Cine Acquisition Research Articles

Machine learning evaluation of LV outflow obstruction in hypertrophic cardiomyopathy using three-chamber cardiovascular magnetic resonance

Left ventricular outflow tract obstruction (LVOTO) is common in hypertrophic cardiomyopathy (HCM), but relationships between anatomical metrics and obstruction are poorly understood. We aimed to develop machine learning methods to evaluate LVOTO in HCM patients and quantify relationships between anatomical metrics and obstruction. This retrospective analysis of 1905 participants of the HCM Registry quantified 11 anatomical metrics derived from 14 landmarks automatically detected on the three-chamber long axis cine CMR images. Linear and logistic regression was used to quantify strengths of relationships with the presence of LVOTO (defined by resting Doppler pressure drop of > 30 mmHg), using the area under the receiver operating characteristic (AUC). Intraclass correlation coefficients between the network predictions and three independent observers showed similar agreement to that between observers. The distance from anterior mitral valve leaflet tip to basal septum (AML-BS) was most highly correlated with Doppler pressure drop (R2 = 0.19, p < 10–5). Multivariate stepwise regression found the best predictive model included AML-BS, AML length to aortic valve diameter ratio, AML length to LV width ratio, and midventricular septal thickness metrics (AUC 0.84). Excluding AML-BS, metrics grouped according to septal hypertrophy, LV geometry, and AML anatomy each had similar associations with LVOTO (AUC 0.71, 0.71, 0.68 respectively, p = ns), significantly less than their combination (AUC 0.77, p < 0.05 for each). Anatomical metrics derived from a standard three-chamber CMR cine acquisition can be used to highlight risk of LVOTO, and suggest further investigation if necessary. A combination of geometric factors is required to provide the best risk prediction.

Body mass index associated differences in cardiac stress energetics in type 2 diabetes

Abstract Background Patients with T2D and heart disease have normal body mass index (BMI), suggesting that diabetes and obesity mediate cardiovascular change by different mechanisms. Changes in cardiac energy metabolism in lean diabetic patients during exercise stress have not been previously reported. Objectives We aimed to assess if there are BMI-associated differences in cardiac stress metabolism in patients with T2D. Methods Twenty-five overweight T2D patients (O-T2D) and eleven lean T2D patients (LnT2D), age- and ethnicity-matched and with no other comorbidities were studied. Patients were on oral hypoglycaemics only and were free of diabetes complications. Participants underwent rest and dobutamine stress phosphorus magnetic resonance spectroscopy (31P-MRS) and cardiovascular magnetic resonance (CMR) at 3T for the assessment of myocardial phosphocreatine to ATP ratio (PCr/ATP) as a measure of myocardial energetics, biventricular volumes, rest and stress left ventricular (LV) ejection fraction, global longitudinal shortening, and mitral in-flow E/A ratio for assessment of diastolic function and perfusion. Intravenous Dobutamine was administered at a dose of 10μg/kg/min, increasing at 90 second intervals up to a maximum of 40 μg/kg/min to achieve a target heart rate of 65% of the age-predicted maximal heart rate. Mean rate pressure product (RPP) was recorded at rest and stress. Heart rate was maintained at target for the duration of the 31P-MRS and stress CMR cine, mitral in-flow and perfusion acquisitions. Results The cardiac volumes, systolic or diastolic function and LV mass were similar between LnT2D and O-T2D. Although the O-T2D patients had a numerically lower rest and stress PCr/ATP ratio, this did not reach statistical significance. Resting PCr/ATP was reduced in LnT2D and O-T2D patients similarly. However, LnT2D showed a greater reduction in PCr/ATP (stress PCr/ATP LnT2D 1.51±0.2 vs O-T2D 1.41±0.25, p=0.02) despite similar increases in RPP. Stress myocardial blood flow (MBF) was also significantly lower in the O-T2D patients. There were significant correlations of BMI with LV mass (r=0.35, p=0.03); stress LVEF (r=−0.34, p=0.04); stress MBF stress (r=−0.53, p=0.001) and stress E/A (r=0.46, p=0.01) (figure 1). Conclusions Despite their better stress perfusion and similar glycaemic control, LnT2D show worse metabolic reserve characterised by more significant decrements in energetics in response to hemodynamic stress compared to overweight patients with T2D. Higher BMI correlates inversely with stress myocardial blood flow and with stress left ventricular ejection fraction. The presence of these subtle alterations in measures of stress metabolism and perfusion might signify a distinct metabolic phenotype of “lean diabetic cardiomyopathy”. Future studies are needed to further delineate alterations in cardiac energy metabolism in lean and overweight/obese type 2 diabetes patients, and their role in the development of cardiac dysfunction. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): Wellcome TrustBHF

Radiation exposure during transcatheter aortic valve implantation (TAVI): comparison of balloon-expandable versus self-expandable prostheses

