Search

Your search keyword '"Vasanawala, Shreyas S"' showing total 579 results
Start Over Author "Vasanawala, Shreyas S"
579 results on '"Vasanawala, Shreyas S"'

1. On the Foundation Model for Cardiac MRI Reconstruction

Author

Zhang, Chi, Loecher, Michael, Alkan, Cagan, Yurt, Mahmut, Vasanawala, Shreyas S., and Ennis, Daniel B.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

In recent years, machine learning (ML) based reconstruction has been widely investigated and employed in cardiac magnetic resonance (CMR) imaging. ML-based reconstructions can deliver clinically acceptable image quality under substantially accelerated scans. ML-based reconstruction, however, also requires substantial data and computational time to train the neural network, which is often optimized for a fixed acceleration rate or image contrast. In practice, imaging parameters are often tuned to best suit the diagnosis, which may differ from the training data. This can result in degraded image quality, and multiple trained networks are needed to fulfill the clinical demands. In this study, we propose a foundation model that uses adaptive unrolling, channel-shifting, and Pattern and Contrast-Prompt-UNet (PCP-UNet) to tackle the problem. In particular, the undersampled data goes through a different number of unrolled iterations according to its acceleration rate. Channel-shifting improves reconstructed data quality. The PCP-UNet is equipped with an image contrast and sampling pattern prompt. In vivo CMR experiments were performed using mixed combinations of image contrasts, acceleration rates, and (under)sampling patterns. The proposed foundation model has significantly improved image quality for a wide range of CMR protocols and outperforms the conventional ML-based method., Comment: For MICCAI CMRxRecon Challenge 2024 team CardiAxs

Published
2024

2. AutoSamp: Autoencoding k-space Sampling via Variational Information Maximization for 3D MRI

Author

Alkan, Cagan, Mardani, Morteza, Liao, Congyu, Li, Zhitao, Vasanawala, Shreyas S., and Pauly, John M.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing
Abstract

Accelerated MRI protocols routinely involve a predefined sampling pattern that undersamples the k-space. Finding an optimal pattern can enhance the reconstruction quality, however this optimization is a challenging task. To address this challenge, we introduce a novel deep learning framework, AutoSamp, based on variational information maximization that enables joint optimization of sampling pattern and reconstruction of MRI scans. We represent the encoder as a non-uniform Fast Fourier Transform that allows continuous optimization of k-space sample locations on a non-Cartesian plane, and the decoder as a deep reconstruction network. Experiments on public 3D acquired MRI datasets show improved reconstruction quality of the proposed AutoSamp method over the prevailing variable density and variable density Poisson disc sampling for both compressed sensing and deep learning reconstructions. We demonstrate that our data-driven sampling optimization method achieves 4.4dB, 2.0dB, 0.75dB, 0.7dB PSNR improvements over reconstruction with Poisson Disc masks for acceleration factors of R = 5, 10, 15, 25, respectively. Prospectively accelerated acquisitions with 3D FSE sequences using our optimized sampling patterns exhibit improved image quality and sharpness. Furthermore, we analyze the characteristics of the learned sampling patterns with respect to changes in acceleration factor, measurement noise, underlying anatomy, and coil sensitivities. We show that all these factors contribute to the optimization result by affecting the sampling density, k-space coverage and point spread functions of the learned sampling patterns., Comment: Accepted for publication in IEEE Transactions on Medical Imaging (TMI)

Published
2023
Full Text
View/download PDF

3. Coil Sketching for computationally-efficient MR iterative reconstruction

Author

Oscanoa, Julio A., Ong, Frank, Iyer, Siddharth S., Li, Zhitao, Sandino, Christopher M., Ozturkler, Batu, Ennis, Daniel B., Pilanci, Mert, and Vasanawala, Shreyas S.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing
Abstract

Purpose: Parallel imaging and compressed sensing reconstructions of large MRI datasets often have a prohibitive computational cost that bottlenecks clinical deployment, especially for 3D non-Cartesian acquisitions. One common approach is to reduce the number of coil channels actively used during reconstruction as in coil compression. While effective for Cartesian imaging, coil compression inherently loses signal energy, producing shading artifacts that compromise image quality for 3D non-Cartesian imaging. We propose coil sketching, a general and versatile method for computationally-efficient iterative MR image reconstruction. Theory and Methods: We based our method on randomized sketching algorithms, a type of large-scale optimization algorithms well established in the fields of machine learning and big data analysis. We adapt the sketching theory to the MRI reconstruction problem via a structured sketching matrix that, similar to coil compression, considers high-energy virtual coils obtained from principal component analysis. But, unlike coil compression, it also considers random linear combinations of the remaining low-energy coils, effectively leveraging information from all coils. Results: First, we performed ablation experiments to validate the sketching matrix design on both Cartesian and non-Cartesian datasets. The resulting design yielded both improved computational efficiency and preserved signal-to-noise ratio (SNR) as measured by the inverse g-factor. Then, we verified the efficacy of our approach on high-dimensional non-Cartesian 3D cones datasets, where coil sketching yielded up to three-fold faster reconstructions with equivalent image quality. Conclusion: Coil sketching is a general and versatile reconstruction framework for computationally fast and memory-efficient reconstruction., Comment: 19 pages, 7 figures, 3 tables

Published
2023

4. Automated MRI Field of View Prescription from Region of Interest Prediction by Intra-stack Attention Neural Network

Author

Lei, Ke, Syed, Ali B., Zhu, Xucheng, Pauly, John M., and Vasanawala, Shreyas S.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

Manual prescription of the field of view (FOV) by MRI technologists is variable and prolongs the scanning process. Often, the FOV is too large or crops critical anatomy. We propose a deep-learning framework, trained by radiologists' supervision, for automating FOV prescription. An intra-stack shared feature extraction network and an attention network are used to process a stack of 2D image inputs to generate output scalars defining the location of a rectangular region of interest (ROI). The attention mechanism is used to make the model focus on the small number of informative slices in a stack. Then the smallest FOV that makes the neural network predicted ROI free of aliasing is calculated by an algebraic operation derived from MR sampling theory. We retrospectively collected 595 cases between February 2018 and February 2022. The framework's performance is examined quantitatively with intersection over union (IoU) and pixel error on position, and qualitatively with a reader study. We use the t-test for comparing quantitative results from all models and a radiologist. The proposed model achieves an average IoU of 0.867 and average ROI position error of 9.06 out of 512 pixels on 80 test cases, significantly better (P<0.05) than two baseline models and not significantly different from a radiologist (P>0.12). Finally, the FOV given by the proposed framework achieves an acceptance rate of 92% from an experienced radiologist.

Published
2022

5. Artifact- and content-specific quality assessment for MRI with image rulers

Author

Lei, Ke, Pauly, John M., and Vasanawala, Shreyas S.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Artificial Intelligence, Computer Science - Computer Vision and Pattern Recognition
Abstract

In clinical practice MR images are often first seen by radiologists long after the scan. If image quality is inadequate either patients have to return for an additional scan, or a suboptimal interpretation is rendered. An automatic image quality assessment (IQA) would enable real-time remediation. Existing IQA works for MRI give only a general quality score, agnostic to the cause of and solution to low-quality scans. Furthermore, radiologists' image quality requirements vary with the scan type and diagnostic task. Therefore, the same score may have different implications for different scans. We propose a framework with multi-task CNN model trained with calibrated labels and inferenced with image rulers. Labels calibrated by human inputs follow a well-defined and efficient labeling task. Image rulers address varying quality standards and provide a concrete way of interpreting raw scores from the CNN. The model supports assessments of two of the most common artifacts in MRI: noise and motion. It achieves accuracies of around 90%, 6% better than the best previous method examined, and 3% better than human experts on noise assessment. Our experiments show that label calibration, image rulers, and multi-task training improve the model's performance and generalizability.

