Tricuspid valve flow measurement using a deep learning framework for automated valve-tracking 2D phase contrast.

PURPOSE: Tricuspid valve flow velocities are challenging to measure with cardiovascular MR, as the rapidly moving valvular plane prohibits direct flow evaluation, but they are vitally important to diastolic function evaluation. We developed an automated valve-tracking 2D method for measuring flow through the dynamic tricuspid valve. METHODS: Nine healthy subjects and 2 patients were imaged. The approach uses a previously trained deep learning network, TVnet, to automatically track the tricuspid valve plane from long-axis cine images. Subsequently, the tracking information is used to acquire 2D phase contrast (PC) with a dynamic (moving) acquisition plane that tracks the valve. Direct diastolic net flows evaluated from the dynamic PC sequence were compared with flows from 2D-PC scans acquired in a static slice localized at the end-systolic valve position, and also ventricular stroke volumes (SVs) using both planimetry and 2D PC of the great vessels. RESULTS: The mean tricuspid valve systolic excursion was 17.8 ± 2.5 mm. The 2D valve-tracking PC net diastolic flow showed excellent correlation with SV by right-ventricle planimetry (bias ± 1.96 SD = -0.2 ± 10.4 mL, intraclass correlation coefficient [ICC] = 0.92) and aortic PC (-1.0 ± 13.8 mL, ICC = 0.87). In comparison, static tricuspid valve 2D PC also showed a strong correlation but had greater bias (p = 0.01) versus the right-ventricle SV (10.6 ± 16.1 mL, ICC = 0.61). In most (8 of 9) healthy subjects, trace regurgitation was measured at begin-systole. In one patient, valve-tracking PC displayed a high-velocity jet (380 cm/s) with maximal velocity agreeing with echocardiography. CONCLUSION: Automated valve-tracking 2D PC is a feasible route toward evaluation of tricuspid regurgitant velocities, potentially solving a major clinical challenge.

Non-invasive pressure-volume loops using the elastance model and CMR: a porcine validation at transient pre-loads.

Aims: Pressure-volume (PV) loops have utility in the evaluation of cardiac pathophysiology but require invasive measurements. Recently, a time-varying elastance model to derive PV loops non-invasively was proposed, using left ventricular (LV) volume by cardiovascular magnetic resonance (CMR) and brachial cuff pressure as inputs. Validation was performed using CMR and pressure measurements acquired on the same day, but not simultaneously, and without varying pre-loads. This study validates the non-invasive elastance model used to estimate PV loops at varying pre-loads, compared with simultaneous measurements of invasive pressure and volume from real-time CMR, acquired concurrent to an inferior vena cava (IVC) occlusion. Methods and results: We performed dynamic PV loop experiments under CMR guidance in 15 pigs (n = 7 naïve, n = 8 with ischaemic cardiomyopathy). Pre-load was altered by IVC occlusion, while simultaneously acquiring invasive LV pressures and volumes from real-time CMR. Pairing pressure and volume signals yielded invasive PV loops, and model-based PV loops were derived using real-time LV volumes. Haemodynamic parameters derived from invasive and model-based PV loops were compared. Across 15 pigs, 297 PV loops were recorded. Intra-class correlation coefficient (ICC) agreement was excellent between model-based and invasive parameters: stroke work (bias = 0.007 ± 0.03 J, ICC = 0.98), potential energy (bias = 0.02 ± 0.03 J, ICC = 0.99), ventricular energy efficiency (bias = -0.7 ± 2.7%, ICC = 0.98), contractility (bias = 0.04 ± 0.1 mmHg/mL, ICC = 0.97), and ventriculoarterial coupling (bias = 0.07 ± 0.15, ICC = 0.99). All haemodynamic parameters differed between naïve and cardiomyopathy animals (P < 0.05). The invasive vs. model-based PV loop dice similarity coefficient was 0.88 ± 0.04. Conclusion: An elastance model-based estimation of PV loops and associated haemodynamic parameters provided accurate measurements at transient loading conditions compared with invasive PV loops.

Evaluation of 12-lead electrocardiogram at 0.55T for improved cardiac monitoring in magnetic resonance imaging.