Abstract Transcatheter aortoc valve implantation (TAVI) is performed under fluoroscopic control and can be associated with the need for long fluoroscopy times and repeated cine acquisitions in angulated projections. The procedural steps for TAVI with balloon-expandable and self-expanding prostheses differ and may be associated with significant differences in radiation exposure. Published data regarding patient or operator radiation exposure in TAVI are limited to small series. We therefore analyzed the influence of prosthesis type on radiation exposure in a consecutive series of 1185 patients who underwent TAVI between 2016 and 2021. A cohort of 1185 consecutive patients undergoing TAVI between 2016 and 2021 was analyzed. Radiation exposure was determined by evaluating overall fluoroscopy time, the number of acquired cine sequences, and total dose-area product (DAP). After eliminating patients treated via a non-transfemoral approach, data between patients underging TAVI with self-expanding prostheses and patients undergoing TAVI with balloon-expandable prostheses were compared. Out of the total patient cohort, 46 patients were excluded due to treatment via a non-transfemoral approach. Of the remaining 1139 patients, 437 (38%) were treated with self-expandable prostheses and 702 (62%) were treated with balloon-expandable prostheses. Median age was 81 years, 45% of patients were female. Body weight was slightly but significantly higher in patients treated by balloon-expandable prostheses (median 73 kg vs. 79 kg, p&amp;lt;0.001). Median fluoroscopy time was 453 s (IQR 365–603 s) for self-expandable prostheses vs. 414 s (IQR 341–540 s) for balloon-expandable prostheses (p=0.002). Also, the number of cine acqusitions was significantly higher (median 12 vs. 7; p&amp;lt;0.001), and total DAP was 6442 mGy cm2 (IQR 4452–9669 mGy cm2) for self-expandable vs. 4798 (IQR 3353–6981 mGy cm2) for balloon-expandable prostheses (p&amp;lt;0.001, see Fig. 1). In multivariable analysis, male sex, higher body weight and use of a self-expandable prosthesis were independently associated with higher total DAP. Transcatheter aortic valve implantation using balloon-expandable prostheses is associated with significantly lower total dose-area product than treatment with self-expandable prostheses. This may be particularly relevant for operators who perform the procedure frequently. Funding Acknowledgement Type of funding sources: None.

Fetal 3D cardiovascular cine image acquisition using radialsampling and compressed sensing.

To explore a fetal 3D cardiovascular cine acquisition using a radial image acquisition and compressed-sensing reconstruction and compare image quality and scan time with conventional multislice 2D imaging. Volumetric fetal cardiac data were acquired in 26 volunteers using a radial 3D balanced SSFP pulse sequence. Cardiac gating was performed using a Doppler ultrasound device. Images were reconstructed using a parallel-imaging and compressed-sensing algorithm. Multiplanar reformatting to standard cardiac views was performed before image analysis. Clinical 2D images were used for comparison. Qualitative and quantitative image evaluation were performed by two experienced observers (scale: 1-4). Volumes, mass, and function were assessed. Average scan time for the 3D imaging was 6min, including one localizer. A 2D imaging stack covering the entire heart including localizer sequences took at least 6.5min, depending on planning complexity. The 3D acquisition was successful in 7 of 26 subjects (27%). Overall image contrast and perceived resolution were lower in the 3D images. Nonetheless, the 3D images had, on average, a moderate cardiac diagnostic quality (median [range]: 3 [1-4]). Standard clinical 2D acquisitions had a high cardiac diagnostic quality (median [range]: 4 [3, 4]). Cardiac measurements were not different between 2D and 3D images (all p> 0.16). The presented free-breathing whole-heart fetal 3D radial cine MRI acquisition and reconstruction method enables retrospective visualization of all cardiac views while keeping examination times short. This proof-of-concept work produced images with diagnostic quality, while at the same time reducing the planning complexity to a single localizer.

Spatiotemporal variational neural network for reconstruction of highly accelerated cardiac cine MRI