Published
2021
Full Text
View/download PDF

6. Deep Learning Automated Background Phase Error Correction for Abdominopelvic 4D Flow MRI.

Author

You, Sophie, Masutani, Evan M, Alley, Marcus T, Vasanawala, Shreyas S, Taub, Pam R, Liau, Joy, Roberts, Anne C, and Hsiao, Albert

Subjects
Abdomen, Humans, Vascular Diseases, Image Interpretation, Computer-Assisted, Magnetic Resonance Imaging, Image Enhancement, Blood Flow Velocity, Retrospective Studies, Middle Aged, Male, Hemodynamics, Deep Learning, Biomedical Imaging, Neurosciences, Clinical Research, Medical and Health Sciences, Nuclear Medicine & Medical Imaging
Abstract

Background Four-dimensional (4D) flow MRI has the potential to provide hemodynamic insights for a variety of abdominopelvic vascular diseases, but its clinical utility is currently impaired by background phase error, which can be challenging to correct. Purpose To assess the feasibility of using deep learning to automatically perform image-based background phase error correction in 4D flow MRI and to compare its effectiveness relative to manual image-based correction. Materials and Methods A convenience sample of 139 abdominopelvic 4D flow MRI acquisitions performed between January 2016 and July 2020 was retrospectively collected. Manual phase error correction was performed using dedicated imaging software and served as the reference standard. After reserving 40 examinations for testing, the remaining examinations were randomly divided into training (86% [85 of 99]) and validation (14% [14 of 99]) data sets to train a multichannel three-dimensional U-Net convolutional neural network. Flow measurements were obtained for the infrarenal aorta, common iliac arteries, common iliac veins, and inferior vena cava. Statistical analyses included Pearson correlation, Bland-Altman analysis, and F tests with Bonferroni correction. Results A total of 139 patients (mean age, 47 years ± 14 [standard deviation]; 108 women) were included. Inflow-outflow correlation improved after manual correction (ρ = 0.94, P < .001) compared with that before correction (ρ = 0.50, P < .001). Automated correction showed similar results (ρ = 0.91, P < .001) and demonstrated very strong correlation with manual correction (ρ = 0.98, P < .001). Both correction methods reduced inflow-outflow variance, improving mean difference from -0.14 L/min (95% limits of agreement: -1.61, 1.32) (uncorrected) to 0.05 L/min (95% limits of agreement: -0.32, 0.42) (manually corrected) and 0.05 L/min (95% limits of agreement: -0.38, 0.49) (automatically corrected). There was no significant difference in inflow-outflow variance between manual and automated correction methods (P = .10). Conclusion Deep learning automated phase error correction reduced inflow-outflow bias and variance of volumetric flow measurements in four-dimensional flow MRI, achieving results comparable with manual image-based phase error correction. © RSNA, 2021 See also the editorial by Roldán-Alzate and Grist in this issue.

Published
2022

7. Memory-efficient Learning for High-Dimensional MRI Reconstruction

Author

Wang, Ke, Kellman, Michael, Sandino, Christopher M., Zhang, Kevin, Vasanawala, Shreyas S., Tamir, Jonathan I., Yu, Stella X., and Lustig, Michael

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

Deep learning (DL) based unrolled reconstructions have shown state-of-the-art performance for under-sampled magnetic resonance imaging (MRI). Similar to compressed sensing, DL can leverage high-dimensional data (e.g. 3D, 2D+time, 3D+time) to further improve performance. However, network size and depth are currently limited by the GPU memory required for backpropagation. Here we use a memory-efficient learning (MEL) framework which favorably trades off storage with a manageable increase in computation during training. Using MEL with multi-dimensional data, we demonstrate improved image reconstruction performance for in-vivo 3D MRI and 2D+time cardiac cine MRI. MEL uses far less GPU memory while marginally increasing the training time, which enables new applications of DL to high-dimensional MRI., Comment: 14 pages, 8 figures

Published
2021

8. Unsupervised MRI Reconstruction with Generative Adversarial Networks

Author

Cole, Elizabeth K., Pauly, John M., Vasanawala, Shreyas S., and Ong, Frank

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Machine Learning, Statistics - Machine Learning
Abstract

Deep learning-based image reconstruction methods have achieved promising results across multiple MRI applications. However, most approaches require large-scale fully-sampled ground truth data for supervised training. Acquiring fully-sampled data is often either difficult or impossible, particularly for dynamic contrast enhancement (DCE), 3D cardiac cine, and 4D flow. We present a deep learning framework for MRI reconstruction without any fully-sampled data using generative adversarial networks. We test the proposed method in two scenarios: retrospectively undersampled fast spin echo knee exams and prospectively undersampled abdominal DCE. The method recovers more anatomical structure compared to conventional methods.

Published
2020

9. multiMap: A Gradient Spoiled Sequence for Simultaneously Measuring B1+, B0, T1/M0, T2, T2*, and Fat Fraction of a Slice

Author

Dwork, Nicholas, Kerr, Adam B., Johnson, Ethan M. I., Baron, Corey A., Vasanawala, Shreyas S., Larson, Peder E. Z., Bush, Adam M., and Pauly, John

Subjects
Electrical Engineering and Systems Science - Image and Video Processing
Abstract

We propose multiMap, a single scan that can generate several quantitative maps simultaneously. The sequence acquires multiple images in a time-efficient manner, which can be modeled for T_2, T2*, main- and transmit-field inhomogeneity, T_1:equilibrium magnetization, and water and fat content. The sequence is constructed so that cross-dependencies between parameters are isolated. Thus, each parameter can be estimated independently. Estimates of all parameters are shown on bottle phantoms, the brain, and the knee. The results are compared to estimates from established techniques.

Published
2020

10. Analysis of Deep Complex-Valued Convolutional Neural Networks for MRI Reconstruction

Author

Cole, Elizabeth K., Cheng, Joseph Y., Pauly, John M., and Vasanawala, Shreyas S.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning, Physics - Medical Physics
Abstract

Many real-world signal sources are complex-valued, having real and imaginary components. However, the vast majority of existing deep learning platforms and network architectures do not support the use of complex-valued data. MRI data is inherently complex-valued, so existing approaches discard the richer algebraic structure of the complex data. In this work, we investigate end-to-end complex-valued convolutional neural networks - specifically, for image reconstruction in lieu of two-channel real-valued networks. We apply this to magnetic resonance imaging reconstruction for the purpose of accelerating scan times and determine the performance of various promising complex-valued activation functions. We find that complex-valued CNNs with complex-valued convolutions provide superior reconstructions compared to real-valued convolutions with the same number of trainable parameters, over a variety of network architectures and datasets.

Published
2020

11. Quantification of the Hemodynamic Changes of Cirrhosis with Free‐Breathing Self‐Navigated MRI

Author

Brunsing, Ryan L, Brown, Dustin, Almahoud, Hashem, Kono, Yuko, Loomba, Rohit, Vodkin, Irene, Sirlin, Claude B, Alley, Marcus T, Vasanawala, Shreyas S, and Hsiao, Albert

Subjects
Digestive Diseases, Chronic Liver Disease and Cirrhosis, Biomedical Imaging, Liver Disease, Blood Flow Velocity, Hemodynamics, Humans, Liver Cirrhosis, Magnetic Resonance Imaging, Portal Vein, Retrospective Studies, 4D flow, liver, cirrhosis, splanchnic, portal vein, TIPS, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging
Abstract

BackgroundNon-invasive assessment of the hemodynamic changes of cirrhosis might help guide management of patients with liver disease but are currently limited.PurposeTo determine whether free-breathing 4D flow MRI can be used to quantify the hemodynamic effects of cirrhosis and introduce hydraulic circuit indexes of severity.Study typeRetrospective.PopulationForty-seven patients including 26 with cirrhosis.Field strength/sequence3 T/free-breathing 4D flow MRI with soft gating and golden-angle view ordering.AssessmentMeasurements of the supra-celiac abdominal aorta, supra-renal abdominal aorta (SRA), celiac trunk (CeT), superior mesenteric artery (SMA), splenic artery (SpA), common hepatic artery (CHA), portal vein (PV), and supra-renal inferior vena cava (IVC) were made by two radiologists. Measures of hepatic vascular resistance (hepatic arterial relative resistance [HARR]; portal resistive index [PRI]) were proposed and calculated.Statistical analysisBland-Altman, Pearson's correlation, Tukey's multiple comparison, and Cohen's kappa. P

Published
2021

12. Accelerating cardiac cine MRI using a deep learning-based ESPIRiT reconstruction

Author

Sandino, Christopher M., Lai, Peng, Vasanawala, Shreyas S., and Cheng, Joseph Y.

