Olivier Jaubert¹, Valery Ozenne^2,3,4,5, Maxime Yon^2,3,4, Javier Montalt-Tordera¹, Jennifer Steeden¹, Vivek Muthurangu¹, and Bruno Quesson^2,3,4
1Institute of Cardiovascular Science, University College London, London, United Kingdom, ²Electrophysiology and Heart Modeling Institute, Foundation Bordeaux Université, IHU Lyric, Bordeaux, France, ³Centre de recherche Cardio-Thoracique de Bordeaux, Universite Bordeaux, Bordeaux, France, ⁴Centre de recherche Cardio-Thoracique de Bordeaux, INSERM, Bordeaux, France, ⁵Centre de Résonance Magnétique des Systèmes Biologiques, Bordeaux, France

A novel thermometry acquisition and a fast deep learning based image reconstruction were combined for cardiac interventional thermometry at high spatial (0.86×0.86mm²) and temporal (0.97s) resolutions, robust to motion and susceptibility artefact and independent of external ECG-gating. The method was tested in phantom and in-vivo in a sheep. The proposed deep learning method outperformed the state-of-the-art algorithm in terms of SNR and paves the way for clinical studies.

Yasmina Al Khalil¹, Sina Amirrajab¹, Cristian Lorenz², Jürgen Weese², Josien Pluim¹, and Marcel Breeuwer^1,3
1Biomedical Engineering, Eindhoven University of Technology, Eindhoven, Netherlands, ²Philips Research Laboratories, Hamburg, Germany, ³Philips Healthcare, MR R&D - Clinical Science, Best, Netherlands

While recent deep-learning-based approaches in automatic cardiac magnetic resonance image segmentation have shown great promise to alleviate the need for manual segmentation, most are not applicable to realistic clinical scenarios. This is largely due to training on mainly homogeneous datasets, without variation in acquisition parameters and pathology. In this work, we develop a model applicable in multi-center, multi-disease, and multi-view settings, where we combine heart region detection, augmentation through synthesis and multi-fusion segmentation to address various aspects of segmenting heterogeneous cardiac data. Our experiments demonstrate competitive results in both short-axis and long-axis MR images, without physically acquiring more training data.

Dominique Franson¹, James Ahad¹, Yuchi Liu², Alexander Fyrdahl², and Nicole Seiberlich^2
1Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States, ²University of Michigan, Ann Arbor, MI, United States

Through-time non-Cartesian GRAPPA enables real-time, free-breathing evaluation of the left ventricle, but it requires extra calibration data. This work proposes a self-calibrated spiral GRAPPA method that uses an interleaved acquisition to eliminate the calibration scan. A long spiral trajectory is used to improve robustness to motion. Data were acquired at 12 LV slices with spatiotemporal resolutions of 2.08x2.08 mm² and 32.72 ms/phase in 71 seconds. In a preliminary study of 6 volunteers comparing the proposed method to a gold-standard, functional values are within the limits of agreement in Bland-Altman analyses and are not statistically significantly different in paired t-tests (p<0.05).

Jakob Meineke¹, Christophe Schülke¹, Kay Nehrke¹, and Jochen Keupp^1
1Philips Research, Hamburg, Germany

Whole-Heart, Free-Breathing 3D Cardiac Cine MRI with a cartesian, water-selective, balanced SSFP acquisition and spiral-like undersampling is combined with fast, non-rigid motion-compensated reconstruction. It demonstrates the feasibility of routine cardiac cine MRI with minimal planning effort and minimum patient discomfort by avoiding the need for breathholds. The combined acquisition and reconstruction time is below 5 minutes, including computation of deformation vector fields, and minimizes image production delay by starting reconstruction on partially acquired data.

Xiao Chen¹, Yuan Zheng², Eric Z Chen¹, Zhongqi Zhang², Yu Ding², Jian Xu², Terrence Chen¹, and Shanhui Sun^1
1United Imaging Intelligence, Cambridge, MA, United States, ²UIH America, Inc., Houston, TX, United States

Cine CMR is a routinely performed technique for heart anatomy and function analysis. In clinics, repeated breathholdings are performed to acquire multiple 2D slices to cover the whole heart, which further exacerbates the discomfort and challenges for young and severe-diseased patients. In this study, we halved the breathholding time for cine CMR using MB technique. Multiple sampling patterns were designed and studied for the best reconstruction quality. A NN leveraging the Siamese structure was designed to reconstruct the accelerated MB data and achieved great image quality. 

