the malignancy in the clinical setting.
fastMRI+, Clinical pathology annotations for knee and brain fully sampled magnetic resonance imaging data
Improving speed and image quality of Magnetic Resonance Imaging (MRI) using deep learning reconstruction is an active area of research. The fastMRI dataset contains large volumes of raw MRI data, which has enabled significant advances in this field. While the impact of the fastMRI dataset is unquestioned, the dataset currently lacks clinical expert pathology annotations, critical to addressing clinically relevant reconstruction frameworks and exploring important questions regarding rendering of specific pathology using such novel approaches. This work introduces fastMRI+, which consists of 16154 subspecialist expert bounding box annotations and 13 study-level labels for 22 different pathology categories on the fastMRI knee dataset, and 7570 subspecialist expert bounding box annotations and 643 study-level labels for 30 different pathology categories for the fastMRI brain dataset. The fastMRI+â€‰dataset is open access and aims to support further research and advancement of medical imaging in MRI reconstruction and beyond.
Bayesian Uncertainty Estimation of Learned Variational MRI Reconstruction
Recent deep learning approaches focus on improving quantitative scores of dedicated benchmarks, and therefore only reduce the observation-related (aleatoric) uncertainty. However, the model-immanent (epistemic) uncertainty is less frequently systematically analyzed. In this work, we introduce a Bayesian variational framework to quantify the epistemic uncertainty. To this end, we solve the linear inverse problem of undersampled MRI reconstruction in a variational setting. The associated energy functional is composed of a data fidelity term and the total deep variation (TDV) as a learned parametric regularizer. To estimate the epistemic uncertainty we draw the parameters of the TDV regularizer from a multivariate Gaussian distribution, whose mean and covariance matrix are learned in a stochastic optimal control problem. In several numerical experiments, we demonstrate that our approach yields competitive results for undersampled MRI reconstruction. Moreover, we can accurately quantify the pixelwise epistemic uncertainty, which can serve radiologists as an additional resource to visualize reconstruction reliability.
Virtual mouse brain histology from multi-contrast MRI via deep learning