BACKGROUND: The 12-lead electrocardiogram (ECG) is a standard diagnostic tool for monitoring cardiac ischemia and heart rhythm during cardiac interventional procedures and stress testing. These procedures can benefit from magnetic resonance imaging (MRI) information; however, the MRI scanner magnetic field leads to ECG distortion that limits ECG interpretation. This study evaluated the potential for improved ECG interpretation in a "low field" 0.55T MRI scanner. METHODS: The 12-lead ECGs were recorded inside 0.55T, 1.5T, and 3T MRI scanners, as well as at scanner table "home" position in the fringe field and outside the scanner room (seven pigs). To assess interpretation of ischemic ECG changes in a 0.55T MRI scanner, ECGs were recorded before and after coronary artery occlusion (seven pigs). ECGs was also recorded for five healthy human volunteers in the 0.55T scanner. ECG error and variation were assessed over 2-minute recordings for ECG features relevant to clinical interpretation: the PR interval, QRS interval, J point, and ST segment. RESULTS: ECG error was lower at 0.55T compared to higher field scanners. Only at 0.55T table home position, did the error approach the guideline recommended 0.025 mV ceiling for ECG distortion (median 0.03 mV). At scanner isocenter, only in the 0.55T scanner did J point error fall within the 0.1 mV threshold for detecting myocardial ischemia (median 0.03 mV in pigs and 0.06 mV in healthy volunteers). Correlation of J point deviation inside versus outside the 0.55T scanner following coronary artery occlusion was excellent at scanner table home position (r2 = 0.97), and strong at scanner isocenter (r2 = 0.92). CONCLUSION: ECG distortion is improved in 0.55T compared to 1.5T and 3T MRI scanners. At scanner home position, ECG distortion at 0.55T is low enough that clinical interpretation appears feasible without need for more cumbersome patient repositioning. At 0.55T scanner isocenter, ST segment changes during coronary artery occlusion appear detectable but distortion is enough to obscure subtle ST segment changes that could be clinically relevant. Reduced ECG distortion in 0.55T scanners may simplify the problem of suppressing residual distortion by ECG cable positioning, averaging, and filtering and could reduce current restrictions on ECG monitoring during interventional MRI procedures.

Dynamic lung water MRI during exercise stress.

PURPOSE: Exercise-induced dyspnea caused by lung water is an early heart failure symptom. Dynamic lung water quantification during exercise is therefore of interest to detect early stage disease. This study developed a time-resolved 3D MRI method to quantify transient lung water dynamics during rest and exercise stress. METHODS: The method was evaluated in 15 healthy subjects and 2 patients with heart failure imaged in transitions between rest and exercise, and in a porcine model of dynamic extravascular lung water accumulation through mitral regurgitation (n = 5). Time-resolved images were acquired at 0.55T using a continuous 3D stack-of-spirals proton density weighted sequence with 3.5 mm isotropic resolution, and derived using a motion corrected sliding-window reconstruction with 90-s temporal resolution in 20-s increments. A supine MRI-compatible pedal ergometer was used for exercise. Global and regional lung water density (LWD) and percent change in LWD (ΔLWD) were automatically quantified. RESULTS: A ΔLWD increase of 3.3 ± 1.5% was achieved in the animals. Healthy subjects developed a ΔLWD of 7.8 ± 5.0% during moderate exercise, peaked at 16 ± 6.8% during vigorous exercise, and remained unchanged over 10 min at rest (-1.4 ± 3.5%, p = 0.18). Regional LWD were higher posteriorly compared the anterior lungs (rest: 33 ± 3.7% vs 20 ± 3.1%, p < 0.0001; peak exercise: 36 ± 5.5% vs 25 ± 4.6%, p < 0.0001). Accumulation rates were slower in patients than healthy subjects (2.0 ± 0.1%/min vs 2.6 ± 0.9%/min, respectively), whereas LWD were similar at rest (28 ± 10% and 28 ± 2.9%) and peak exercise (ΔLWD 17 ± 10% vs 16 ± 6.8%). CONCLUSION: Lung water dynamics can be quantified during exercise using continuous 3D MRI and a sliding-window image reconstruction.

Reshaping the Ventricle From Within: MIRTH (Myocardial Intramural Remodeling by Transvenous Tether) Ventriculoplasty in Swine.

MIRTH (Myocardial Intramural Remodeling by Transvenous Tether) is a transcatheter ventricular remodeling procedure. A transvenous tension element is placed within the walls of the beating left ventricle and shortened to narrow chamber dimensions. MIRTH uses 2 new techniques: controlled intramyocardial guidewire navigation and EDEN (Electrocardiographic Radial Depth Navigation). MIRTH caused a sustained reduction in chamber dimensions in healthy swine. Midventricular implants approximated papillary muscles. MIRTH shortening improved myocardial contractility in cardiomyopathy in a dose-dependent manner up to a threshold beyond which additional shortening reduced performance. MIRTH may help treat dilated cardiomyopathy. Clinical investigation is warranted.

Impact of Vasodilation on Oxygen-Enhanced Functional Lung MRI at 0.55 T.