Alena Kollmann¹, David Lohr¹, Markus Ankenbrand², Maya Bille¹, Maxim Terekhov¹, Michael Hock¹, Ibrahim Elabyad¹, Theresa Reiter^1,3, Florian Schnitter³, Wolfgang Bauer^1,3, Ulrich Hofmann³, and Laura Schreiber^1
1Chair of Cellular and Molecular Imaging, Comprehensive Heart Failure Center (CHFC), University Hospital Wuerzburg, Wuerzburg, Germany, ²Center for Computational and Theoretical Biology (CCTB), University of Wuerzburg, Wuerzburg, Germany, ³Department of Internal Medicine I, University Hospital Wuerzburg, Wuerzburg, Germany

Cardiac magnetic resonance (CMR) is considered the gold standard for evaluating cardiac function. Tools for automatic segmentation already exist in the clinical context. To bring the benefits of automatic segmentation to preclinical research, we use a deep learning based model.  We demonstrate that for training small data sets with parameters deviating from the clinical situation (high resolution images of porcine hearts acquired at 7T in this case) are sufficient to achieve good correlation with manual segmentation. We obtain DICE-scores of 0.87 (LV) and 0.85 (myocardium) and find high agreement of the calculated functional parameters (Pearson‘s r between 0.95 and 0.99).

Thomas James Fletcher¹, Carlos Velasco¹, Gastão Cruz¹, Alina Schneider¹, René Michael Botnar¹, and Claudia Prieto^1
1School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom

Dictionary generation and pattern matching are two important bottlenecks in cardiac MRF. Dictionaries must be recalculated for each new scan as they depend on the subject’s heart rate variability and both dictionaries and the length of pattern matching grow exponentially with the number of parameters being considered. We propose a conditional invertible neural network capable of both dictionary generation and parameter estimation for T₁, T₂ and T_1ρ cardiac MRF. The network achieves excellent results on EPG-generated data (inner product >0.999, parameters’ relative error <1.5%) and good results for in-vivo data (mean relative errors for myocardium ranging from 2.2% to 15.1%).

Jesse Ian Hamilton^1,2 and Nicole Seiberlich^1,2
1Radiology, University of Michigan, Ann Arbor, MI, United States, ²Biomedical Engineering, University of Michigan, Ann Arbor, MI, United States

This work introduces a physics-based deep learning reconstruction that can enable real-time (free-breathing and ungated) spiral 2D bSSFP functional cardiac MRI on a 0.55T scanner. The proposed method extends the concept of a deep image prior and does not require prior training on reference data. In addition, GROG is used to reduce the computational complexity of the forward model calculation by shifting spiral k-space data to their nearest Cartesian grid points, allowing use of FFT rather than NUFFT operations. Results are demonstrated in healthy subjects with functional measurements compared to gold standard acquisitions at 1.5T.

Sina Amirrajab¹, Yasmina Al Khalil¹, Cristian Lorenz², Jürgen Weese², Josien Pluim¹, and Marcel Breeuwer^1,3
1Biomedical Engineering Department, Eindhoven University of Technology, Eindhoven, Netherlands, ²Philips Research Laboratories, Hamburg, Germany, ³MR R&D - Clinical Science, Philips Healthcare, Best, Netherlands

 There has been considerable interest in the MR physics-based simulation of a database of virtual cardiac MR images for the development of deep-learning analysis networks. However, the employment of such a database is limited or shows suboptimal performance due to the realism gap, missing textures, and the simplified appearance of simulated images. In this work we 1) provide image simulation on virtual XCAT subjects with varying anatomies, and 2) propose sim2real translation network to improve image realism. Our usability experiments suggest that sim2real data exhibits a good potential to augment training data and boost the performance of a segmentation algorithm.

 

Zihao Chen^1,2, Hsu-Lei Lee¹, Xianglun Mao¹, Tianle Cao^1,2, Yibin Xie¹, Debiao Li^1,2, and Anthony Christodoulou^1,2
1Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States, ²Department of Bioengineering, University of California, Los Angeles, Los Angeles, CA, United States

CMR multitasking is promising for quantitative cardiac imaging, and can achieve fast three-slice quantification when combined with simultaneous multi-slice (SMS) acquisition. However, slow non-Cartesian iterative reconstruction is a barrier to clinical adoption. Deep learning can accelerate reconstruction, but the SMS training data are currently limited. Here we propose a data-consistent deep subspace transfer learning strategy that trains on single-slice T1 CMR multitasking data but is applied to SMS-encoded T1 CMR multitasking image reconstruction. The proposed strategy is >40x faster than the conventional SMS reconstruction, resulting in an equally better image quality and comparably precise T1 as in single-slice reconstruction.

Cheng William Hong¹, Naeim Bahrami², Vipul R Sheth¹, and Shreyas Vasanawala^1
1Department of Radiology, Stanford University, Palo Alto, CA, United States, ²GE Healthcare, Menlo Park, CA, United States

Cardiac MRI plays an important role in defining cardiac anatomy and function in children with heart disease. Prescription of diagnostic planes in cardiac MRI requires specialized training, limiting availability. Machine-learning based prescription has the potential to allow rapid, consistent acquisition of these planes independent of operator, but performance in congenital heart disease is unknown. This study evaluates the utility of such a system which demonstrates performance in children with mild to moderate heart disease similar to healthy adult volunteers. As expected, the system does not generate acceptable results in severe intracardiac anomalies.