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): Siemens Healthcare GmbH; NIH Grant. Background Acquisition of cardiac cine MRI usually requires the subject to repeatedly hold their breath for several seconds. Deep learning-based reconstruction of highly undersampled data could allow higher accelerations and has been proposed for several applications. The variational network [1] (VN) was one of the first network architectures for MRI reconstruction, replicating iterative reconstruction with a regularization term learnt from training data. Purpose To expand the concept of VNs to spatiotemporal data from cardiac cine acquisitions. Reconstructions are presented and compared using 3D, 2D spatial-only, and 2D+t spatiotemporal convolutions. Methods A VN with intermediate quantities in k-space [2] was trained to reconstruct cine sequences from retrospectively undersampled data. The network architecture is illustrated in Fig. 1. It consisted of 10 cascades, each resembling a gradient descent step of an iterative reconstruction. Each refinement step contained a Residual U-Net (cf. Fig. 1d). Pre-estimated coil sensitivity maps were used in all transformations between k-space and image space. The convolution operations were implemented as either 2D spatial-only, 3D, or separable 2D+t spatiotemporal convolutions. A 2D+t convolution consisted of a 2D spatial convolution followed by an activation function and a 1D temporal convolution (cf. Fig. 1e). The OCMR dataset [3] was used for training and evaluation. It contains 183 fully sampled cine slices from 74 subjects in different imaging planes with 15 to 38 coils and 15 to 31 temporal frames at 1.5T and 3T. The training set contained 142 slices from 33 individuals, which includes all multi-slice acquisitions from the dataset. The validation set consisted of 22 single slices and the network performance was evaluated on the remaining, previously unseen 19 single-slice acquisitions. The k-space data was retrospectively undersampled using a variable density mask [4]. Sensitivity-based coil combination of the fully sampled k-space was used as ground truth. Coil sensitivity maps were estimated with ESPIRiT. Six networks were trained using the different convolution types mentioned above, each for acceleration factors R=8 and R=12. Results Exemplary reconstructions and quantitative results are given in Fig. 2. At R=8, reconstruction with 2D+t and 3D convolutions reached high structural similarity (SSIM) scores of around 0.95, while the 2D convolutions only reached 0.74. Artifacts are clearly visible in reconstructions learnt with 2D kernels. At R=12, reconstruction quality degraded most considerably with 2D convolution kernels. Conclusion Reconstruction of highly accelerated cardiac cine data using VNs yielded very good preliminary results. Convolutions that exploit temporal correlations are clearly beneficial over spatial-only convolutions. 2D+t spatiotemporal convolutions may be preferable over 3D convolutions due to reduced model capacity with similar reconstruction capability. Network design with 2D+t convolutionExample reconstruction and time profiles

Real time volumetric MRI for 3D motion tracking via geometry-informed deep learning.

To develop a geometry-informed deep learning framework for volumetric MRI with sub-second acquisition time in support of 3D motion tracking, which is highly desirable for improved radiotherapy precision but hindered by the long image acquisition time. A 2D-3D deep learning network with an explicitly defined geometry module that embeds geometric priors of the k-space encoding pattern was investigated, where a 2D generation network first augmented the sparsely sampled image dataset by generating new 2D representations of the underlying 3D subject. A geometry module then unfolded the 2D representations to the volumetric space. Finally, a 3D refinement network took the unfolded 3D data and outputted high-resolution volumetric images. Patient-specific models were trained for seven abdominal patients to reconstruct volumetric MRI from both orthogonal cine slices and sparse radial samples. To evaluate the robustness of the proposed method to longitudinal patient anatomy and position changes, we tested the trained model on separate datasets acquired more than one month later and evaluated 3D target motion tracking accuracy using the model-reconstructed images by deforming a reference MRI with gross tumor volume (GTV) contours to a 5-min time series of both ground truth and model-reconstructed volumetric images with a temporal resolution of 340ms. Across the seven patients evaluated, the median distances between model-predicted and ground truth GTV centroids in the superior-inferior direction were 0.4 ± 0.3mm and 0.5 ± 0.4mm for cine and radial acquisitions, respectively. The 95-percentile Hausdorff distances between model-predicted and ground truth GTV contours were 4.7 ± 1.1mm and 3.2 ± 1.5mm for cine and radial acquisitions, which are of the same scale as cross-plane image resolution. Incorporating geometric priors into deep learning model enables volumetric imaging with high spatial and temporal resolution, which is particularly valuable for 3D motion tracking and has the potential of greatly improving MRI-guided radiotherapy precision.

Feasibility of Single-Breath-Hold Compressed Sensing Real-Time Cine Imaging for Assessment of Ventricular Function and Left Ventricular Strain in Cardiac Magnetic Resonance

To explore the feasibility of single-breath-hold compressed sensing real-time cine imaging (CS-cine) in the assessment of ventricular function and left ventricular (LV) strain. A total of 70 subjects were enrolled prospectively, and all subjects underwent cardiac magnetic resonance imaging (cardiac MRI) using both the standard steady-state free procession cine (sta-cine) acquisition and a prototype CS-cine sequence. For both CS-cine and sta-cine imaging, continuous short-axis cine images were acquired from the base to the apex to cover the entire left ventricle, and long-axis cine images including two-, three-, and four-chamber views were also acquired. The scanning range, number of slices, slice thickness and intervals were kept identical for the two cine images of the same participant. Subjective evaluation of the image quality was performed on all cine images. For both sequences, the conventional function parameters of the left and the right ventricles and LV strain values were assessed with post-processing software analysis. The cine image quality, conventional ventricular function parameters, and LV strain values were compared between the two cine groups and the differences were examined. Inter- and intraobserver agreements for CS-cine images were measured using intraclass correlation coefficient ( ICC). Bland-Altman analysis was performed to assess reproducibility between the two cine methods. The median scanning time of CS-cine was 21 s versus 272 s for sta-cine ( P<0.001). The median image quality scores of two groups were significantly different, 4 points for sta-cine and 2 points for CS-cine ( P<0.001). Bi-ventricular end-diastolic volumes (EDV), stroke volume (SV) and ejection fraction (EF) were significantly smaller in CS-cine ( P<0.001). Nevertheless, no significant differences between the two groups in bi-ventricular ESV or LV mass were observed ( P>0.05). LV strain parameters, including the peak radial strain, peak circumferential strain and peak longitudinal strain derived from LV mid-ventricular slice, were significantly different in the two sequences ( P<0.001). Moreover, CS-cine-derived functional parameters and strain measurements have a good correlation with those of sta-cine (for RV function parameters, and left ventricular PLS, PCS values, more than 95% points fell within the limits of agreement [ LoA]; meanwhile, more than 91% points fell within the LoA for other parameters) and inter- and intraobserver agreements were strong ( ICC=0.88 to 0.99) for CS-cine. CS-cine can well realize the rapid acquisition of cine images for quantitative analysis of cardiac function, and the conventional ventricular function parameters and LV globalized strain values obtained from CS-cine imaging have good reproducibility.

