Nicola Martini, Daniele Della Latta, Gianmarco Santini, Gabriele Valvano, Andrea Barison, Francesco Avogliero, Daniele De Marchi, Luigi Landini, Dante Chiappino

We proposed a fully automated approach for the segmental analysis of T1 mapping using a fully convolutional neural network architecture. T1 maps acquired using the MOLLI sequence from 394 subjects were considered. Excellent segmentation results are demonstrated by high Jaccard (0.969±0.023) and a Dice (0.984±0.012) indexes. No significant difference in the obtained segmental T1 values compared to manual measurements was found, with a mismatch percentage ranging from 0.95% to 3.14% across segments.

Fei Han, Ziwu Zhou, Vahid Kouzehkonan, Yu Gao, Yingli Yang, Peng Hu

Cardiac CINE MRI is widely used for evaluating ventricular wall motion and cardiac function. Conventional cardiac CINE consists of ECG-triggered k-space segmented 2D acquisitions, each performed within a breath-hold. In this study, we propose an ECG-free, cardiac CINE protocol that covers the entire LV within a single breath-hold. Our solution is based on a highly accelerated real-time imaging that is enabled by our recently proposed parallel imaging and deep learning combined (PI-DL) image reconstruction. In this study, we evaluated the proposed solution in healthy volunteers and compare its performance with cardiac CINE images acquired using conventional protocol. 

Zechen Zhou, Shuo Chen, Jiayi Wu, Xihai Zhao, Peter Börnert, Chun Yuan

Turbo spin echo (TSE) imaging with variable flip angle (VFA) is commonly used for three-dimensional (3D) high resolution intracranial vessel wall imaging. However, different tissues may experience various blurring effects particularly for longer TSE factor. In this study, a deep convolutional neural network is trained to provide a solution for this special deblurring problem. Combined with a signal-to-noise ratio (SNR)-priority VFA design scheme, the developed technique can provide a better tradeoff across scan efficiency, point spread function and SNR for 3D TSE acquisitions. Preliminary results have demonstrated its improvement for sharper delineation of intracranial vessel wall and plaque boundaries at isotropic 0.5mm resolution.

Yibin Xie, Ruiyuan Lin, Yuhua Chen, Yubo Zhang, Feng Shi, Yanan Fei, Zixin Deng, Derenik Haghverdian, Madhvi Kannan, Hyuk-Jae Chang, C.-C. Jay Kuo, Debiao Li

Coronary MRA is an attractive imaging tool to offer noninvasive, radiation-free evaluation of coronary artery disease. However, long scan time and sensitivity to motion limit its current clinical applications. In this paper, we propose a super resolution reconstruction framework based on 3D generative adversarial network (GAN) to allow substantial acceleration (10x plus) and potentially whole-heart coronary MRA within a breath-hold. Preliminary results demonstrated significantly improved vessel sharpness and image quality metrics in highly under-sampled coronary MRA dataset.  

Jifan Li, Shuo Chen, Xihai Zhao, Chun Yuan, Rui Li

In this work, we proposed an automatic approach for segmentation of carotid vessel wall in multi-contrast blackblood images, using a fine-tuning U-net neural network. The U-net network consists of an encoder path that captures context and reduces data dimension and a symmetric decoder path that enables precise localization and high resolution. The fine-tuning was utilized to accommodate multi-contrast images input. The pixel-level sensitivity, specificity and IoU of our model achieved 0.869, 0.987 and 0.751 on the test data set, respectively. 

Joshua van Amerom, David Lloyd, Maria Kuklisova Murgasova, Anthony Price, Shaihan Malik, Milou Van Poppel, Kuberan Pushparajah, Mary Rutherford, Reza Razavi, Joseph Hajnal

Cine fetal cardiac imaging with whole heart coverage presents numerous challenges due to maternal and fetal motion as well as high heart rate and lack of reliable independent synchronisation signal. To overcome these challenges, highly accelerated multi-slice 2D dynamic images were retrospectively reordered in time and aligned in space to produce a 3D cine using scattered interpolation methods with outlier rejection. The proposed methods were tested on ten human fetal subjects and validated using a numerical phantom.

Chenxi Hu, Dana Peters

The accuracy of cardiac T2 mapping in small structures, such as the subendocardium, right ventricle, and left atrium, is impaired by the partial-voluming effect between myocardium and blood. In this work, we propose a dark-blood T2 mapping method that nulls the blood prior to the T2 quantification, so that partial-voluming is reduced and T2 estimation accuracy in these small structures is improved. Both conventional and the proposed dark-blood T2 mapping methods were performed in healthy subjects. The results demonstrated the potential of dark-blood T2 mapping to improve the clinical assessment of the subendocardium, right ventricle, and left atrium. 

James Pilla, Gabor Mizsei, Jerry Zsido II, Norman Butler, Yoshiaki Saito, Gabrielle Pilla, Akito Imai, Keitaro Okamoto,, Christopher Gade, Mehrdad Pourfathi, Kai Ruppert, Stephen Kadlecek, Sarmad Siddiqui, Rahim Rizi, Joseph Gorman III, Robert Gorman, Terence Gade

Adverse remodeling after a myocardial infarct has been linked to the elevated wall stress in the myocardium adjacent to the infarct (i.e. border zone). Perturbed metabolism in this region could drive the transition from compensated to adverse remodeling. To evaluate regional metabolism this study compared the uptake and intracellular conversion of [1-¹³C]pyruvate using hyperpolarized ¹³C.  An infarct model of ventricular remodeling was used to investigate region metabolism.  Pharmacologic stress produced an increase in remote metabolite flux compared to border zone region which may provide a metabolic mechanism for the established association of mechanical stress and adverse cardiac remodeling following infarct.

Jessica Webb, Jurgen Runge, Omar Darwish, Alessandro Polcaro, Torben Schneider, Gerald Carr-White, Jordi Martorell, David Nordsletten, Reza Razavi, Ralph Sinkus

Heart Failure with preserved Ejection Fraction is common, associated with high morbidity and mortality, and is challenging to diagnose. We have developed a novel patient friendly non-invasive technique to quantify myocardial stiffness using transient MR Elastography. Aortic valve closure results in the propagation of a transient shear wave through the myocardium. Torsional wave propagation can be visualised using a 1D pencil beam navigator positioned longitudinally along the myocardial septum, using four breath holds each 15 seconds. Providing a temporal resolution of 0.3ms, we observe increased myocardial stiffness in HFpEF patients compared to healthy volunteers (torsional speed 5.5 ± 1.1 m/s in volunteers and 10.0 ± 0.7 m/s in Patients; stiffness 36 ± 12.0 kPa and 108 ± 15.2 kPa, respectively) .

stanislas Rapacchi, Thomas Troalen, Maxime Guye, Monique Bernard, Alexis Jacquier, Frank Kober

To perform high-resolution cardiac function MRI, thin-slice Cine MRI at 7T is accelerated using Simultaneous Multi-Slice (SMS) technique. Additionally, Hadamard encoding strategy along the temporal dimension is superimposed on the CAIPIRINHA phase shift. The Hadamard-decoded data serve as embedded reference for SMS image reconstruction. Additional in-plane L1-SPIRiT reconstruction allows for limited noise amplification. Results show excellent slice separation, satisfactory SNR and CNR for assessment of cardiac function. However, SAR restrictions impose a lower flip angle for SMS acquisitions that result in poorer blood-to-myocardium contrast.