Subjects
Electrical Engineering and Systems Science - Signal Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

A novel neural network architecture, known as DL-ESPIRiT, is proposed to reconstruct rapidly acquired cardiac MRI data without field-of-view limitations which are present in previously proposed deep learning-based reconstruction frameworks. Additionally, a novel convolutional neural network based on separable 3D convolutions is integrated into DL-ESPIRiT to more efficiently learn spatiotemporal priors for dynamic image reconstruction. The network is trained on fully-sampled 2D cardiac cine datasets collected from eleven healthy volunteers with IRB approval. DL-ESPIRiT is compared against a state-of-the-art parallel imaging and compressed sensing method known as $l_1$-ESPIRiT. The reconstruction accuracy of both methods is evaluated on retrospectively undersampled datasets (R=12) with respect to standard image quality metrics as well as automatic deep learning-based segmentations of left ventricular volumes. Feasibility of this approach is demonstrated in reconstructions of prospectively undersampled data which were acquired in a single heartbeat per slice., Comment: 29 pages, 9 figures, 1 table, 7 supplementary videos, Submitted to Magnetic Resonance in Medicine

Published
2019

13. Wasserstein GANs for MR Imaging: from Paired to Unpaired Training

Author

Lei, Ke, Mardani, Morteza, Pauly, John M., and Vasanawala, Shreyas S.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

Lack of ground-truth MR images impedes the common supervised training of neural networks for image reconstruction. To cope with this challenge, this paper leverages unpaired adversarial training for reconstruction networks, where the inputs are undersampled k-space and naively reconstructed images from one dataset, and the labels are high-quality images from another dataset. The reconstruction networks consist of a generator which suppresses the input image artifacts, and a discriminator using a pool of (unpaired) labels to adjust the reconstruction quality. The generator is an unrolled neural network -- a cascade of convolutional and data consistency layers. The discriminator is also a multilayer CNN that plays the role of a critic scoring the quality of reconstructed images based on the Wasserstein distance. Our experiments with knee MRI datasets demonstrate that the proposed unpaired training enables diagnostic-quality reconstruction when high-quality image labels are not available for the input types of interest, or when the amount of labels is small. In addition, our adversarial training scheme can achieve better image quality (as rated by expert radiologists) compared with the paired training schemes with pixel-wise loss.

Published
2019
Full Text
View/download PDF

14. Extreme MRI: Large-Scale Volumetric Dynamic Imaging from Continuous Non-Gated Acquisitions

Author

Ong, Frank, Zhu, Xucheng, Cheng, Joseph Y., Johnson, Kevin M., Larson, Peder E. Z., Vasanawala, Shreyas S., and Lustig, Michael

Subjects
Physics - Medical Physics, Electrical Engineering and Systems Science - Image and Video Processing
Abstract

Purpose: To develop a framework to reconstruct large-scale volumetric dynamic MRI from rapid continuous and non-gated acquisitions, with applications to pulmonary and dynamic contrast enhanced (DCE) imaging. Theory and Methods: The problem considered here requires recovering hundred-gigabytes of dynamic volumetric image data from a few gigabytes of k-space data, acquired continuously over several minutes. This reconstruction is vastly under-determined, heavily stressing computing resources as well as memory management and storage. To overcome these challenges, we leverage intrinsic three dimensional (3D) trajectories, such as 3D radial and 3D cones, with ordering that incoherently cover time and k-space over the entire acquisition. We then propose two innovations: (1) A compressed representation using multi-scale low rank matrix factorization that constrains the reconstruction problem, and reduces its memory footprint. (2) Stochastic optimization to reduce computation, improve memory locality, and minimize communications between threads and processors. We demonstrate the feasibility of the proposed method on DCE imaging acquired with a golden-angle ordered 3D cones trajectory and pulmonary imaging acquired with a bit-reversed ordered 3D radial trajectory. We compare it with "soft-gated" dynamic reconstruction for DCE and respiratory resolved reconstruction for pulmonary imaging. Results: The proposed technique shows transient dynamics that are not seen in gating based methods. When applied to datasets with irregular, or non-repetitive motions, the proposed method displays sharper image features. Conclusion: We demonstrated a method that can reconstruct massive 3D dynamic image series in the extreme undersampling and extreme computation setting., Comment: Accepted to Magnetic Resonance in Medicine

Published
2019

15. Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning

Author

Cheng, Joseph Y., Chen, Feiyu, Sandino, Christopher, Mardani, Morteza, Pauly, John M., and Vasanawala, Shreyas S.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning, Physics - Medical Physics, Statistics - Machine Learning
Abstract

Compressed sensing in MRI enables high subsampling factors while maintaining diagnostic image quality. This technique enables shortened scan durations and/or improved image resolution. Further, compressed sensing can increase the diagnostic information and value from each scan performed. Overall, compressed sensing has significant clinical impact in improving the diagnostic quality and patient experience for imaging exams. However, a number of challenges exist when moving compressed sensing from research to the clinic. These challenges include hand-crafted image priors, sensitive tuning parameters, and long reconstruction times. Data-driven learning provides a solution to address these challenges. As a result, compressed sensing can have greater clinical impact. In this tutorial, we will review the compressed sensing formulation and outline steps needed to transform this formulation to a deep learning framework. Supplementary open source code in python will be used to demonstrate this approach with open databases. Further, we will discuss considerations in applying data-driven compressed sensing in the clinical setting., Comment: Submitted to the Special Issue on Computational MRI: Compressed Sensing and Beyond in the IEEE Signal Processing Magazine

Published
2019

16. Deep Residual Network for Off-Resonance Artifact Correction with Application to Pediatric Body Magnetic Resonance Angiography with 3D Cones

Author

Zeng, David Y, Shaikh, Jamil, Nishimura, Dwight G, Vasanawala, Shreyas S, and Cheng, Joseph Y

Subjects
Computer Science - Computer Vision and Pattern Recognition, Physics - Medical Physics
Abstract

Purpose: Off-resonance artifact correction by deep-learning, to facilitate rapid pediatric body imaging with a scan time efficient 3D cones trajectory. Methods: A residual convolutional neural network to correct off-resonance artifacts (Off-ResNet) was trained with a prospective study of 30 pediatric magnetic resonance angiography exams. Each exam acquired a short-readout scan (1.18 ms +- 0.38) and a long-readout scan (3.35 ms +- 0.74) at 3T. 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 to short-readout scans by normalized root-mean-square error (NRMSE), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR). Scans were also compared by scoring on eight anatomical features by two radiologists, using analysis of variance with post-hoc Tukey's test. Reader agreement was determined with intraclass correlation. Results: Long-readout scans were on average 59.3% shorter than short-readout scans. Images from Off-ResNet had superior NRMSE, SSIM, and PSNR compared to uncorrected images across +-1kHz off-resonance (P<0.01). The proposed method had superior NRMSE over -677Hz to +1kHz and superior SSIM and PSNR over +-1kHz compared to autofocus (P<0.01). Radiologic scoring demonstrated that long-readout scans corrected with Off-ResNet were non-inferior to short-readout scans (P<0.01). Conclusion: The proposed method can correct off-resonance artifacts from rapid long-readout 3D cones scans to a non-inferior image quality compared to diagnostically standard short-readout scans.

Published
2018

17. Clinically Deployed Distributed Magnetic Resonance Imaging Reconstruction: Application to Pediatric Knee Imaging

Author

Anderson, Michael J., Tamir, Jonathan I., Turek, Javier S., Alley, Marcus T., Willke, Theodore L., Vasanawala, Shreyas S., and Lustig, Michael

Subjects
Physics - Medical Physics, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Distributed, Parallel, and Cluster Computing, 68W15, 68U10
Abstract

Magnetic resonance imaging is capable of producing volumetric images without ionizing radiation. Nonetheless, long acquisitions lead to prohibitively long exams. Compressed sensing (CS) can enable faster scanning via sub-sampling with reduced artifacts. However, CS requires significantly higher reconstruction computation, limiting current clinical applications to 2D/3D or limited-resolution dynamic imaging. Here we analyze the practical limitations to T2 Shuffling, a four-dimensional CS-based acquisition, which provides sharp 3D-isotropic-resolution and multi-contrast images in a single scan. Our improvements to the pipeline on a single machine provide a 3x overall reconstruction speedup, which allowed us to add algorithmic changes improving image quality. Using four machines, we achieved additional 2.1x improvement through distributed parallelization. Our solution reduced the reconstruction time in the hospital to 90 seconds on a 4-node cluster, enabling its use clinically. To understand the implications of scaling this application, we simulated running our reconstructions with a multiple scanner setup typical in hospitals.

