View Presentation Video

Renske Merton^1,2, Daan Bosshardt^1,2, Gustav J. Strijkers³, Aart J. Nederveen¹, Eric M. Schrauben¹, and Pim van Ooij^1,2
1Radiology and Nuclear Medicine, Amsterdam UMC location University of Amsterdam, Amsterdam, Netherlands, ²Atherosclerosis and Ischemic Syndromes, Amsterdam Cardiovascular Sciences, Amsterdam, Netherlands, ³Biomedical Engineering and Physics, Amsterdam UMC location University of Amsterdam, Amsterdam, Netherlands

Keywords: Vessel Wall, Segmentation, 4D motion, aorta

In this study we introduce 4D aortic displacement and diameter change estimations from isotropic high-resolution 3D CINE CMR as novel biomarkers for aortic stiffness. These could potentially be used as an alternative to pulse wave velocity stiffness measurements which can be unreliable for very stiff aortas as found in for example Marfan syndrome patients. Using deep learning for segmentation, visualization and quantification of aortic motion is demonstrated through 4D aortic motion maps. Time curves of displacement showed the typical cyclic behavior of cardiac motion, with a maximum displacement median of 10.3 (3.6) mm over 14 healthy volunteers.

View Presentation Video

Masami Yoneyama¹, Yasuhiro Goto², Michinobu Nagao³, Shuo Zhang⁴, Kayoko Abe³, Yutaka Hamatani², Kazuo Kodaira², Takumi Ogawa², Mana Kato², Isao Shiina², Christian Stehning⁴, and Marc Van Cauteren^5
1Philips Japan, Tokyo, Japan, ²Department of Radiological Services, Tokyo Women's Medical University, Tokyo, Japan, ³Department of Diagnostic Imaging and Nuclear Medicine, Tokyo Women’s Medical University, Tokyo, Japan, ⁴Philips Healthcare, Hamburg, Germany, ⁵Philips Healthcare, Best, Netherlands

Keywords: Atherosclerosis, Atherosclerosis

Clinical imaging procedure for vascular evaluation and tissue characterization often needs separate MR scans. Both scenarios require multiple imaging sequences with different image contrasts, which results in long exam times and image misalignment due to possible motion between the scans that needs to be corrected in post-processing. In this work, the recently proposed REACT (Relaxation-Enhanced Angiography without ContrasT) technique was further developed and combined with multi-echo data acquisitions and Dixon-based chemical-shift-based water-fat separation. Initial results in patients showed great promise in detection and assessment of vascular and vessel wall lesions in one single scan. 
 

View Presentation Video

Keywords: Vessels, Cardiovascular, deep learning

The balanced steady-state free precession (bSSFP) sequence is widely used for navigated whole-heart coronary magnetic resonance angiography (MRA) for the evaluation of coronary anatomy and abnormalities due to its inherently high blood signal intensity and blood-myocardial contrast. However, the main drawback of this approach is that the scan time is longer and prone to interference with motion artifacts. In this study, we investigated the utility of whole-heart coronary MRA using accelerated bSSFP with compressed sensing artificial intelligence (CSAI) technique at 3 Tesla. The results demonstrated the adopted CS-AI technique yielded high image quality within a clinically feasible acquisition time in healthy subjects and patients with suspected CAD.

View Presentation Video

Carlos Castillo-Passi^1,2,3, Michael G. Crabb¹, Camila Muñoz¹, Karl P. Kunze^1,4, Radhouene Neji ^1,4, Pablo Irarrazaval^2,3,5, Claudia Prieto^1,3,5, and Rene M. Botnar^1,2,3,5
1School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom, ²Institute for Biological and Medical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile, ³Millennium Institute for Intelligent Healthcare Engineering, Santiago, Chile, ⁴MR Research Collaborations, Siemens Healthcare Limited, Camberley, United Kingdom, ⁵Electrical Engineering Department, Pontificia Universidad Católica de Chile, Santiago, Chile

Keywords: Heart, Low-Field MRI

In this work, we demonstrate the feasibility of free-breathing whole-heart magnetic resonance angiography (CMRA) at 0.55T. We implemented an image navigator (iNAV)-based non-rigid motion-corrected reconstruction with 100% respiratory scan efficiency and patch-based low-rank (PROST) denoising. The research sequence was validated on three healthy subjects, showing that whole-heart CMRA provides good depiction of the coronary arteries at 0.55T in under 6 minutes scan time.