Establishing a priori and a posteriori predictive models to assess patients' peak skin dose in interventional cardiology. Part 2: results of the VERIDIC project.

Optimizing patient exposure in interventional cardiology is key to avoid skin injuries. To establish predictive models of peak skin dose (PSD) during percutaneous coronary intervention (PCI), chronic total occlusion percutaneous coronary intervention (CTO), and transcatheter aortic valve implantation (TAVI) procedures. A total of 534 PCI, 219 CTO, and 209 TAVI were collected from 12 hospitals in eight European countries. Independent associations between PSD and clinical and technical dose determinants were examined for those procedures using multivariate statistical analysis. A priori and a posteriori predictive models were built using stepwise multiple linear regressions. A fourfold cross-validation was performed, and models' performance was evaluated using the root mean square error (RMSE), mean absolute percentage error (MAPE), coefficient of determination (R²), and linear correlation coefficient (r). Multivariate analysis proved technical parameters to overweight clinical complexity indices with PSD mainly affected by fluoroscopy time, tube voltage, tube current, distance to detector, and tube angulation for PCI. For CTO, these were body mass index, tube voltage, and fluoroscopy contribution. For TAVI, these parameters were sex, fluoroscopy time, tube voltage, and cine acquisitions. When benchmarking the predictive models, the correlation coefficients were r = 0.45 for the a priori model and r = 0.89 for the a posteriori model for PCI. These were 0.44 and 0.67, respectively, for the CTO a priori and a posteriori models, and 0.58 and 0.74, respectively, for the TAVI a priori and a posteriori models. A priori predictive models can help operators estimate the PSD before performing the intervention while a posteriori models are more accurate estimates and can be useful in the absence of skin dose mapping solutions.

Abstract 10639: Abbreviated Cardiac Magnetic Resonance Imaging versus Echocardiography for Interval Assessment of Systolic Function in Duchenne Muscular Dystrophy: Patient Satisfaction, Clinical Utility and Image Quality

Introduction: Interval assessment of systolic function in patients with Duchenne Muscular Dystrophy (DMD) is challenging by echocardiography (echo). Cardiac magnetic resonance imaging (CMR) has been limited due to study duration and patient experience. We developed an abbreviated CMR (aCMR) protocol which consisted only of cine acquisitions in the short axis, 4 chamber and vertical long axis planes, and compared it to echo in regard to: exam duration, patient satisfaction, clinical utility and image quality scores. Methods: DMD patients were recruited prospectively to undergo simultaneous echo and aCMR. Exclusion criteria included need for continuous positive pressure ventilation (PPV) and inability to lay supine ≥ 30 minutes. Subjects completed a patient satisfaction survey (PSS) comparing modalities. The clinical cardiologist completed a survey (CUS) assessing the clinical utility of each modality. A blinded expert reviewer completed an image quality survey (GQAS) for each modality. Results were compared using the Wilcoxon signed-rank test and Spearman correlation. Results: 19 DMD patients participated. PSS scores and exam duration were equivalent between modalities, while CUS and GQAS scores favored aCMR. ACMR scored particularly higher than echo in RV visualization and judgement of atrial size. Older age had a statistically significant negative correlation with echo GQAS and CUS scores, as well as aCMR PSS scores. Nighttime PPV requirement and non-ambulatory status were associated with worse echo CUS scores. Poor image quality precluding quantification existed in 4 (21%) echo and 0 aCMR studies. Conclusions: The aCMR protocol resulted in improved clinical utility and quality scores relative to echocardiography, without detriment in patient satisfaction or exam duration. For DMD patients known to have challenging echo acoustic windows, aCMR should be considered in lieu of echocardiography for interval assessment of systolic function.

Deep Learning for Classification and Selection of Cine CMR Images to Achieve Fully Automated Quality-Controlled CMR Analysis From Scanner to Report.