Published
2018

18. Highly Scalable Image Reconstruction using Deep Neural Networks with Bandpass Filtering

Author

Cheng, Joseph Y., Chen, Feiyu, Alley, Marcus T., Pauly, John M., and Vasanawala, Shreyas S.

Subjects
Computer Science - Computer Vision and Pattern Recognition, Physics - Medical Physics
Abstract

To increase the flexibility and scalability of deep neural networks for image reconstruction, a framework is proposed based on bandpass filtering. For many applications, sensing measurements are performed indirectly. For example, in magnetic resonance imaging, data are sampled in the frequency domain. The introduction of bandpass filtering enables leveraging known imaging physics while ensuring that the final reconstruction is consistent with actual measurements to maintain reconstruction accuracy. We demonstrate this flexible architecture for reconstructing subsampled datasets of MRI scans. The resulting high subsampling rates increase the speed of MRI acquisitions and enable the visualization rapid hemodynamics., Comment: 9 pages, 10 figures

Published
2018

19. Development and Clinical Implementation of Next Generation Very Light Weight and Extremely Flexible Receiver Arrays for Pediatric MRI

Author

Vasanawala, Shreyas S., Stormont, Robert, Lindsay, Scott, Grafendorfer, Thomas, Cheng, Joseph Y., Pauly, John M., Scott, Greig, Guzman, Jorge X., Taracila, Victor, Chirayath, Dan, and Robb, Fraser

Subjects
Physics - Medical Physics
Abstract

We develop a novel next generation light-weight highly flexible pediatric coil array, combine it with a high-density pediatric posterior array or conventional posterior phased array, and determine feasibility of pediatric clinical use. A highly flexible 16 element MRI receiver coil was constructed with low-profile noise controlling preamplifiers that minimized reactive and resistive coupling. Element decoupling was assessed in flat and highly flexed states. With IRB approval and informed consent and assent, 24 consecutive subjects undergoing torso or extremity MRI were prospectively recruited. Care team members were surveyed on preference for the coil versus conventional coils and diagnostic acceptability of the images was recorded. Confidence interval of proportion of diagnostic exams was calculated. The array without cable weighed 480 grams, demonstrated good flexibility while maintaining element decoupling. The coil was preferred by all nurses and anesthesiologists involved in the care of the patients. Technologists preferred the coil in 96 percent of cases, and 23 of 24 exams were diagnostically adequate with 85 percent confidence interval of 90-100 percent. Light-weight highly flexible coil arrays can be constructed that maintain element decoupling. Pediatric clinical image quality is likely to be diagnostic, with acceptance by members of the care team., Comment: 22 pages, 7 figures, 2 tables

Published
2017

20. 4D flow MRI quantification of mitral and tricuspid regurgitation: Reproducibility and consistency relative to conventional MRI

Author

Feneis, Jennifer F, Kyubwa, Espoir, Atianzar, Kimberly, Cheng, Joseph Y, Alley, Marcus T, Vasanawala, Shreyas S, Demaria, Anthony N, and Hsiao, Albert

Subjects
Heart Disease, Biomedical Imaging, Cardiovascular, Clinical Research, Adult, Aged, Aged, 80 and over, Blood Flow Velocity, Female, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Imaging, Three-Dimensional, Magnetic Resonance Imaging, Male, Middle Aged, Mitral Valve, Mitral Valve Insufficiency, Observer Variation, Reproducibility of Results, Retrospective Studies, Stroke Volume, Time Factors, Tricuspid Valve, Tricuspid Valve Insufficiency, Young Adult, mitral regurgitation, tricuspid regurgitation, cardiac MRI, 4D flow MRI, quantification, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging
Abstract

BackgroundIn patients with mitral or tricuspid valve regurgitation, evaluation of regurgitant severity is essential for determining the need for surgery. While transthoracic echocardiography is widely accessible, it has limited reproducibility for grading inlet valve regurgitation. Multiplanar cardiac MRI is the quantitative standard but requires specialized local expertise, and is thus not widely available. Volumetric 4D flow MRI has potential for quantitatively grading the severity of inlet valve regurgitation in adult patients.PurposeTo evaluate the accuracy and reproducibility of volumetric 4D flow MRI for quantification of inlet valvular regurgitation compared to conventional multiplanar MRI, which may simplify and improve accessibility of cardiac MRI.Study typeThis retrospective, HIPAA-compliant imaging-based comparison study was conducted at a single institution.SubjectsTwenty-one patients who underwent concurrent multiplanar and 4D flow cardiac MRI between April 2015 and January 2017.Field strength/sequences3T; steady-state free-precession (SSFP), 2D phase contrast (2D-PC), and postcontrast 4D flow.AssessmentWe evaluated the intertechnique (4D flow vs. 2D-PC), intermethod (direct vs. indirect measurement), interobserver and intraobserver reproducibility of measurements of regurgitant flow volume (RFV), fraction (RF), and volume (RVol).Statistical testsStatistical analysis included Pearson correlation, Bland-Altman statistics, and intraclass correlation coefficients.ResultsThere was high concordance between 4D flow and multiplanar MRI, whether using direct or indirect methods of quantifying regurgitation (r = 0.813-0.985). Direct interrogation of the regurgitant jet with 4D flow showed high intraobserver consistency (r = 0.976-0.999) and interobserver consistency (r = 0.861-0.992), and correlated well with traditional indirect measurements obtained as the difference between stroke volume and forward outlet valve flow.Data conclusion4D flow MRI provides highly reproducible measurements of mitral and tricuspid regurgitant volume, and may be used in place of conventional multiplanar MRI.Level of evidence4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;48:1147-1158.

Published
2018

21. Upstream Machine Learning in Radiology

Author

Sandino, Christopher M., Cole, Elizabeth K., Alkan, Cagan, Chaudhari, Akshay S., Loening, Andreas M., Hyun, Dongwoon, Dahl, Jeremy, Imran, Abdullah-Al-Zubaer, Wang, Adam S., and Vasanawala, Shreyas S.

Published
2021
Full Text
View/download PDF

22. The impact of computed high b-value images on the diagnostic accuracy of DWI for prostate cancer: A receiver operating characteristics analysis.

Author

Ning, Peigang, Shi, Dapeng, Sonn, Geoffrey A, Vasanawala, Shreyas S, Loening, Andreas M, Ghanouni, Pejman, Obara, Piotr, Shin, Lewis K, Fan, Richard E, Hargreaves, Brian A, and Daniel, Bruce L

Subjects
Prostate, Humans, Prostatic Neoplasms, Image Interpretation, Computer-Assisted, Diffusion Magnetic Resonance Imaging, Area Under Curve, Sensitivity and Specificity, Retrospective Studies, ROC Curve, Aged, Middle Aged, Male, Image-Guided Biopsy, Image Interpretation, Computer-Assisted
Abstract

To evaluate the performance of computed high b value diffusion-weighted images (DWI) in prostate cancer detection. 97 consecutive patients who had undergone multiparametric MRI of the prostate followed by biopsy were reviewed. Five radiologists independently scored 138 lesions on native high b-value images (b = 1200 s/mm2), apparent diffusion coefficient (ADC) maps, and computed high b-value images (contrast equivalent to b = 2000 s/mm2) to compare their diagnostic accuracy. Receiver operating characteristic (ROC) analysis and McNemar's test were performed to assess the relative performance of computed high b value DWI, native high b-value DWI and ADC maps. No significant difference existed in the area under the curve (AUC) for ROCs comparing B1200 (b = 1200 s/mm2) to computed B2000 (c-B2000) in 5 readers. In 4 of 5 readers c-B2000 had significantly increased sensitivity and/or decreased specificity compared to B1200 (McNemar's p

Published
2018

23. Pelvic Blood Flow Predicts Fibroid Volume and Embolic Required for Uterine Fibroid Embolization: A Pilot Study With 4D Flow MR Angiography.