Shishuai Wang¹, Pedro F Ferreira², Zohya Khalique², Margarita Gorodezky², Malte Roehl², Jialin Pu³, Dudley J Pennell², Sonia Nielles-Vallespin², and Andrew D Scott^2
1Department of Physics, Imperial College London, London, United Kingdom, ²Cardiovascular Magnetic Resonance Unit, Royal Brompton Hospital and National Heart and Lung Institute, Imperial College London, London, United Kingdom, ³College of Information And Communication Engineering, Harbin Engineering University, Harbin, China

Spiral trajectories are time efficient but are susceptible to off-resonance artefacts. Many approaches to off-resonance correction require a high-quality B0 field map. Here we train a convolutional neural network to remove the off-resonance artefacts using simulated cardiac spiral MRI and validate our methods by applying this network to diffusion tensor cardiovascular MR (DTCMR) data acquired with spiral trajectories.

Darian Viezzer^1,2, Thomas Hadler^1,2, Edyta Blaszczyk^1,2, Maximilian Fenski^1,3, Jan Gröschel^1,2, Steffen Lange⁴, and Jeanette Schulz-Menger^1,2,3
1Charité Universitätsmedizin Berlin, Working Group on Cardiovascular Magnetic Resonance, Experimental and Clinical Research Center, a joint cooperation between the Charité Universitätsmedizin Berlin and the Max-Delbrück-Center for Molecular Medicine, Berlin, Germany, ²DZHK (German Centre for Cardiovascular Research), partner site Berlin, Berlin, Germany, ³Department of Cardiology and Nephrology, Helios Hospital Berlin-Buch, Berlin, Germany, ⁴Faculty for Computer Sciences, Hochschule Darmstadt (University of Applied Sciences), Darmstadt, Germany

Parametric mapping images contain to a large extent irrelevant background information. In order to hide some of it, we incorporated bounding box information into a U-net based segmentation network. Our dataset consisted of 845 training, 102 validation and 146 test T1 maps of native and post-contrast myocardium from different clinical studies, including healthy volunteers and patients with inflammatory heart disease, muscular dystrophies or chronic myocardial infarction. While cropping the image input improved the segmentation itself, a second input of the bounding box mask reduced the mean absolute and mean squared T1 deviation, which is clinically preferred.

Xitong Wang¹, Junyu Wang¹, Ruixi Zhou^1,2, and Michael Salerno^1,3,4,5
1Department of Biomedical Engineering, University of Virginia, Charottesville, VA, United States, ²School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, China, ³Department of Medicine, Stanford University, Palo Alto, CA, United States, ⁴Department of Medicine, University of Virginia, Charottesville, VA, United States, ⁵Department of Radiology, University of Virginia, Charottesville, VA, United States

Cardiac cine images are the gold standard techniques for analyzing cardiac function. Clinical breath-held ECG-gated Cartesian cine imaging requires 10-12 breath-holds to cover the left ventricle.  Our group has developed a continuous-acquisition respiratory and cardiac self-gated cine sequence (SPARCS) for free-breathing cardiac function acquisition using 2D spiral acquisition can suffer from through-plane motion. Thus, we develop a rapid 3D self-gated cine technique using a variable density undersampled stack of spiral gradient echo sequence to provide 3D coverage of the ventricle enabling thinner slice acquisition and improved robustness to through-plane motion.

Antonella Meloni¹, Laura Pistoia¹, Pietro Giuliano², Nicola Giunta², Nicolò Schicchi³, Emanuele Grassedonio⁴, Stefania Renne⁵, Vincenzo Positano¹, Lorella Pitrolo⁶, Maria Grazia Roberti⁷, Paola Maria Grazia Sanna⁸, Filippo Cademartiri¹, and Alessia Pepe^1
1Fondazione G. Monasterio CNR-Regione Toscana, Pisa, Italy, ²"ARNAS" Civico, Di Cristina Benfratelli, Palermo, Italy, ³Azienda Ospedaliero-Universitaria Ospedali Riuniti "Umberto I-Lancisi-Salesi", Ancona, Italy, ⁴Policlinico "Paolo Giaccone", Palermo, Italy, ⁵Presidio Ospedaliero “Giovanni Paolo II”, Lamezia Terme (CZ), Italy, ⁶Ospedale "V. Cervello", Palermo, Italy, ⁷Azienda Ospedaliero-Universitaria OO.RR. Foggia, Foggia, Italy, ⁸Azienda Ospedaliero-Universitaria di Sassari, Sassari, Italy

Seven-hundred and nine patients with thalassemia major who performed a baseline and a 1st follow-up CMR scan after 18 months were followed prospectively in order to evaluate the predictive value of changes in CMR parameters (myocardial iron, biventricular function, and replacement myocardial fibrosis) for cardiac complications. During a mean follow-up of 89.4±33.3 months, cardiac events (heart failure, arrhythmias, and pulmonary hypertension) were recorded in 7.1% of patients. In the univariate Cox regression analysis, cardiac iron clearance and replacement myocardial fibrosis were identified as univariate prognosticators but in the multivariate analysis only myocardial fibrosis remained an independent predictor factor.