Introduction: Deep learning demonstrates great promise for automated analysis of CMR. However, existing limitations, such as insufficient quality control and selection of target acquisitions from the full CMR exam, are holding back the introduction of deep learning tools in the clinical environment. This study aimed to develop a framework for automated detection and quality-controlled selection of standard cine sequences images from clinical CMR exams, prior to analysis of cardiac function.Materials and Methods: Retrospective study of 3,827 subjects that underwent CMR imaging. We used a total of 119,285 CMR acquisitions, acquired with scanners of different magnetic field strengths and from different vendors (1.5T Siemens and 1.5T and 3.0T Phillips). We developed a framework to select one good acquisition for each conventional cine class. The framework consisted of a first pre-processing step to exclude still acquisitions; two sequential convolutional neural networks (CNN), the first (CNNclass) to classify acquisitions in standard cine views (2/3/4-chamber and short axis), the second (CNNQC) to classify acquisitions according to image quality and orientation; a final algorithm to select one good acquisition of each class. For each CNN component, 7 state-of-the-art architectures were trained for 200 epochs, with cross entropy loss and data augmentation. Data were divided into 80% for training, 10% for validation, and 10% for testing.Results: CNNclass selected cine CMR acquisitions with accuracy ranging from 0.989 to 0.998. Accuracy of CNNQC reached 0.861 for 2-chamber, 0.806 for 3-chamber, and 0.859 for 4-chamber. The complete framework was presented with 379 new full CMR studies, not used for CNN training/validation/testing, and selected one good 2-, 3-, and 4-chamber acquisition from each study with sensitivity to detect erroneous cases of 89.7, 93.2, and 93.9%, respectively.Conclusions: We developed an accurate quality-controlled framework for automated selection of cine acquisitions prior to image analysis. This framework is robust and generalizable as it was developed on multivendor data and could be used at the beginning of a pipeline for automated cine CMR analysis to obtain full automatization from scanner to report.

A data-driven semantic segmentation model for direct cardiac functional analysis based on undersampled radial MR cine series.

Imageacquisition and subsequent manual analysis of cardiac cine MRIis time-consuming. The purpose of this study was to train and evaluate a 3D artificial neural network for semanticsegmentation ofradially undersampledcardiacMRI to accelerate both scan time and postprocessing. A database ofCartesianshort-axisMR images of the heart(148,500 images, 484 examinations)was assembled fromanopenly accessibledatabaseandradial undersampling wassimulated.A3D U-Net architecture waspretrainedfor segmentation of undersampled spatiotemporal cineMRI.Transfer learning wasthen performed using samples from asecond database, comprising108non-Cartesian radial cine series of the midventricularmyocardium to optimize the performancefor authentic data.The performance was evaluated for different levelsof undersampling by theDice similarity coefficient(DSC)with respecttoreferencelabels, as well as by deriving ventricular volumes and myocardial masses. Without transfer learning,thepretrained modelperformedmoderately ontrueradial data [maximum number of projections tested, P=196;DSC =0.87(left ventricle),DSC=0.76(myocardium),and DSC =0.64(right ventricle)]. After transfer learning with authentic data, the predictions achieved human levelevenforhighundersamplingrates (P=33, DSC=0.95, 0.87, and 0.93)withoutsignificant difference compared withsegmentations derived fromfully sampled data. A 3D U-Net architecture can be used for semantic segmentation of radially undersampled cine acquisitions, achieving a performance comparable with human experts in fully sampled data. This approach can jointly acceleratetime-consumingcineimage acquisition andcumbersome manual image analysis.

Intracardiac echocardiography- versus transesophageal echocardiography-guided left atrial appendage occlusion with Watchman FLX.

Watchman FLX has been recently approved for left atrial appendage occlusion (LAAO) in the US. Intracardiac echocardiography (ICE) - which is already commonly used to guide trans-septal access - can serve as an alternative to TEE, simplifying the procedure and reducing associated costs. Herein, we report our experience with ICE-guided LAAO with Watchman FLX. This cohort study included the first 190 consecutive patients who underwent LAAO with Watchman FLX in our center. LAAO was successful in all patients without significant peri-procedural, device-related complications in either group. Compared to TEE, we observed a significant reduction in procedural times when using ICE. In addition, there was a potentially clinically relevant reduction in fluoroscopy dose, mainly secondary to fewer cine acquisition runs. At follow-up, no cases of device embolism were noted, whereas the rate of device-related thrombosis and peri-device leaks were comparable between groups. ICE-guided LAAO with Watchman FLX is safe and feasible, with a significant reduction in procedural time and potential reduction in fluoroscopy dose when compared to TEE.