Author

Malone, Christopher D, Banerjee, Arjun, Alley, Marcus T, Vasanawala, Shreyas S, Roberts, Anne C, and Hsiao, Albert

Subjects
Fibroid Tumors (Uterine), Clinical Research, Biomedical Imaging, Adult, Embolization, Therapeutic, Female, Humans, Image Enhancement, Leiomyoma, Magnetic Resonance Angiography, Middle Aged, Pelvis, Pilot Projects, Predictive Value of Tests, Retrospective Studies, Treatment Outcome, Uterine Neoplasms, flow MRI, MR angiography, uterine fibroid embolization, vascular and interventional radiology, Clinical Sciences, Nuclear Medicine & Medical Imaging
Abstract

OBJECTIVE:We report here an initial experience using 4D flow MRI in pelvic imaging-specifically, in imaging uterine fibroids. We hypothesized that blood flow might correlate with fibroid volume and that quantifying blood flow might help to predict the amount of embolic required to achieve stasis at subsequent uterine fibroid embolization (UFE). MATERIALS AND METHODS:Thirty-three patients with uterine fibroids and seven control subjects underwent pelvic MRI with 4D flow imaging. Of the patients with fibroids, 10 underwent 4D flow imaging before UFE and seven after UFE; in the remaining 16 patients with fibroids, UFE had yet to be performed. Four-dimensional flow measurements were performed using Arterys CV Flow. The flow fraction of the internal iliac artery was expressed as the ratio of internal iliac artery flow to external iliac artery flow and was compared between groups. The flow ratios between the internal iliac arteries on each side were calculated. Fibroid volume versus internal iliac flow fraction, embolic volume versus internal iliac flow fraction, and embolic volume ratio between sides versus the ratio of internal iliac artery flows between sides were compared. RESULTS:The mean internal iliac flow fraction was significantly higher in the 26 patients who underwent imaging before UFE (mean ± standard error, 0.78 ± 0.06) than in the seven patients who underwent imaging after UFE (0.48 ± 0.07, p < 0.01) and in the seven control patients without fibroids (0.48 ± 0.08, p < 0.0001). The internal iliac flow fraction correlated well with fibroid volumes before UFE (r = 0.7754, p < 0.0001) and did not correlate with fibroid volumes after UFE (r = -0.3051, p = 0.51). The ratio of embolic required to achieve stasis between sides showed a modest correlation with the ratio of internal iliac flow (r = 0.6776, p = 0.03). CONCLUSION:Internal iliac flow measured by 4D flow MRI correlates with fibroid volume and is predictive of the ratio of embolic required to achieve stasis on each side at subsequent UFE and may be useful for preprocedural evaluation of patients with uterine fibroids.

Published
2018

24. AutoSamp: Autoencoding k-Space Sampling via Variational Information Maximization for 3D MRI

Author

Alkan, Cagan, Mardani, Morteza, Liao, Congyu, Li, Zhitao, Vasanawala, Shreyas S., and Pauly, John M.

Abstract

Accelerated MRI protocols routinely involve a predefined sampling pattern that undersamples the k-space. Finding an optimal pattern can enhance the reconstruction quality, however this optimization is a challenging task. To address this challenge, we introduce a novel deep learning framework, AutoSamp, based on variational information maximization that enables joint optimization of sampling pattern and reconstruction of MRI scans. We represent the encoder as a non-uniform Fast Fourier Transform that allows continuous optimization of k-space sample locations on a non-Cartesian plane, and the decoder as a deep reconstruction network. Experiments on public 3D acquired MRI datasets show improved reconstruction quality of the proposed AutoSamp method over the prevailing variable density and variable density Poisson disc sampling for both compressed sensing and deep learning reconstructions. We demonstrate that our data-driven sampling optimization method achieves 4.4dB, 2.0dB, 0.75dB, 0.7dB PSNR improvements over reconstruction with Poisson Disc masks for acceleration factors of R =5, 10, 15, 25, respectively. Prospectively accelerated acquisitions with 3D FSE sequences using our optimized sampling patterns exhibit improved image quality and sharpness. Furthermore, we analyze the characteristics of the learned sampling patterns with respect to changes in acceleration factor, measurement noise, underlying anatomy, and coil sensitivities. We show that all these factors contribute to the optimization result by affecting the sampling density, k-space coverage and point spread functions of the learned sampling patterns.

Published
2025
Full Text
View/download PDF

25. Current and potential imaging applications of ferumoxytol for magnetic resonance imaging

Author

Toth, Gerda B, Varallyay, Csanad G, Horvath, Andrea, Bashir, Mustafa R, Choyke, Peter L, Daldrup-Link, Heike E, Dosa, Edit, Finn, John Paul, Gahramanov, Seymur, Harisinghani, Mukesh, Macdougall, Iain, Neuwelt, Alexander, Vasanawala, Shreyas S, Ambady, Prakash, Barajas, Ramon, Cetas, Justin S, Ciporen, Jeremy, DeLoughery, Thomas J, Doolittle, Nancy D, Fu, Rongwei, Grinstead, John, Guimaraes, Alexander R, Hamilton, Bronwyn E, Li, Xin, McConnell, Heather L, Muldoon, Leslie L, Nesbit, Gary, Netto, Joao P, Petterson, David, Rooney, William D, Schwartz, Daniel, Szidonya, Laszlo, and Neuwelt, Edward A

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Cardiovascular, Biomedical Imaging, Neurosciences, 4.1 Discovery and preclinical testing of markers and technologies, Detection, screening and diagnosis, Adolescent, Adult, Animals, Atlases as Topic, Child, Preschool, Contrast Media, Female, Ferrosoferric Oxide, Hematinics, Humans, Kidney, Magnetic Resonance Imaging, Male, Middle Aged, Predictive Value of Tests, Renal Elimination, Renal Insufficiency, Chronic, Reproducibility of Results, chronic kidney disease, magnetic resonance imaging, nephrotoxicity, Urology & Nephrology, Clinical sciences
Abstract

Contrast-enhanced magnetic resonance imaging is a commonly used diagnostic tool. Compared with standard gadolinium-based contrast agents, ferumoxytol (Feraheme, AMAG Pharmaceuticals, Waltham, MA), used as an alternative contrast medium, is feasible in patients with impaired renal function. Other attractive imaging features of i.v. ferumoxytol include a prolonged blood pool phase and delayed intracellular uptake. With its unique pharmacologic, metabolic, and imaging properties, ferumoxytol may play a crucial role in future magnetic resonance imaging of the central nervous system, various organs outside the central nervous system, and the cardiovascular system. Preclinical and clinical studies have demonstrated the overall safety and effectiveness of this novel contrast agent, with rarely occurring anaphylactoid reactions. The purpose of this review is to describe the general and organ-specific properties of ferumoxytol, as well as the advantages and potential pitfalls associated with its use in magnetic resonance imaging. To more fully demonstrate the applications of ferumoxytol throughout the body, an imaging atlas was created and is available online as supplementary material.

Published
2017

27. T2 shuffling: Sharp, multicontrast, volumetric fast spin‐echo imaging

Author

Tamir, Jonathan I, Uecker, Martin, Chen, Weitian, Lai, Peng, Alley, Marcus T, Vasanawala, Shreyas S, and Lustig, Michael

Subjects
Engineering, Biomedical Engineering, Clinical Research, Bioengineering, 4.1 Discovery and preclinical testing of markers and technologies, Adult, Child, Foot, Humans, Imaging, Three-Dimensional, Magnetic Resonance Imaging, Signal-To-Noise Ratio, fast spin-echo, T-2 imaging, compressed sensing, multicontrast, T2 imaging, Nuclear Medicine & Medical Imaging, Biomedical engineering
Abstract

PurposeA new acquisition and reconstruction method called T2 Shuffling is presented for volumetric fast spin-echo (three-dimensional [3D] FSE) imaging. T2 Shuffling reduces blurring and recovers many images at multiple T2 contrasts from a single acquisition at clinically feasible scan times (6-7 min).Theory and methodsThe parallel imaging forward model is modified to account for temporal signal relaxation during the echo train. Scan efficiency is improved by acquiring data during the transient signal decay and by increasing echo train lengths without loss in signal-to-noise ratio (SNR). By (1) randomly shuffling the phase encode view ordering, (2) constraining the temporal signal evolution to a low-dimensional subspace, and (3) promoting spatio-temporal correlations through locally low rank regularization, a time series of virtual echo time images is recovered from a single scan. A convex formulation is presented that is robust to partial voluming and radiofrequency field inhomogeneity.ResultsRetrospective undersampling and in vivo scans confirm the increase in sharpness afforded by T2 Shuffling. Multiple image contrasts are recovered and used to highlight pathology in pediatric patients. A proof-of-principle method is integrated into a clinical musculoskeletal imaging workflow.ConclusionThe proposed T2 Shuffling method improves the diagnostic utility of 3D FSE by reducing blurring and producing multiple image contrasts from a single scan. Magn Reson Med 77:180-195, 2017. © 2016 Wiley Periodicals, Inc.