BACKGROUND: Oxygen-enhanced magnetic resonance imaging (OE-MRI) can be used to assess regional lung function without ionizing radiation. Inhaled oxygen acts as a T1-shortening contrast agent to increase signal in T1-weighted (T1w) images. However, increase in proton density from pulmonary hyperoxic vasodilation may also contribute to the measured signal enhancement. Our aim was to quantify the relative contributions of the T1-shortening and vasodilatory effects of oxygen to signal enhancement in OE-MRI in both swine and healthy volunteers. METHODS: We imaged 14 anesthetized female swine (47 ± 8 kg) using a prototype 0.55 T high-performance MRI system while experimentally manipulating oxygenation and blood volume independently through oxygen titration, partial occlusion of the vena cava for volume reduction, and infusion of colloid fluid (6% hydroxyethyl starch) for volume increase. Ten healthy volunteers were imaged before, during, and after hyperoxia. Two proton density-weighted (PDw) and 2 T1w ultrashort echo time images were acquired per experimental state. The median PDw and T1w percent signal enhancement (PSE), compared with baseline room air, was calculated after image registration and correction for lung volume changes. Differences in median PSE were compared using Wilcoxon signed rank test. RESULTS: The PSE in PDw images after 100% oxygen was similar in swine (1.66% ± 1.41%, P = 0.01) and in healthy volunteers (1.99% ± 1.79%, P = 0.02), indicating that oxygen-induced pulmonary vasodilation causes ~2% lung proton density increase. The PSE in T1w images after 100% oxygen was also similar (swine, 9.20% ± 1.68%, P < 0.001; healthy volunteers, 10.10% ± 3.05%, P < 0.001). The PSE in T1w enhancement was oxygen dose-dependent in anesthetized swine, and we measured a dose-dependent PDw image signal increase from infused fluids. CONCLUSIONS: The contribution of oxygen-induced vasodilation to T1w OE-MRI signal was measurable using PDw imaging and was found to be ~2% in both anesthetized swine and in healthy volunteers. This finding may have implications for patients with regional or global hypoxia or vascular dysfunction undergoing OE-MRI and suggest that PDw imaging may be useful to account for oxygen-induced vasodilation in OE-MRI.

Dynamic pressure-volume loop analysis by simultaneous real-time cardiovascular magnetic resonance and left heart catheterization.

BACKGROUND: Left ventricular (LV) contractility and compliance are derived from pressure-volume (PV) loops during dynamic preload reduction, but reliable simultaneous measurements of pressure and volume are challenging with current technologies. We have developed a method to quantify contractility and compliance from PV loops during a dynamic preload reduction using simultaneous measurements of volume from real-time cardiovascular magnetic resonance (CMR) and invasive LV pressures with CMR-specific signal conditioning. METHODS: Dynamic PV loops were derived in 16 swine (n = 7 naïve, n = 6 with aortic banding to increase afterload, n = 3 with ischemic cardiomyopathy) while occluding the inferior vena cava (IVC). Occlusion was performed simultaneously with the acquisition of dynamic LV volume from long-axis real-time CMR at 0.55 T, and recordings of invasive LV and aortic pressures, electrocardiogram, and CMR gradient waveforms. PV loops were derived by synchronizing pressure and volume measurements. Linear regression of end-systolic- and end-diastolic- pressure-volume relationships enabled calculation of contractility. PV loops measurements in the CMR environment were compared to conductance PV loop catheter measurements in 5 animals. Long-axis 2D LV volumes were validated with short-axis-stack images. RESULTS: Simultaneous PV acquisition during IVC-occlusion was feasible. The cardiomyopathy model measured lower contractility (0.2 ± 0.1 mmHg/ml vs 0.6 ± 0.2 mmHg/ml) and increased compliance (12.0 ± 2.1 ml/mmHg vs 4.9 ± 1.1 ml/mmHg) compared to naïve animals. The pressure gradient across the aortic band was not clinically significant (10 ± 6 mmHg). Correspondingly, no differences were found between the naïve and banded pigs. Long-axis and short-axis LV volumes agreed well (difference 8.2 ± 14.5 ml at end-diastole, -2.8 ± 6.5 ml at end-systole). Agreement in contractility and compliance derived from conductance PV loop catheters and in the CMR environment was modest (intraclass correlation coefficient 0.56 and 0.44, respectively). CONCLUSIONS: Dynamic PV loops during a real-time CMR-guided preload reduction can be used to derive quantitative metrics of contractility and compliance, and provided more reliable volumetric measurements than conductance PV loop catheters.