View Presentation Video

Joshua F P van Amerom¹, Ye Tian², Datta S Goolaub¹, John Wood^3,4, Jon Detterich⁴, Krishna S Nayak², and Christopher K Macgowan^1,5
1Translational Medicine, Hospital for Sick Children, Toronto, ON, Canada, ²University of Southern California, Los Angeles, CA, United States, ³Department of Biomedical Engineering, University of Southern California, Los Angeles, CA, United States, ⁴Division of Cardiology, Children's Hospital Los Angeles, Los Angeles, CA, United States, ⁵Department ofMedical Biophysics, University of Toronto, Toronto, ON, Canada

Keywords: Heart, Fetus, Cine, 4D, Whole Heart, Morphology

Spiral SSFP of the fetal heart at low-field has the advantage of low artifact and high sampling efficiency, compared to conventional systems. This work demonstrates the feasibility of spiral SSFP on a high-performance 0.55 T scanner to produce high spatiotemporal resolution real-time 2D and motion-corrected slice-to-volume super-resolution reconstructed 3D cine images of the heart and great vessels of the human fetus in utero. Patient comfort is improved with the use of a low-field scanner, while motion-robust slice-to-volume reconstruction allows for free-breathing acquisitions that do not require precise scan plane prescriptions during acquisition.

View Presentation Video

Manuel A Morales¹, Siyeop Yoon¹, Ahmed Fahmy¹, Jennifer Rodriguez¹, Daniel B Herzka², Warren J Manning¹, and Reza Nezafat^1
1BIDMC, Boston, MA, United States, ²Case Western Reserve University, Cleveland, OH, United States

Keywords: Heart, Pulse Sequence Design

Combined cardiac MRI and exercise (Ex-CMR) is a stress imaging test with promising applications in cardiovascular disease (CVD). Ex-CMR myocardial tagging could provide insights into myocardial deformation post-exercise. We sought to develop an SSFP-based highly-accelerate (12-fold) free-breathing ECG-free real time tagging sequence for quantification of beat-to-beat variation of myocardial deformation at rest and after exercise. The proposed real time sequence achieved a spatiotemporal resolution of 2 x 2 mm² and 27.6 ms. Feasibility of real time imaging at 3T post-exercise was demonstrated in patients with heart failure with preserved (HFpEF) and reduced (HFrEF) ejection fraction. 

View Presentation Video

Omer Burak Demirel^1,2, Chi Zhang^1,2, Burhaneddin Yaman^1,2, Merve Gulle^1,2, Tim Leiner³, Peter Kellman⁴, and Mehmet Akçakaya^1,2
1Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, United States, ²Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States, ³Department of Radiology, Mayo Clinical, Rochester, MN, United States, ⁴National Heart-Lung and Blood Institute, Bethesda, MD, United States

Keywords: Heart, Heart

Real-time cine CMR is an ECG-free free-breathing alternative for functionally assessing the heart. To achieve sufficient spatio-temporal resolutions, these require rapid imaging, e.g. compressed sensing (CS) with radial trajectories. However, at high accelerations, CS may suffer from residual aliasing and temporal blurring. Recently, deep learning (DL) reconstruction has gained immense interest for fast MRI. Yet, for free-breathing real-time cine, where subjects have different breathing and cardiac motion patterns, database learning of spatiotemporal correlations has been difficult. Here, we propose a physics-guided DL reconstruction trained in a subject-specific manner. Proposed method improves image quality compared to database-trained DL and conventional methods. 