Published
2017

29. Safety and technique of ferumoxytol administration for MRI.

Author

Vasanawala, Shreyas S, Nguyen, Kim-Lien, Hope, Michael D, Bridges, Mellena D, Hope, Thomas A, Reeder, Scott B, and Bashir, Mustafa R

Subjects
Macrophages, Humans, Neoplasms, Neoplasm Metastasis, Iodine, Gadolinium, Ferric Compounds, Contrast Media, Magnetic Resonance Imaging, Perfusion, Ferrosoferric Oxide, Clinical Trials as Topic, Patient Safety, contrast agent, ferumoxytol, iron oxide, safety, Biomedical Imaging, Biomedical Engineering, Nuclear Medicine & Medical Imaging
Abstract

Ferumoxytol is an ultrasmall superparamagnetic iron oxide agent marketed for the treatment of anemia. There has been increasing interest in its properties as an MRI contrast agent as well as greater awareness of its adverse event profile. This mini-review summarizes the current state of knowledge of the risks of ferumoxytol and methods of administration.

Published
2016

30. Qualitative grading of aortic regurgitation: a pilot study comparing CMR 4D flow and echocardiography

Author

Chelu, Raluca G, van den Bosch, Annemien E, van Kranenburg, Matthijs, Hsiao, Albert, van den Hoven, Allard T, Ouhlous, Mohamed, Budde, Ricardo PJ, Beniest, Kirsten M, Swart, Laurens E, Coenen, Adriaan, Lubbers, Marisa M, Wielopolski, Piotr A, Vasanawala, Shreyas S, Roos-Hesselink, Jolien W, and Nieman, Koen

Subjects
Biomedical and Clinical Sciences, Cardiovascular Medicine and Haematology, Clinical Sciences, Clinical Research, Cardiovascular, Biomedical Imaging, Heart Disease, Detection, screening and diagnosis, 4.2 Evaluation of markers and technologies, Adult, Aortic Valve, Aortic Valve Insufficiency, Female, Humans, Imaging, Three-Dimensional, Magnetic Resonance Imaging, Male, Pilot Projects, Reproducibility of Results, Sensitivity and Specificity, Severity of Illness Index, Cardiac, Phase contrast, CMR 4D flow imaging, Eddy currents correction, Aortic regurgitation, Flow visualization, Cardiorespiratory Medicine and Haematology, Nuclear Medicine & Medical Imaging, Cardiovascular medicine and haematology
Abstract

Over the past 10 years there has been intense research in the development of volumetric visualization of intracardiac flow by cardiac magnetic resonance (CMR).This volumetric time resolved technique called CMR 4D flow imaging has several advantages over standard CMR. It offers anatomical, functional and flow information in a single free-breathing, ten-minute acquisition. However, the data obtained is large and its processing requires dedicated software. We evaluated a cloud-based application package that combines volumetric data correction and visualization of CMR 4D flow data, and assessed its accuracy for the detection and grading of aortic valve regurgitation using transthoracic echocardiography as reference. Between June 2014 and January 2015, patients planned for clinical CMR were consecutively approached to undergo the supplementary CMR 4D flow acquisition. Fifty four patients(median age 39 years, 32 males) were included. Detection and grading of the aortic valve regurgitation using CMR4D flow imaging were evaluated against transthoracic echocardiography. The agreement between 4D flow CMR and transthoracic echocardiography for grading of aortic valve regurgitation was good (j = 0.73). To identify relevant,more than mild aortic valve regurgitation, CMR 4D flow imaging had a sensitivity of 100 % and specificity of 98 %. Aortic regurgitation can be well visualized, in a similar manner as transthoracic echocardiography, when using CMR 4D flow imaging.

Published
2016

32. Improved quantification and mapping of anomalous pulmonary venous flow with four‐dimensional phase‐contrast MRI and interactive streamline rendering

Author

Hsiao, Albert, Yousaf, Ufra, Alley, Marcus T, Lustig, Michael, Chan, Frandics Pak, Newman, Beverley, and Vasanawala, Shreyas S

Subjects
Biomedical and Clinical Sciences, Cardiovascular Medicine and Haematology, Clinical Sciences, Heart Disease, Bioengineering, Cardiovascular, Biomedical Imaging, Pediatric, Clinical Research, 4.2 Evaluation of markers and technologies, Detection, screening and diagnosis, Adolescent, Adult, Algorithms, Blood Flow Velocity, Child, Child, Preschool, Humans, Image Enhancement, Imaging, Three-Dimensional, Infant, Magnetic Resonance Angiography, Observer Variation, Pulmonary Veins, Reproducibility of Results, Sensitivity and Specificity, User-Computer Interface, Young Adult, flow, pulmonary, shunt, structural, veins, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences
Abstract

BackgroundCardiac MRI is routinely performed for quantification of shunt flow in patients with anomalous pulmonary veins, but can be technically-challenging to perform. Four-dimensional phase-contrast (4D-PC) MRI has potential to simplify this exam. We sought to determine whether 4D-PC may be a viable clinical alternative to conventional 2D phase-contrast MR imaging.MethodsWith institutional review board approval and HIPAA-compliance, we retrospectively identified all patients with anomalous pulmonary veins who underwent cardiac MRI at either 1.5 Tesla (T) or 3T with parallel-imaging compressed-sensing (PI-CS) 4D-PC between April, 2011 and October, 2013. A total of 15 exams were included (10 male, 5 female). Algorithms for interactive streamline visualization were developed and integrated into in-house software. Blood flow was measured at the valves, pulmonary arteries and veins, cavae, and any associated shunts. Pulmonary veins were mapped to their receiving atrial chamber with streamlines. The intraobserver, interobserver, internal consistency of flow measurements, and consistency with conventional MRI were then evaluated with Pearson correlation and Bland-Altman analysis.ResultsTriplicate measurements of blood flow from 4D-PC were highly consistent, particularly at the aortic and pulmonary valves (cv 2-3%). Flow measurements were reproducible by a second observer (ρ = 0.986-0.999). Direct measurements of shunt volume from anomalous veins and intracardiac shunts matched indirect estimates from the outflow valves (ρ = 0.966). Measurements of shunt fraction using 4D-PC using any approach were more consistent with ventricular volumetric displacements than conventional 2D-PC (ρ = 0.972-0.991 versus 0.929).ConclusionShunt flow may be reliably quantified with 4D-PC MRI, either indirectly or with detailed delineation of flow from multiple shunts. The 4D-PC may be a more accurate alternative to conventional MRI.

Published
2015

33. Congenital heart disease assessment with 4D flow MRI

Author

Vasanawala, Shreyas S, Hanneman, Kate, Alley, Marcus T, and Hsiao, Albert

Subjects
Biomedical and Clinical Sciences, Cardiovascular Medicine and Haematology, Clinical Sciences, Biomedical Imaging, Pediatric Research Initiative, Congenital Structural Anomalies, Heart Disease, Pediatric, Cardiovascular, Clinical Research, Algorithms, Cardiac-Gated Imaging Techniques, Heart Defects, Congenital, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Imaging, Three-Dimensional, Magnetic Resonance Angiography, Magnetic Resonance Imaging, Cine, Myocardial Perfusion Imaging, Reproducibility of Results, Sensitivity and Specificity, 4D flow, cardiac, congenital, phase contrast, time-resolved, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences
Abstract

With improvements in surgical and medical management, patients with congenital heart disease (CHD) are often living well into adulthood. MRI provides critical data for diagnosis and monitoring of these patients, yielding information on cardiac anatomy, blood flow, and cardiac function. Though historically these exams have been complex and lengthy, four-dimensional (4D) flow is emerging as a single fast technique for comprehensive assessment of CHD. The 4D flow consists of a volumetric time-resolved acquisition that is gated to the cardiac cycle, providing a time-varying vector field of blood flow as well as registered anatomic images. In this article, we provide an overview of MRI evaluation of congenital heart disease by means of example of three relatively common representative conditions: tetralogy of Fallot, aortic coarctation, and anomalous pulmonary venous drainage. Then 4D flow data acquisition, data correction, and postprocessing techniques are reviewed. We conclude with several examples that highlight the comprehensive nature of the evaluation of congenital heart disease with 4D flow.