Biventricular Pressure-Volume Loop Assessment Before and After Pulmonary Valve Replacement in Tetralogy of Fallot.

Patients with tetralogy of Fallot (TOF) may undergo pulmonary valve replacement (PVR) after initial full repair. We investigated indices of biventricular function, work and efficiency of TOF patients' using noninvasive pressure-volume (PV) loop analysis on cardiovascular magnetic resonance (CMR) images and compared pre-and post PVR groups. Biventricular segmentations of steady state free precession CMR images were performed using custom validated software (Segment version 2.0 R7067). Brachial cuff pressure estimated left ventricular (LV) systolic pressure. Right ventricular (RV) inputs were obtained from pre-PVR cardiac catheterization data. Biventricular PV loops were then derived using a time-varying elastance model. Twenty seven patients were studied: (22 pre-PVR, 5 post-PVR), mean age of 20±10.5 years and 83% male. RV stroke volume significantly differed before and after PVR (73.2±25 ml vs. 41±10 mL, P =0.01). RV stroke work (SW) and mean external power (MEP) were significantly less post-PVR, but there were no significant differences in the LV hemodynamic indices. TOF patients have reduced RV SW and MEP post-PVR suggesting improved hemodynamics. Noninvasive biventricular PV loop analysis shows potential for integration into standard CMR imaging of TOF and provides hemodynamic data that could influence management decisions.

Imaging gravity-induced lung water redistribution with automated inline processing at 0.55 T cardiovascular magnetic resonance.

BACKGROUND: Quantitative assessment of dynamic lung water accumulation is of interest to unmask latent heart failure. We develop and validate a free-breathing 3D ultrashort echo time (UTE) sequence with automated inline image processing to image changes in lung water density (LWD) using high-performance 0.55 T cardiovascular magnetic resonance (CMR). METHODS: Quantitative lung water CMR was performed on 15 healthy subjects using free-breathing 3D stack-of-spirals proton density weighted UTE at 0.55 T. Inline image reconstruction and automated image processing was performed using the Gadgetron framework. A gravity-induced redistribution of LWD was provoked by sequentially acquiring images in the supine, prone, and again supine position. Quantitative validation was performed in a phantom array of vials containing mixtures of water and deuterium oxide. RESULTS: The phantom experiment validated the capability of the sequence in quantifying water density (bias ± SD 4.3 ± 4.8%, intraclass correlation coefficient, ICC = 0.97). The average global LWD was comparable between imaging positions (supine 24.7 ± 3.4%, prone 22.7 ± 3.1%, second supine 25.3 ± 3.6%), with small differences between imaging phases (first supine vs prone 2.0%, p < 0.001; first supine vs second supine - 0.6%, p = 0.001; prone vs second supine - 2.7%, p < 0.001). In vivo test-retest repeatability in LWD was excellent (- 0.17 ± 0.91%, ICC = 0.97). A regional LWD redistribution was observed in all subjects when repositioning, with a predominant posterior LWD accumulation when supine, and anterior accumulation when prone (difference in anterior-posterior LWD: supine - 11.6 ± 2.7%, prone 5.5 ± 2.7%, second supine - 11.4 ± 2.9%). Global LWD maps were calculated inline within 23.2 ± 0.3 s following the image reconstruction using the automated pipeline. CONCLUSIONS: Redistribution of LWD due to gravitational forces can be depicted and quantified using a validated free-breathing 3D proton density weighted UTE sequence and inline automated image processing pipeline on a high-performance 0.55 T CMR system.

Transcatheter Myotomy to Relieve Left Ventricular Outflow Tract Obstruction: The Septal Scoring Along the Midline Endocardium Procedure in Animals.