B-PO04-111 ICE VS TEE-GUIDED LEFT ATRIAL APPENDAGE OCCLUSION WITH WATCHMAN FLX

Watchman FLX has been recently approved for LAAO in the US. While TEE is the standard imaging modality to guide its deployment, ICE - which is already used for transeptal access - can serve as an alternative simplifying the procedure and reducing associated costs. To compare the procedural and clinical outcomes of using ICE vs TEE to guide LAAO with Watchman FLX. The study included consecutive patients who underwent Watchman FLX LAAO with TEE (first n=20 patients) or ICE (the following n=20). LAAO was performed as a standalone procedure (n=7), or in following ablation of extra-LAA sites (n=33). For LAAO, ICE was positioned in the LSPV and in the anterior LA just proximal to the MVA to assess LAA anatomy and guide device deployment. Demographic characteristic, procedure parameters, and 45-days clinical outcomes were compared between TEE- and ICE-guided groups. Age, gender, BMI, and stroke risk where comparable between the two groups. LAAO was successful in all patients, with the correct device size always selected on the first attempt. There was no significant difference in procedural times and volume of iodinated contrast used (ΔICE-TEE: +2 min, 95% CI -21 to 25 min; 95% CI +9 mL, -8 to 25 mL), with a potential clinically relevant reduction in fluoroscopy dose with ICE (ΔICE-TEE: DAP -26264 mGy/cm2, 95% CI -57469 to 4940 mGy/cm2; AK -60 mGy, 95% CI -330 to 209 mGy). This reduction in fluoroscopy dose is mainly secondary to fewer cine acquisition runs with ICE (ΔICE-TEE: -31 secs; 95% CI -86 to 27 secs). There were no significant peri- or early procedural device-related complications in either group. At a median of 8 (6-11) weeks, no device-related thrombosis and 9 small (< 5 mm) peri-device leaks were detected on first the follow-up TEE. The rate of leaks was similar between the groups (TEE, n=4; ICE, n=5). ICE-guided Watchman FLX implant is safe and feasible, with a potential reduction in fluoroscopy dose when compared to TEE.

Occupational and Patient Radiation Dose and Quality Implications of Femoral Access Imaging During Coronary Angiography.

PurposeCardiologists often perform angiography of the common femoral artery (CFA) access site to evaluate whether the anatomy is suitable for deployment of a vascular closure device or to assess whether iatrogenic vessel damage has occurred. The choice of acquisition mode has radiation dose implications. The objective of this study was to investigate the influence of the selected type of CFA x-ray imaging mode (fluoro save, cine acquisition and digital subtraction angiography (DSA)) and tube angle on patient and staff dose during coronary angiography.Materials and MethodsAssessment of image quality for the different modes was performed to determine whether lower dose modes provide images of sufficient clinical quality to be routinely employed. Radiation dose levels for the patients (n=782), cardiologists (n=17), scrub nurses (n=27) and scout nurses (n=32) were measured in a prospective single-centre study between February 2017 and August 2019. Three Philips angiographic units and DoseAware dose monitoring systems were used.ResultsAmong the acquisition modes, fluoro save provided acceptable diagnostic quality for visualizing femoral access points and diagnosing pathology in 99% of cases. Average patient dose area product (DAP) was 83.95, 742.50, and 3363.41mGy2 and average patient air kerma (AK) was 0.87, 8.44, and 18.61mGy for fluoro save, cine, and DSA acquisitions, respectively. The use of higher dose imaging modes, imaging in the contralateral view and utilizing steeper TA was associated with a higher patient dose. Due to staff dose being highly correlated with DAP and AK, it was difficult to observe any association between staff dose and CFA imaging mode. However, this does not discount a potential increase in occupational dose due to the use of cine angiography or digital subtraction angiography during CFA imaging.ConclusionDSA of the CFA should be avoided during transfemoral coronary angiography unless critical to diagnostic analysis. It is recommended that fluoroscopic operators consider utilizing lower dose modes in the ipsilateral orientation ≤32° TA to reduce the risk of patient and staff radiation exposure.

Free-breathing Accelerated Cardiac MRI Using Deep Learning: Validation in Children and Young Adults.

Background Obtaining ventricular volumetry and mass is key to most cardiac MRI but challenged by long multibreath-hold acquisitions. Purpose To assess the image quality and performance of a highly accelerated, free-breathing, two-dimensional cine cardiac MRI sequence incorporating deep learning (DL) reconstruction compared with reference standard balanced steady-state free precession (bSSFP). Materials and Methods A DL algorithm was developed to reconstruct custom 12-fold accelerated bSSFP cardiac MRI cine images from coil sensitivity maps using 15 iterations of separable three-dimensional convolutions and data consistency steps. The model was trained, validated, and internally tested in 10, two, and 10 adult human volunteers, respectively, based on vendor partner-supplied fully sampled bSSFP acquisitions. For prospective external clinical validation, consecutive children and young adults undergoing cardiac MRI from September through December 2019 at a single children's hospital underwent both conventional and highly accelerated short-axis bSSFP cine acquisitions in one MRI examination. Two radiologists scored overall and volumetric three-dimensional mesh image quality of all short-axis stacks on a five-point Likert scale and manually segmented endocardial and epicardial contours. Scan times and image quality were compared using the Wilcoxon rank sum test. Measurement agreement was assessed with intraclass correlation coefficient and Bland-Altman analysis. Results Fifty participants (mean age, 16 years ± 4 [standard deviation]; range, 5-30 years; 29 men) were evaluated. The mean prescribed acquisition times of accelerated scans (non-breath-held) and bSSFP (excluding breath-hold time) were 0.9 minute ± 0.3 versus 3.0 minutes ± 1.9 (P < .001). Overall and three-dimensional mesh image quality scores were, respectively, 3.8 ± 0.6 versus 4.3 ± 0.6 (P < .001) and 4.0 ± 1.0 versus 4.4 ± 0.8 (P < .001). Raters had strong agreement between all bSSFP and DL measurements, with intraclass correlation coefficients of 0.76 to 0.97, near-zero mean differences, and narrow limits of agreement. Conclusion With slightly lower image quality yet much faster speed, deep learning reconstruction may allow substantially shorter acquisition times of cardiac MRI compared with conventional balanced steady-state free precession MRI performed for ventricular volumetry. © RSNA, 2021 Online supplemental material is available for this article.