Published
2015

34. Clinical performance of a free-breathing spatiotemporally accelerated 3-D time-resolved contrast-enhanced pediatric abdominal MR angiography

Author

Zhang, Tao, Yousaf, Ufra, Hsiao, Albert, Cheng, Joseph Y, Alley, Marcus T, Lustig, Michael, Pauly, John M, and Vasanawala, Shreyas S

Subjects
Biomedical Imaging, Pediatric, Clinical Research, Detection, screening and diagnosis, 4.2 Evaluation of markers and technologies, Abdomen, Adolescent, Arteries, Child, Child, Preschool, Contrast Media, Female, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Imaging, Three-Dimensional, Infant, Infant, Newborn, Magnetic Resonance Angiography, Male, Reproducibility of Results, Respiratory Mechanics, Sensitivity and Specificity, Spatio-Temporal Analysis, Children, Compressed sensing, Magnetic resonance angiography, Parallel imaging, Spatiotemporal acceleration, Paediatrics and Reproductive Medicine, Nuclear Medicine & Medical Imaging
Abstract

BackgroundPediatric contrast-enhanced MR angiography is often limited by respiration, other patient motion and compromised spatiotemporal resolution.ObjectiveTo determine the reliability of a free-breathing spatiotemporally accelerated 3-D time-resolved contrast-enhanced MR angiography method for depicting abdominal arterial anatomy in young children.Materials and methodsWith IRB approval and informed consent, we retrospectively identified 27 consecutive children (16 males and 11 females; mean age: 3.8 years, range: 14 days to 8.4 years) referred for contrast-enhanced MR angiography at our institution, who had undergone free-breathing spatiotemporally accelerated time-resolved contrast-enhanced MR angiography studies. A radio-frequency-spoiled gradient echo sequence with Cartesian variable density k-space sampling and radial view ordering, intrinsic motion navigation and intermittent fat suppression was developed. Images were reconstructed with soft-gated parallel imaging locally low-rank method to achieve both motion correction and high spatiotemporal resolution. Quality of delineation of 13 abdominal arteries in the reconstructed images was assessed independently by two radiologists on a five-point scale. Ninety-five percent confidence intervals of the proportion of diagnostically adequate cases were calculated. Interobserver agreements were also analyzed.ResultsEleven out of 13 arteries achieved acceptable image quality (mean score range: 3.9-5.0) for both readers. Fair to substantial interobserver agreement was reached on nine arteries.ConclusionFree-breathing spatiotemporally accelerated 3-D time-resolved contrast-enhanced MR angiography frequently yields diagnostic image quality for most abdominal arteries in young children.

Published
2015

35. Ferumoxytol as an off-label contrast agent in body 3T MR angiography: a pilot study in children

Author

Ruangwattanapaisarn, Nichanan, Hsiao, Albert, and Vasanawala, Shreyas S

Subjects
Biomedical and Clinical Sciences, Cardiovascular Medicine and Haematology, Clinical Sciences, Pediatric, Biomedical Imaging, Clinical Research, Cardiovascular, Adolescent, Cardiac-Gated Imaging Techniques, Child, Child, Preschool, Contrast Media, Feasibility Studies, Female, Ferrosoferric Oxide, Humans, Infant, Infant, Newborn, Magnetic Resonance Angiography, Male, Off-Label Use, Pilot Projects, Retrospective Studies, Signal-To-Noise Ratio, Paediatrics and Reproductive Medicine, Nuclear Medicine & Medical Imaging, Clinical sciences, Paediatrics
Abstract

BackgroundFerumoxytol is an ultrasmall superparamagnetic iron oxide (USPIO) nanoparticle agent used to treat iron deficiency anemia in adults with chronic kidney disease.ObjectiveWe aim to determine the feasibility of using ferumoxytol for clinical pediatric cardiac and vascular imaging.Material and methodsWe retrospectively identified 23 consecutive children who underwent MRI with ferumoxytol (11 males; mean age: 7.4 years, range: 3 days-18 years), yielding 12 abdominal MR angiography and 15 cardiac MRI studies. Medical records were reviewed for the clinical indication, ferumoxytol dose, injection rate, sedation and any complication. A two-reader consensus scored the images on a five-point scale for overall image quality and delineation of various anatomical structures. Signal-to-background ratios for abdominal aorta and inferior vena cava for abdominal cases and blood pool-myocardium contrast ratios for cardiac cases were calculated. The confidence intervals for obtaining a score of three or above for each image parameter were calculated by using adjusted Wald method.ResultsFor abdominal MR angiography, average scores for overall image quality, as well as delineation of the hepatic artery, superior mesenteric artery, renal artery and veins were 4.5, 4.3, 4.3, 3.7 and 4.7, respectively. For cardiac exams, the average scores for overall image quality, systemic arteries, pulmonary arteries, pulmonary veins, valves and ventricles were 4.4, 4.6, 4.1, 4.8, 4.1 and 4.7, respectively. For all parameters, the lower bound for the proportion of cases to have a score of 3 or above was 65%. Signal-to-background ratios for aorta and abdominal veins averaged 86 +/- 74 and 86 +/- 77 for full-dose images, and 23 and 18 for half-dose images, respectively. Mean blood pool to myocardium contrast ratio was 3:3.ConclusionFerumoxytol can provide excellent image quality for pediatric body MR angiography/MR venography at a dose of 1.5 or 3 mg Fe/kg. Further investigation should be directed toward understanding the lowest dose that can be administered.

Published
2015

36. Inlet and outlet valve flow and regurgitant volume may be directly and reliably quantified with accelerated, volumetric phase‐contrast MRI

Author

Hsiao, Albert, Tariq, Umar, Alley, Marcus T, Lustig, Michael, and Vasanawala, Shreyas S

Subjects
Biomedical and Clinical Sciences, Cardiovascular Medicine and Haematology, Clinical Sciences, Heart Disease, Clinical Research, Congenital Structural Anomalies, Cardiovascular, Pediatric, Biomedical Imaging, Adolescent, Algorithms, Child, Child, Preschool, Feasibility Studies, Female, Heart Defects, Congenital, Heart Valve Diseases, Humans, Image Interpretation, Computer-Assisted, Infant, Magnetic Resonance Angiography, Male, Retrospective Studies, flow, structural, tricuspid, mitral, regurgitation, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences
Abstract

PurposeTo determine whether it is feasible to use solely an accelerated 4D phase-contrast magnetic resonance imaging (4D-PC MRI) acquisition to quantify net and regurgitant flow volume through each of the cardiac valves.Materials and methodsAccelerated, 4D-PC MRI examinations performed between March 2010 through June 2011 as part of routine MRI examinations for congenital, structural heart disease were retrospectively reviewed and analyzed using valve-tracking visualization and quantification algorithms developed in Java and OpenGL. Excluding patients with transposition or single ventricle physiology, a total of 34 consecutive pediatric patients (19 male, 15 female; mean age 6.9 years; age range 10 months to 15 years) were identified. 4D-PC flow measurements were compared at each valve and against routine measurements from conventional cardiac MRI using Bland-Altman and Pearson correlation analysis.ResultsInlet and outlet valve net flow were highly correlated between all valves (P = 0.940-0.985). The sum of forward flow at the outlet valve and regurgitant flow at the inlet valve were consistent with volumetric displacements in each ventricle (P = 0.939-0.948). These were also highly consistent with conventional planar MRI measurements with net flow (P = 0.923-0.935) and regurgitant fractions (P = 0.917-0.972) at the outlet valve and ventricular volumes (P = 0.925-0.965).ConclusionIt is possible to obtain consistent measurements of net and regurgitant blood flow across the inlet and outlet valves relying solely on accelerated 4D-PC. This may facilitate more efficient clinical quantification of valvular regurgitation.