BACKGROUND: Left ventricular outflow tract obstruction complicates hypertrophic cardiomyopathy and transcatheter mitral valve replacement. Septal reduction therapies including surgical myectomy and alcohol septal ablation are limited by surgical morbidity or coronary anatomy and high pacemaker rates, respectively. We developed a novel transcatheter procedure, mimicking surgical myotomy, called Septal Scoring Along the Midline Endocardium (SESAME). METHODS: SESAME was performed in 5 naive pigs and 5 pigs with percutaneous aortic banding-induced left ventricular hypertrophy. Fluoroscopy and intracardiac echocardiography guided the procedures. Coronary guiding catheters and guidewires were used to mechanically enter the basal interventricular septum. Imparting a tip bend to the guidewire enabled intramyocardial navigation with multiple df. The guidewire trajectory determined the geometry of SESAME myotomy. The myocardium was lacerated using transcatheter electrosurgery. Cardiac function and tissue characteristics were assessed by cardiac magnetic resonance at baseline, postprocedure, and at 7- or 30-day follow-up. RESULTS: SESAME myotomy along the intended trajectory was achieved in all animals. The myocardium splayed after laceration, increasing left ventricular outflow tract area (753 to 854 mm2, P=0.008). Two naive pigs developed ventricular septal defects due to excessively deep lacerations in thin baseline septa. No hypertrophy model pig, with increased septal thickness and left ventricular mass compared with naive pigs, developed ventricular septal defects. One animal developed left axis deviation on ECG but no higher conduction block was seen in any animal. Coronary artery branches were intact on angiography with no infarction on cardiac magnetic resonance late gadolinium imaging. Cardiac magnetic resonance chamber volumes, function, flow, and global strain were preserved. No myocardial edema was evident on cardiac magnetic resonance T1 mapping. CONCLUSIONS: This preclinical study demonstrated feasibility of SESAME, a novel transcatheter myotomy to relieve left ventricular outflow tract obstruction. This percutaneous procedure using available devices, with a safe surgical precedent, is readily translatable into patients.

Atrioventricular plane displacement and regional function to predict outcome in pulmonary arterial hypertension.

To investigate if left and right atrioventricular plane displacement (AVPD) or regional contributions to SV are prognostic for outcome in patients with pulmonary arterial hypertension (PAH). Seventy-one patients with PAH and 20 sex- and age-matched healthy controls underwent CMR. Myocardial borders and RV insertion points were defined at end diastole and end systole in cine short-axis stacks to compute biventricular volumes, lateral (SVlat%) and septal (SVsept%) contribution to stroke volume. Eight atrioventricular points were defined at end diastole and end systole in 2-, 3- and 4-chamber cine long-axis views for computation of AVPD and longitudinal contribution to stroke volume (SVlong%). Cut-off values for survival analysis were defined as two standard deviations above or below the mean of the controls. Outcome was defined as death or lung transplantation. Median follow-up time was 3.6 [IQR 3.7] years. Patients were 57 ± 19 years (65% women) and controls 58 ± 15 years (70% women). Biventricular AVPD, SVlong% and ejection fraction (EF) were lower and SVlat% was higher, while SVsept% was lower in PAH compared with controls. In PAH, transplantation-free survival was lower below cut-off for LV-AVPD (hazard ratio [HR] = 2.1, 95%CI 1.2-3.9, p = 0.02) and RV-AVPD (HR = 9.8, 95%CI 4.6-21.1, p = 0.005). In Cox regression analysis, lower LV-AVPD and RV-AVPD inferred lower transplantation-free survival (LV: HR = 1.16, p = 0.007; RV: HR = 1.11, p = 0.01; per mm decrease). LV-SVlong%, RV-SVlong%, LV-SVlat%, RV-SVlat%, SVsept% and LV- and RVEF did not affect outcome. Low left and right AVPD were associated with outcome in PAH, but regional contributions to stroke volume and EF were not.

MVnet: automated time-resolved tracking of the mitral valve plane in CMR long-axis cine images with residual neural networks: a multi-center, multi-vendor study.

BACKGROUND: Mitral annular plane systolic excursion (MAPSE) and left ventricular (LV) early diastolic velocity (e') are key metrics of systolic and diastolic function, but not often measured by cardiovascular magnetic resonance (CMR). Its derivation is possible with manual, precise annotation of the mitral valve (MV) insertion points along the cardiac cycle in both two and four-chamber long-axis cines, but this process is highly time-consuming, laborious, and prone to errors. A fully automated, consistent, fast, and accurate method for MV plane tracking is lacking. In this study, we propose MVnet, a deep learning approach for MV point localization and tracking capable of deriving such clinical metrics comparable to human expert-level performance, and validated it in a multi-vendor, multi-center clinical population. METHODS: The proposed pipeline first performs a coarse MV point annotation in a given cine accurately enough to apply an automated linear transformation task, which standardizes the size, cropping, resolution, and heart orientation, and second, tracks the MV points with high accuracy. The model was trained and evaluated on 38,854 cine images from 703 patients with diverse cardiovascular conditions, scanned on equipment from 3 main vendors, 16 centers, and 7 countries, and manually annotated by 10 observers. Agreement was assessed by the intra-class correlation coefficient (ICC) for both clinical metrics and by the distance error in the MV plane displacement. For inter-observer variability analysis, an additional pair of observers performed manual annotations in a randomly chosen set of 50 patients. RESULTS: MVnet achieved a fast segmentation (<1 s/cine) with excellent ICCs of 0.94 (MAPSE) and 0.93 (LV e') and a MV plane tracking error of -0.10 ± 0.97 mm. In a similar manner, the inter-observer variability analysis yielded ICCs of 0.95 and 0.89 and a tracking error of -0.15 ± 1.18 mm, respectively. CONCLUSION: A dual-stage deep learning approach for automated annotation of MV points for systolic and diastolic evaluation in CMR long-axis cine images was developed. The method is able to carefully track these points with high accuracy and in a timely manner. This will improve the feasibility of CMR methods which rely on valve tracking and increase their utility in a clinical setting.