Elastic AlignedSENSE for Dynamic MR Reconstruction: A Proof of Concept in Cardiac Cine.

Numerous methods in the extensive literature on magnetic resonance imaging (MRI) reconstruction exploit temporal redundancy to accelerate cardiac cine. Some of them include motion compensation, which involves high computational costs and long runtimes. In this work, we proposed a method—elastic alignedSENSE (EAS)—for the direct reconstruction of a motion-free image plus a set of nonrigid deformations to reconstruct a 2D cardiac sequence. The feasibility of the proposed approach was tested in 2D Cartesian and golden radial multi-coil breath-hold cardiac cine acquisitions. The proposed approach was compared against parallel imaging compressed sense (sPICS) and group-wise motion corrected compressed sense (GWCS) reconstructions. EAS provides better results on objective measures with considerable less runtime when an acceleration factor is higher than 10×. Subjective assessment of an expert, however, invited proposing the combination of EAS and GWCS as a preferable alternative to GWCS or EAS in isolation.

The Burden of Radiation Exposure During Transcatheter Closure of Atrial Septal Defect

Radiation reduction in the pediatric cardiac catheterization laboratory is well-suited for targeted quality improvement (QI) interventions. Transcatheter atrial septal defect (ASD) closure was chosen for this QI project based on a homogenous procedural population and inter-operator variability in radiation usage, with the aim to reduce radiation exposure during ASD device closure by 50% over 1 year. The aim for this project was defined and a Key Driver Diagram (KDD) was created with three domains for change: modification of procedural practice, reporting and monitoring/feedback, and team engagement. All patients undergoing attempted transcatheter ASD closure were considered for inclusion. The primary outcome, % reduction in median radiation dose (DAP/Kg), was determined through comparison with a historical cohort. Additional radiation metrics, procedural characteristics, and adverse events (AE) were compared to the historical cohort. Radiation exposure (DAP/kg) was reduced by 55% with a median dose reduction from 26 (15, 61) in a historical cohort to 12 (6, 22) in the intervention population (p <0.001). Fluoroscopy time and cine acquisition utilization significantly decreased. Procedure time, procedural success (defined as successful delivery of the device) and AE did not increase in the QI cohort. Successful practice changes included standardized procedural strategies to limit fluoroscopy and cine acquisition, improved fluoroscopic practice, engagement of the multidisciplinary team, and feedback with data reporting by electronic and in-person reminders. In conclusion, application of QI methodologies such as KDD with engagement of a multidisciplinary team can effectively reduce radiation in the pediatric catheterization laboratory.

Radiation Exposure Reduction and Patient Outcome by Using Very Low Frame Rate Fluoroscopy Protocol (3.8 + 7.5 fps) During Percutaneous Coronary Intervention.

Objectives: In this study, we intend to analyze the feasibility and efficacy of very low frame rate fluoroscopy (VLFF) protocol using a combination of 3.8 and 7.5 fps while performing Percutaneous Coronary Intervention (PCI).Methods: A retrospective cohort including 193 patients undergoing PCI under the VLFF protocol (Post-VLFF group) was compared with a retrospective cohort of 133 patients, who underwent PCI prior to implementation of VLFF protocol (Pre-VLFF group). In the Pre-VLFF group, all PCIs were performed using fluoroscopy frame rate of 15 fps. In the Post-VLFF group, 3.8 fps was used to guide catheter engagement, coronary lesion wiring, pre-and post-dilation, and 7.5 fps was used for lesion assessment and stent placement. Increasing use of fluoroscopic storage in place of cineangiography was also encouraged. Cine acquisition in both groups was performed at 15 fps. Primary endpoint was radiation exposure measured by Air Kerma. Secondary endpoints were procedure related outcomes and patient related outcomes (Major Adverse Cardiac Events including all-cause mortality, Target Lesion Failure, Myocardial Infarction, and Stroke).RESULTS: Post-VLFF group showed 74.7% reduction in Air Kerma as compared to Pre-VLFF group (433 ± 27 mGy vs. 1,714 ± 140 mGy; p < 0.0001), with no increase in the fluoroscopy time (15.38 ± 0.98 min Post-VLFF vs. 17.06 ± 1.29 min Pre-VLFF; p = 0.529) and contrast volume (116.5 ± 4.9 ml Post-VLFF vs. 116.7 ± 6 ml Pre-VLFF; p = 0.700). Both groups had comparable procedural success and complications rates as well as incidence of MACE.Conclusions: The very low frame rate fluoroscopy protocol is a feasible, effective, and safe method to significantly reduce the radiation exposure during PCI without any compromise on procedural and patient outcomes.