Published
2015

37. Robust 4D flow denoising using divergence‐free wavelet transform

Author

Ong, Frank, Uecker, Martin, Tariq, Umar, Hsiao, Albert, Alley, Marcus T, Vasanawala, Shreyas S, and Lustig, Michael

Subjects
Communications Engineering, Engineering, Algorithms, Artifacts, Blood Flow Velocity, Child, Coronary Circulation, Female, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Magnetic Resonance Angiography, Male, Reproducibility of Results, Sensitivity and Specificity, Signal-To-Noise Ratio, Wavelet Analysis, four-dimensional flow, wavelet denoising, divergence-free, Biomedical Engineering, Nuclear Medicine & Medical Imaging, Biomedical engineering
Abstract

PurposeTo investigate four-dimensional flow denoising using the divergence-free wavelet (DFW) transform and compare its performance with existing techniques.Theory and methodsDFW is a vector-wavelet that provides a sparse representation of flow in a generally divergence-free field and can be used to enforce "soft" divergence-free conditions when discretization and partial voluming result in numerical nondivergence-free components. Efficient denoising is achieved by appropriate shrinkage of divergence-free wavelet and nondivergence-free coefficients. SureShrink and cycle spinning are investigated to further improve denoising performance.ResultsDFW denoising was compared with existing methods on simulated and phantom data and was shown to yield better noise reduction overall while being robust to segmentation errors. The processing was applied to in vivo data and was demonstrated to improve visualization while preserving quantifications of flow data.ConclusionDFW denoising of four-dimensional flow data was shown to reduce noise levels in flow data both quantitatively and visually.

Published
2015

40. Evaluation of Valvular Insufficiency and Shunts with Parallel-imaging Compressed-sensing 4D Phase-contrast MR Imaging with Stereoscopic 3D Velocity-fusion Volume-rendered Visualization

Author

Hsiao, Albert, Lustig, Michael, Alley, Marcus T, Murphy, Mark J, and Vasanawala, Shreyas S

Subjects
Cardiovascular, Biomedical Imaging, Clinical Research, Heart Disease, Detection, screening and diagnosis, 4.2 Evaluation of markers and technologies, Adolescent, Algorithms, Child, Child, Preschool, Contrast Media, Data Compression, Echocardiography, Doppler, Color, Female, Heart Defects, Congenital, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Imaging, Three-Dimensional, Infant, Magnetic Resonance Imaging, Male, Retrospective Studies, Sensitivity and Specificity, Young Adult, Medical and Health Sciences, Nuclear Medicine & Medical Imaging
Abstract

PurposeTo assess the potential of compressed-sensing parallel-imaging four-dimensional (4D) phase-contrast magnetic resonance (MR) imaging and specialized imaging software in the evaluation of valvular insufficiency and intracardiac shunts in patients with congenital heart disease.Materials and methodsInstitutional review board approval was obtained for this HIPAA-compliant study. Thirty-four consecutive retrospectively identified patients in whom a compressed-sensing parallel-imaging 4D phase-contrast sequence was performed as part of routine clinical cardiac MR imaging between March 2010 and August 2011 and who had undergone echocardiography were included. Multiplanar, volume-rendered, and stereoscopic three-dimensional velocity-fusion visualization algorithms were developed and implemented in Java and OpenGL. Two radiologists independently reviewed 4D phase-contrast studies for each of 34 patients (mean age, 6 years; age range, 10 months to 21 years) and tabulated visible shunts and valvular regurgitation. These results were compared with color Doppler echocardiographic and cardiac MR imaging reports, which were generated without 4D phase-contrast visualization. Cohen κ statistics were computed to assess interobserver agreement and agreement with echocardiographic results.ResultsThe 4D phase-contrast acquisitions were performed, on average, in less than 10 minutes. Among 123 valves seen in 34 4D phase-contrast studies, 29 regurgitant valves were identified, with good agreement between observers (k=0.85). There was also good agreement with the presence of at least mild regurgitation at echocardiography (observer 1, κ=0.76; observer 2, κ=0.77) with high sensitivity (observer 1, 75%; observer 2, 82%) and specificity (observer 1, 97%; observer 2, 95%) relative to the reference standard. Eight intracardiac shunts were identified, four of which were not visible with conventional cardiac MR imaging but were detected with echocardiography. No intracardiac shunts were found with echocardiography alone.ConclusionWith velocity-fusion visualization, the compressed-sensing parallel-imaging 4D phase-contrast sequence can augment conventional cardiac MR imaging by improving sensitivity for and depiction of hemodynamically significant shunts and valvular regurgitation.

Published
2012

43. Coil sketching for computationally efficient MR iterative reconstruction.

Author

Oscanoa, Julio A., Ong, Frank, Iyer, Siddharth S., Li, Zhitao, Sandino, Christopher M., Ozturkler, Batu, Ennis, Daniel B., Pilanci, Mert, and Vasanawala, Shreyas S.

Subjects
THREE-dimensional imaging, OPTIMIZATION algorithms, PRINCIPAL components analysis, IMAGE reconstruction, MAGNETIC resonance imaging
Abstract

Purpose: Parallel imaging and compressed sensing reconstructions of large MRI datasets often have a prohibitive computational cost that bottlenecks clinical deployment, especially for three‐dimensional (3D) non‐Cartesian acquisitions. One common approach is to reduce the number of coil channels actively used during reconstruction as in coil compression. While effective for Cartesian imaging, coil compression inherently loses signal energy, producing shading artifacts that compromise image quality for 3D non‐Cartesian imaging. We propose coil sketching, a general and versatile method for computationally‐efficient iterative MR image reconstruction. Theory and Methods: We based our method on randomized sketching algorithms, a type of large‐scale optimization algorithms well established in the fields of machine learning and big data analysis. We adapt the sketching theory to the MRI reconstruction problem via a structured sketching matrix that, similar to coil compression, considers high‐energy virtual coils obtained from principal component analysis. But, unlike coil compression, it also considers random linear combinations of the remaining low‐energy coils, effectively leveraging information from all coils. Results: First, we performed ablation experiments to validate the sketching matrix design on both Cartesian and non‐Cartesian datasets. The resulting design yielded both improved computatioanal efficiency and preserved signal‐to‐noise ratio (SNR) as measured by the inverse g‐factor. Then, we verified the efficacy of our approach on high‐dimensional non‐Cartesian 3D cones datasets, where coil sketching yielded up to three‐fold faster reconstructions with equivalent image quality. Conclusion: Coil sketching is a general and versatile reconstruction framework for computationally fast and memory‐efficient reconstruction. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

46. AutoSamp: Autoencoding MRI Sampling via Variational Information Maximization

Author

Alkan, Cagan, Mardani, Morteza, Vasanawala, Shreyas S., and Pauly, John M.

Subjects
Image and Video Processing (eess.IV), FOS: Electrical engineering, electronic engineering, information engineering, Electrical Engineering and Systems Science - Image and Video Processing
Abstract

Accelerated MRI protocols routinely involve a predefined sampling pattern that undersamples the k-space. Finding an optimal pattern can enhance the reconstruction quality, however this optimization is a challenging task. To address this challenge, we introduce a novel deep learning framework, AutoSamp, based on variational information maximization that enables joint optimization of sampling pattern and reconstruction of MRI scans. We represent the encoder as a non-uniform Fast Fourier Transform that allows continuous optimization of k-space sample locations on a non-Cartesian plane, and the decoder as a deep reconstruction network. Experiments on public MRI datasets show improved reconstruction quality of the proposed AutoSamp method over the prevailing variable density and variable density Poisson disc sampling. We demonstrate that our data-driven sampling optimization method achieves 4.4dB, 2.0dB, 0.75dB, 0.7dB PSNR improvements over reconstruction with Poisson Disc masks for acceleration factors of R = 5, 10, 15, 25, respectively. Furthermore, we analyze the characteristics of the learned sampling patterns with respect to changes in acceleration factor, measurement noise, underlying anatomy, and coil sensitivities. We show that all these factors contribute to the optimization result by affecting the sampling density, k-space coverage and point spread functions of the learned sampling patterns.

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results