Non-invasive quantification of pressure-volume loops from cardiovascular magnetic resonance at rest and during dobutamine stress.

Non-invasive quantification of pressure-volume (PV) loops from brachial pressure and cardiovascular magnetic resonance is a validated method but its application has been limited to resting heart rates. The aim of this study was to improve the previous method and validate it against invasive left-ventricular pressure measurements in an experimental porcine model, and further apply it to 16 healthy humans at rest and during dobutamine stress. In addition, the improved method calculates the arterial elastance which provides the computation of the ratio of effective arterial (Ea ) to maximal ventricular elastance (Emax ) representing the ventricular-arterial coupling. In the porcine model, the differences between the improved non-invasively derived PV loops and invasively measured PV loops were for stroke work (mean ± SD) 0.00 ± 0.03 J, ventricular efficiency -1.1 ± 0.4%, and contractility 1.1 ± 0.1 mmHg/ml. In human subjects during dobutamine stress, stroke work increased from 1.3 ± 0.3 to 1.7 ± 0.4 J, ventricular efficiency from 71 ± 4 to 82 ± 4%, contractility from 1.3 ± 0.2 to 2.3 ± 0.6 mmHg/ml, and the ratio of arterial to ventricular elastance decreased from 0.96 to 0.56. The improved method for non-invasive PV loops constitutes a more robust diagnostic tool for cardiac disease states in a wider range of study cohorts at both rest and during stress.

Automated left atrial time-resolved segmentation in MRI long-axis cine images using active contours.

BACKGROUND: Segmentation of the left atrium (LA) is required to evaluate atrial size and function, which are important imaging biomarkers for a wide range of cardiovascular conditions, such as atrial fibrillation, stroke, and diastolic dysfunction. LA segmentations are currently being performed manually, which is time-consuming and observer-dependent. METHODS: This study presents an automated image processing algorithm for time-resolved LA segmentation in cardiac magnetic resonance imaging (MRI) long-axis cine images of the 2-chamber (2ch) and 4-chamber (4ch) views using active contours. The proposed algorithm combines mitral valve tracking, automated threshold calculation, edge detection on a radially resampled image, edge tracking based on Dijkstra's algorithm, and post-processing involving smoothing and interpolation. The algorithm was evaluated in 37 patients diagnosed mainly with paroxysmal atrial fibrillation. Segmentation accuracy was assessed using the Dice similarity coefficient (DSC) and Hausdorff distance (HD), with manual segmentations in all time frames as the reference standard. For inter-observer variability analysis, a second observer performed manual segmentations at end-diastole and end-systole on all subjects. RESULTS: The proposed automated method achieved high performance in segmenting the LA in long-axis cine sequences, with a DSC of 0.96 for 2ch and 0.95 for 4ch, and an HD of 5.5 mm for 2ch and 6.4 mm for 4ch. The manual inter-observer variability analysis had an average DSC of 0.95 and an average HD of 4.9 mm. CONCLUSION: The proposed automated method achieved performance on par with human experts analyzing MRI images for evaluation of atrial size and function. Video Abstract.

Quantification of left ventricular contribution to stroke work by longitudinal and radial force-length loops.