View Presentation Video

Olivier Jaubert¹, Javier Montalt-Tordera¹, Dan Knight², Simon Arridge³, Jennifer Steeden¹, and Vivek Muthurangu^1
1Institute of Cardiovascular Sciences, University College London, London, United Kingdom, ²Department of Cardiology, Royal Free London NHS Foundation Trust, London, United Kingdom, ³Department of Computer Science, University College London, London, United Kingdom

Keywords: Heart, Machine Learning/Artificial Intelligence, Interventional, Interactive

Interactive MR sequences have relatively low spatial and temporal resolution due to the limited acquisition and reconstruction time available. We propose to jointly optimize a variable density spiral acquisition and deep artifact suppression network (via a bandit-based approach) to maximize acquisition and reconstruction efficiency and provide interactive high spatio-temporal resolution images. The proposed approach was characterized in simulations and demonstrated prospectively in-vivo, offering promising performance with improved image quality and good handling of abrupt scan-plane changes.

View Presentation Video

Daniel P Seiter¹, Philip A Corrado¹, Farhan Raza², and Oliver Wieben^1,3
1Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, ²Department of Medicine, Cardiovascular Division, University of Wisconsin-Madison, Madison, WI, United States, ³Radiology, University of Wisconsin-Madison, Madison, WI, United States

Keywords: Heart, Cardiovascular, Exercise, real time

Right heart catheterization during exercise is the current gold standard for diagnosis of heart failure with preserved ejection fraction (HFpEF) but carries the risk of an invasive procedure. In this work, we present preliminary data using real time cardiac magnetic resonance imaging during exercise in a HFpEF cohort with comparison to healthy controls.

View Presentation Video

Qing Zou¹, Sanja Dzelebdzic¹, and Tarique Hussain^1
1University of Texas Southwestern Medical Center, Dallas, TX, United States

Keywords: Heart, Machine Learning/Artificial Intelligence, Reconstruction

We introduce a deep kernel model for the recovery of free-breathing and ungated cardiac MRI from highly undersampled measurements. The proposed scheme uses the cascade of two deep convolutional neural networks for the kernel representation of images. The parameters of the two CNNs in the proposed method are learned from the undersampled measurements directly in this work and hence the framework is unsupervised. The main benefits of the proposed scheme are (a) the elimination of the empirical choice of the feature map and kernel function in the kernel method, and (b) the unsupervised nature of the proposed framework.

View Presentation Video

Charles McGrath¹, Oliver Bieri^2,3, Sebastian Kozerke¹, and Grzegorz Bauman^2,3
1University and ETH Zurich, Zurich, Switzerland, ²Division of Radiological Physics, Department of Radiology, University Hospital Basel, Basel, Switzerland, ³Department of Biomedical Engineering, University of Basel, Basel, Switzerland

Keywords: Flow, Low-Field MRI

Free-running radial phase-contrast balanced steady-state free precession (PC-bSSFP) imaging including slice-select gradient inversion and respiratory and cardiac self-gating enables robust through-plane flow quantification in the aorta with excellent image signal-to-noise ratio (SNR) and contrast on a low-cost low-field 0.55T scanner. Relative to phase-contrast spoiled gradient echo (PC-GRE), PC-bSSFP offers SNR improvements by a factor of about two.