Quantification of Biventricular Myocardial Strain Using CMR Feature Tracking: Reproducibility in Small Animals.

Myocardial strain is a well-validated parameter for evaluating myocardial contraction. Cardiovascular magnetic resonance myocardial feature tracking (CMR-FT) is a novel method for the quantitative measurements of myocardial strain from routine cine acquisitions. In this study, we investigated the influence of temporal resolution on tracking accuracy of CMR-FT and the intraobserver, interobserver, and interstudy reproducibilities for biventricular strain analysis in mice from self-gated CMR at 11.7 T. 12 constitutive nexilin knockout (Nexn-KO) mice, heterozygous (Het, N = 6) and wild-type (WT, N = 6), were measured with a well-established self-gating sequence twice within two weeks. CMR-FT measures of biventricular global and segmental strain parameters were derived. Interstudy, intraobserver, and interobserver reproducibilities were investigated. For the assessment of the impact of the temporal resolution for the outcome in CMR-FT, highly oversampled semi-4 chamber and midventricular short-axis data were acquired and reconstructed with 10 to 80 phases per cardiac cycle. A generally reduced biventricular myocardial strain was observed in Nexn-KO Het mice. Excellent intraobserver and interobserver reproducibility was achieved in all global strains (ICC range from 0.76 to 0.99), where global right ventricle circumferential strain (RCSSAX) showed an only good interobserver reproducibility (ICC 0.65, 0.11-0.89). For interstudy reproducibility, left ventricle longitudinal strain (LLSLAX) was the most reproducible measure of strain (ICC 0.90, 0.71-0.97). The left ventricle radial strain (LRSSAX) (ICC 0.50, 0.10-0.83) showed fair reproducibility and RCSSAX (ICC 0.36, 0.14-0.74) showed only poor reproducibility. In general, compared with global strains, the segmental strains showed relatively lower reproducibility. A minimal temporal resolution of 20 phases per cardiac cycle appeared sufficient for CMR-FT strain analysis. The analysis of myocardial strain from high-resolution self-gated cine images by CMR-FT provides a highly reproducible method for assessing myocardial contraction in small rodent animals. Especially, global LV longitudinal and circumferential strain revealed excellent reproducibility of intra- and interobserver and interstudy measurements.

Cardiac magnetic resonance imaging features prognostic information in patients with suspected myocardial infarction with non-obstructed coronary arteries

BackgroundTo assess the prognostic implications of cardiac magnetic resonance imaging (CMR) in patients with clinical suspicion of myocardial infarction with non-obstructed coronary arteries (MINOCA). MethodsA total of 145 patients (58 ± 15 years, 97 men) were retrospectively enrolled in this single-center, longitudinal observational study. All patients underwent CMR including cine, edema-sensitive, and late gadolinium enhancement acquisitions, within a median of 3 days after cardiac catheterization. Follow-up was performed by medical records chart review and phone interviews; the median follow-up time was 4.2 years. The primary endpoint was defined as a combination of death, stroke, new onset of congestive heart failure, recurrent hospitalization, or the need for an invasive cardiac procedure. ResultsIn 143 (98.6%) cases, CMR revealed the following cardiac pathologies: myocarditis (n = 48, 33.1%), structural cardiomyopathies (n = 40, 27.6%), “true” myocardial infarction (n = 22, 15.1%), hypertensive heart disease (n = 19, 13.1%), and Tako-Tsubo cardiomyopathy (n = 14, 9.7%). Only two patients (1.4%) had a normal CMR examination. There were significant prognostic differences between different etiologies, e.g. myocarditis and Tako-Tsubo cardiomyopathy had a more favorable prognosis then structural cardiomyopathies. Age, end-diastolic volume index and time-to-CMR showed significant association with the primary endpoint in multi-variate Cox regression. ConclusionsCMR performed early after the onset of clinical symptoms allows discrimination between acute myocardial injury from “true” MINOCA in patients presenting with chest pain and elevated cardiac biomarkers, thereby helping to identify the underlying pathology in suspected MINOCA and allowing risk stratification based on the established diagnosis. Furthermore, CMR parameters allow for improved prediction of adverse events compared to clinical and laboratory parameters.