Left ventricular (LV) stroke work (SW) is calculated from the pressure-volume (PV) loop. PV loops do not contain information on longitudinal and radial pumping, leaving their contributions to SW unknown. A conceptual framework is proposed to derive the longitudinal and radial contributions to SW, using ventricular force-length loops reflecting longitudinal and radial pumping. The aim of this study was to develop and validate this framework experimentally and to explore these contributions in healthy controls and heart failure patients. Thirteen swine underwent cardiovascular magnetic resonance (CMR) and LV pressure catheterization at baseline (n = 7) or 1 wk after myocardial infarction (n = 6). CMR and noninvasive PV loop quantification were performed on 26 human controls and 14 patients. Longitudinal and radial forces were calculated as LV pressure multiplied by the myocardial surface areas in the respective directions. Length components were defined as the atrioventricular plane and epicardial displacements, respectively. Contributions to SW were calculated as the area within the respective force-length loop. Summation of longitudinal and radial SW had excellent agreement with PV loop-derived SW (ICC = 0.95, R = 0.96, bias ± SD = -4.5 ± 5.4%) in swine. Longitudinal and radial contributions to SW were ~50/50% in swine and human controls, and 44/56% in patients. Longitudinal pumping required less work than radial to deliver stroke volume in swine (6.8 ± 0.8 vs. 8.7 ± 1.2 mJ/mL, P = 0.0002) and in humans (11 ± 2.1 vs. 17 ± 4.7 mJ/mL, P < 0.0001). In conclusion, longitudinal and radial pumping contribute ~50/50% to SW in swine and human controls and 44/56% in heart failure patients. Longitudinal pumping is more energy efficient than radial pumping in delivering stroke volume.NEW & NOTEWORTHY A novel method for quantifying the contributions of longitudinal and radial pumping to stroke work using global left ventricular force-length loops was proposed and validated, which can be quantified noninvasively using cardiovascular magnetic resonance and brachial cuff pressure. We found that longitudinal and radial pumping contributes equally to stroke work in controls and 44/56% in heart failure patients, and that the longitudinal pumping is more energy efficient in delivering stroke volume than radial pumping.

Evolution of left ventricular function among subjects with ST-elevation myocardial infarction after percutaneous coronary intervention.

BACKGROUND: Atrioventricular plane displacement (AVPD) reflects longitudinal left ventricular (LV) systolic function, and wall thickening (WT) regional radial LV function. The temporal evolution of these measures after STEMI with CMR has not been evaluated. We aimed to investigate how AVPD and WT are affected globally and regionally from the sub-acute to the chronic phase after ST-elevation myocardial infarction (STEMI). METHODS: Healthy volunteers without cardiovascular disease and medication (controls, n = 20) and patients from the CHILL-MI study ( NCT01379261 ) prospectively underwent magnetic resonance imaging (MRI) 2-6 days and 6 months after STEMI (n = 77). CHILL-MI randomized STEMI-patients to cooling therapy initiated before reperfusion or standard of care. AVPD was measured at six points in three long axis cine images and wall thickening in short axis cine images. Infarction was quantified using late gadolinium enhancement (LGE) and used to define infarct and remote segments. RESULTS: There were no difference in AVPD either at acute or chronic phase (p = 0.90 and p = 0.40) or WT (p = 0.85 and p = 0.99) between patients randomized to cooling therapy and standard of care. Therefore, the results are presented for the pooled cohort. Global AVPD was decreased in both the sub-acute (12 ± 2 mm, p < 0.001) and the chronic phase (13 ± 2 mm, p < 0.001) compared to controls (15 ± 2 mm) with a partial recovery of AVPD (p < 0.001) in the chronic phase. Patients with left anterior descending (LAD) and right coronary artery (RCA) infarcts had decreased AVPD in the chronic phase in both infarcted and remote segments. Mean WT was decreased in patients with LAD infarction both in the sub-acute and the chronic phase in both infarcted and remote segments. The decrease in WT in patients with RCA and left circumflex (LCx) infarcts was more affected in the infarcted segments, especially in the chronic phase. CONCLUSION: AVPD was a global rather than regional marker of cardiac function in this STEMI study and this may explain the prognostic importance of local measurements of mitral annular plane systolic excursion (MAPSE). The decrease in WT in remote myocardium even in the chronic phase needs to be taken into consideration when combining functional measurements with infarct quantification for diagnosis of post-ischemic stunning and hibernation.

Valvular imaging in the era of feature-tracking: A slice-following cardiac MR sequence to measure mitral flow.

BACKGROUND: In mitral valve dysfunction, noninvasive measurement of transmitral blood flow is an important clinical examination. Flow imaging of the mitral valve, however, is challenging, since it moves in and out of the image plane during the cardiac cycle. PURPOSE: To more accurately measure mitral flow, a slice-following MRI phase contrast sequence is proposed. This study aimed to implement such a sequence, validate its slice-following functionality in a phantom and healthy subjects, and test its feasibility in patients with mitral valve dysfunction. STUDY TYPE: Prospective. PHANTOM AND SUBJECTS: The slice-following functionality was validated in a cone-shaped phantom by measuring the depicted slice radius. Sixteen healthy subjects and 10 mitral valve dysfunction patients were enrolled at two sites. FIELD STRENGTH/SEQUENCE: 1.5T and 3T gradient echo cine phase contrast. ASSESSMENT: A single breath-hold retrospectively gated sequence using offline feature-tracking of the mitral valve was developed. Valve displacements were measured and imported to the scanner, allowing the slice position to change dynamically based on the cardiac phase. Mitral valve imaging was performed with slice-following and static imaging planes. Validation was performed by comparing mitral stroke volume with planimetric and aortic stroke volume. STATISTICAL TESTS: Measurements were compared using linear regression, Pearson's R, parametric paired t-tests, Bland-Altman analysis, and intraclass correlation coefficient (ICC). RESULTS: Phantom experiments confirmed accurate slice displacements. Slice-following was feasible in all subjects, yielding physiologically accurate mitral flow patterns. In healthy subjects, mitral and aortic stroke volumes agreed, with ICC = 0.72 and 0.90 for static and slice-following planes; with bias ±1 SDs 23.2 ± 13.2 mls and 8.4 ± 10.8 mls, respectively. Agreement with planimetry was stronger, with ICC = 0.84 and 0.96; bias ±1 SDs 13.7 ± 13.7 mls and -2.0 ± 8.8 mls for static and slice-following planes, respectively. DATA CONCLUSION: Slice-following outperformed the conventional sequence and improved the accuracy of transmitral flow, which is important for assessment of diastolic function and mitral regurgitation. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:1412-1421.