Keywords: Flow, Low-Field, Balanced SSFP, Phase-Contrast, Self-gated

View Presentation Video

Christopher W. Roy^1,2, Bastien Milani^1,2, Jérôme Yerly^1,2,3, Salim Si-Mohamed^1,4,5, Estelle Tenisch^1,2, Tobias Rutz^2,6, Milan Prsa^2,7, Ludovica Romanin^1,2,8, Aurélien Bustin^9,10, Davide Piccini^1,8, and Matthias Stuber^1,2,3
1Diagnostic and Interventional Radiology, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland, ²University of Lausanne, Lausanne, Switzerland, ³Center for Biomedical Imaging (CIBM), Lausanne, Switzerland, ⁴University Lyon, Lyon, France, ⁵Radiology, Louis Pradel Hospital, Lyon, France, ⁶Service of Cardiology, Heart and Vessel Department, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland, ⁷Division of Pediatric Cardiology, Woman-Mother-Child Department, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland, ⁸Advanced Clinical Imaging Technology, Siemens Healthcare International AG, Lausanne, Switzerland, ⁹IHU LIRYC, Electrophysiology and Heart Modeling Institute, Université de Bordeaux, Bordeaux, France, ¹⁰Department of Cardiovascular Imaging, Hôpital Cardiologique du Haut-Lévêque, Bordeaux, France

Keywords: Heart, Motion Correction

In this work, we develop, validate, and apply a novel reconstruction framework for intra-bin correction and inter-bin compensation of respiratory motion in cardiac and respiratory motion-resolved free-running whole heart 5D MRI. We demonstrate respiratory resolved images with reduced artifact and without compressing the underlying physiological motion providing a means to increase the acquired resolution, accelerate the acquisition, or investigate respiratory driven changes in cardiac output and blood flow, potentially leading to new MRI driven biomarkers for cardiovascular disease

View Presentation Video

Salim Aymeric Si-Mohamed^1,2, Ludovica Romanin³, Mariana Falcao³, Jerome Yerly², Estelle Tenisch², Tobias Rutz², Charles De Bourguignon⁴, Jurg Schwitter⁵, Matthias Stuber³, Christopher Roy², and Milan Prsa^6
1Radiology department, University of Lyon, Lyon, France, ²CIBM, CHUV, Lausanne, Switzerland, ³CIBM, CHUV, Lausanne, France, ⁴Research radiology department, Hospices Civiles de Lyon, Lyon, France, ⁵Cardiology, CHUV, Lausanne, Switzerland, ⁶Pediatric cardiology department, CHUV, Lausanne, France

Keywords: Cardiomyopathy, Data Analysis, cardiac function

Free-running 5D whole-heart imaging (5D CMR) has been proposed as a means of simplifying CMR exams by capturing the entire 3D cardiac anatomy without the need for ECG gating or breath-holds. We demonstrated that 5D CMR with ferumoxytol enhancement enables the evaluation of cardiac function in comparison to ECG gated 2D CINE images in congenital heart disease patients. In addition, it enables evaluation of the cardiac morphology with improved diagnostic quality in ~40% of the cases. This suggests that the free-running approach has the potential for replacing the conventional 2D imaging for evaluation of both cardiac function and morphology.

View Presentation Video

Marina Awad¹, Junyu Wang², and Michael Salerno^2
1Biomedical Engineering, University of Virginia, Charlottesville, VA, United States, ²Cardiovascular Medicine, Stanford University, Stanford, CA, United States

Keywords: Heart, Perfusion

Cardiovascular magnetic resonance (CMR) is susceptible to motion-induced artifacts from cardiac and respiratory motion, leading to poor image quality. The inter-frame motion artifacts make quantitative analysis for cardiac function evaluation difficult. Hence motion correction is an important pre-processing step before robust quantification of myocardial perfusion. We developed a deep learning-based framework for rapid and accurate motion correction of CMR perfusion imaging using a 2D U-Net that estimates the deformation field from a moving frame to a fixed frame.

View Presentation Video

Keywords: Image Reconstruction, Cardiovascular, T2*

T2* CMR is widely used for detecting hemorrhagic myocardial infarction (MI). However, the conventional T2* CMR (2D breath-held, ECG-gated, multi-gradient-echo T2*) can suffer from limited spatial resolution and motion artifacts. We have recently developed a time-efficient, fully ungated, free breathing, 3D T2* mapping approach using a low-rank tensor (LRT) framework to address the above issues.