Deep learning based assessment of hemodynamics in the coarctation of the aorta: comparison of bidirectional recurrent and convolutional neural networks.

The utilization of numerical methods, such as computational fluid dynamics (CFD), has been widely established for modeling patient-specific hemodynamics based on medical imaging data. Hemodynamics assessment plays a crucial role in treatment decisions for the coarctation of the aorta (CoA), a congenital heart disease, with the pressure drop (PD) being a crucial biomarker for CoA treatment decisions. However, implementing CFD methods in the clinical environment remains challenging due to their computational cost and the requirement for expert knowledge. This study proposes a deep learning approach to mitigate the computational need and produce fast results. Building upon a previous proof-of-concept study, we compared the effects of two different artificial neural network (ANN) architectures trained on data with different dimensionalities, both capable of predicting hemodynamic parameters in CoA patients: a one-dimensional bidirectional recurrent neural network (1D BRNN) and a three-dimensional convolutional neural network (3D CNN). The performance was evaluated by median point-wise root mean square error (RMSE) for pressures along the centerline in 18 test cases, which were not included in a training cohort. We found that the 3D CNN (median RMSE of 3.23 mmHg) outperforms the 1D BRNN (median RMSE of 4.25 mmHg). In contrast, the 1D BRNN is more precise in PD prediction, with a lower standard deviation of the error (±7.03 mmHg) compared to the 3D CNN (±8.91 mmHg). The differences between both ANNs are not statistically significant, suggesting that compressing the 3D aorta hemodynamics into a 1D centerline representation does not result in the loss of valuable information when training ANN models. Additionally, we evaluated the utility of the synthetic geometries of the aortas with CoA generated by using a statistical shape model (SSM), as well as the impact of aortic arch geometry (gothic arch shape) on the model's training. The results show that incorporating a synthetic cohort obtained through the SSM of the clinical cohort does not significantly increase the model's accuracy, indicating that the synthetic cohort generation might be oversimplified. Furthermore, our study reveals that selecting training cases based on aortic arch shape (gothic versus non-gothic) does not improve ANN performance for test cases sharing the same shape.

CARDIOKIN1: Computational Assessment of Myocardial Metabolic Capability in Healthy Controls and Patients With Valve Diseases.

BACKGROUND: Many heart diseases can result in reduced pumping capacity of the heart muscle. A mismatch between ATP demand and ATP production of cardiomyocytes is one of the possible causes. Assessment of the relation between myocardial ATP production (MVATP) and cardiac workload is important for better understanding disease development and choice of nutritional or pharmacologic treatment strategies. Because there is no method for measuring MVATP in vivo, the use of physiology-based metabolic models in conjunction with protein abundance data is an attractive approach. METHOD: We developed a comprehensive kinetic model of cardiac energy metabolism (CARDIOKIN1) that recapitulates numerous experimental findings on cardiac metabolism obtained with isolated cardiomyocytes, perfused animal hearts, and in vivo studies with humans. We used the model to assess the energy status of the left ventricle of healthy participants and patients with aortic stenosis and mitral valve insufficiency. Maximal enzyme activities were individually scaled by means of protein abundances in left ventricle tissue samples. The energy status of the left ventricle was quantified by the ATP consumption at rest (MVATP[rest]), at maximal workload (MVATP[max]), and by the myocardial ATP production reserve, representing the span between MVATP(rest) and MVATP(max). RESULTS: Compared with controls, in both groups of patients, MVATP(rest) was increased and MVATP(max) was decreased, resulting in a decreased myocardial ATP production reserve, although all patients had preserved ejection fraction. The variance of the energetic status was high, ranging from decreased to normal values. In both patient groups, the energetic status was tightly associated with mechanic energy demand. A decrease of MVATP(max) was associated with a decrease of the cardiac output, indicating that cardiac functionality and energetic performance of the ventricle are closely coupled. CONCLUSIONS: Our analysis suggests that the ATP-producing capacity of the left ventricle of patients with valvular dysfunction is generally diminished and correlates positively with mechanical energy demand and cardiac output. However, large differences exist in the energetic state of the myocardium even in patients with similar clinical or image-based markers of hypertrophy and pump function. Registration: URL: https://www.clinicaltrials.gov; Unique identifiers: NCT03172338 and NCT04068740.