Noninvasive Quantification of Pressure-Volume Loops From Brachial Pressure and Cardiovascular Magnetic Resonance.

BACKGROUND: Pressure-volume (PV) loops provide a wealth of information on cardiac function but are not readily available in clinical routine or in clinical trials. This study aimed to develop and validate a noninvasive method to compute individualized left ventricular PV loops. METHODS: The proposed method is based on time-varying elastance, with experimentally optimized model parameters from a training set (n=5 pigs), yielding individualized PV loops. Model inputs are left ventricular volume curves from cardiovascular magnetic resonance imaging and brachial pressure. The method was experimentally validated in a separate set (n=9 pig experiments) using invasive pressure measurements and cardiovascular magnetic resonance images and subsequently applied to human healthy controls (n=13) and patients with heart failure (n=28). RESULTS: There was a moderate-to-excellent agreement between in vivo-measured and model-calculated stroke work (intraclass correlation coefficient, 0.93; bias, -0.02±0.03 J), mechanical potential energy (intraclass correlation coefficient, 0.57; bias, -0.04±0.03 J), and ventricular efficiency (intraclass correlation coefficient, 0.84; bias, 3.5±2.1%). The model yielded lower ventricular efficiency ( P<0.0001) and contractility ( P<0.0001) in patients with heart failure compared with controls, as well as a higher potential energy ( P<0.0001) and energy per ejected volume ( P<0.0001). Furthermore, the model produced realistic values of stroke work and physiologically representative PV loops. CONCLUSIONS: We have developed the first experimentally validated, noninvasive method to compute left ventricular PV loops and associated quantitative measures. The proposed method shows significant agreement with in vivo-derived measurements and could support clinical decision-making and provide surrogate end points in clinical heart failure trials.

Assessment of diastolic function and atrial remodeling by MRI - validation and correlation with echocardiography and filling pressure.

Atrial fibrosis can be estimated noninvasively by magnetic resonance imaging (MRI) using late gadolinium enhancement (LGE), but diastolic dysfunction is clinically assessed by transthoracic echocardiography (TTE), and rarely by MRI. This study aimed to evaluate well-established diastolic parameters using MRI, and validate them with TTE and left ventricular (LV) filling pressures, and to study the relationship between left atrial (LA) remodeling and parameters of diastolic function. The study retrospectively included 105 patients (53 ± 16 years, 39 females) who underwent 3D LGE MRI between 2012 and 2016. Medical charts were reviewed for the echocardiographic diastolic parameters E, A, and e' by TTE, and pressure catheterizations. E and A were measured from in-plane phase-contrast cardiac MRI images, and e' by feature-tracking, and validated with TTE. Interobserver and intraobserver variability was examined. Furthermore, LA volumes, function, and atrial LGE was correlated with diastolic parameters. Evaluation of e' in MRI had strong agreement with TTE (r = 0.75, P < 0.0001), and low interobserver and intraobserver variability. E and A by TTE showed strong agreement to MRI (r = 0.77, P = 0.001; r = 0.73, P = 0.003, for E and A, respectively). Agreement between E/e' by TTE and MRI was strong (r = 0.85, P = 0.0004), and E/e' by TTE correlated moderately to invasive pressures (r = 0.59, P = 0.03). There was a strong relationship between LA LGE and pulmonary capillary wedge pressure (r = 0.81, P = 0.01). In conclusion, diastolic parameters can be measured with good reproducibility by cardiovascular MRI. LA LGE exhibited a strong relationship with pulmonary capillary wedge pressure, an indicator of diastolic function.