Computed Tomography-Based Assessment of Transvalvular Pressure Gradient in Aortic Stenosis.

Background: In patients with aortic stenosis, computed tomography (CT) provides important information about cardiovascular anatomy for treatment planning but is limited in determining relevant hemodynamic parameters such as the transvalvular pressure gradient (TPG). Purpose: In the present study, we aimed to validate a reduced-order model method for assessing TPG in aortic stenosis using CT data. Methods: TPGCT was calculated using a reduced-order model requiring the patient-specific peak-systolic aortic flow rate (Q) and the aortic valve area (AVA). AVA was determined by segmentation of the aortic valve leaflets, whereas Q was quantified based on volumetric assessment of the left ventricle. For validation, invasively measured TPGcatheter was calculated from pressure measurements in the left ventricle and the ascending aorta. Altogether, 84 data sets of patients with aortic stenosis were used to compare TPGCT against TPGcatheter. Results: TPGcatheter and TPGCT were 50.6 ± 28.0 and 48.0 ± 26 mmHg, respectively (p = 0.56). A Bland-Altman analysis revealed good agreement between both methods with a mean difference in TPG of 2.6 mmHg and a standard deviation of 19.3 mmHg. Both methods showed good correlation with r = 0.72 (p < 0.001). Conclusions: The presented CT-based method allows assessment of TPG in patients with aortic stenosis, extending the current capabilities of cardiac CT for diagnosis and treatment planning.

Assessment of 10-Year Left-Ventricular-Remodeling by CMR in Patients Following Aortic Valve Replacement.

Aims: Aortic valve replacement (AVR) may result in reverse cardiac remodeling. We aimed to assess long-term changes in the myocardium following AVR by Cardiac Magnetic Resonance Imaging (CMR). Methods: We prospectively observed the long-term left ventricular (LV) function and structure of 27 patients with AVR [n = 19 with aortic stenosis (AS); n = 8 with aortic regurgitation (AR)] by CMR. Patients underwent CMR before, as well as 1, 5, and 10 years after AVR. We evaluated clinical parameters, LV volumes, mass, geometry, ejection fraction (EF), global myocardial longitudinal strain (MyoGLS), global myocardial circular strain (MyoGCS), hemodynamic forces (HemForces), and Late Gadolinium Enhancement (LGE). Results: The median of LVMI, EDVI, and ESVI decreased in both groups. Patients with AR had higher initial values of EDVI and ESVI and showed a more prominent initial reduction. In AS, MyoGLS improved already after 1 year and remained constant afterward, whereas, in AR no improvement of MyoGLS was found. MyoGCS remained unchanged in the AS group but deteriorated in the AR group over 10 years. Ejection fraction (EF) was higher in AS patients compared to AR 10 years post-AVR. Late gadolinium enhancement (LGE) could be found more frequently in AS patients. Conclusion: CMR was well suited to investigate myocardial changes over a 10-year follow up period in patients with aortic valve disease. Regarding the long-term functional changes following AVR, patients with AR seemed to benefit less from AVR compared to AS patients. Fibrosis was more common in AS, but this did not reflect functional evolution in these patients. Close monitoring seems indispensable to avoid irreversible structural damage of the heart and to perform AVR at an appropriate stage.

Ectopic beats arise from micro-reentries near infarct regions in simulations of a patient-specific heart model.

Ectopic beats are known to be involved in the initiation of a variety of cardiac arrhythmias. Although their location may vary, ectopic excitations have been found to originate from infarct areas, regions of micro-fibrosis and other heterogeneous tissues. However, the underlying mechanisms that link ectopic foci to heterogeneous tissues have yet to be fully understood. In this work, we investigate the mechanism of micro-reentry that leads to the generation of ectopic beats near infarct areas using a patient-specific heart model. The patient-specific geometrical model of the heart, including scar and peri-infarct zones, is obtained through magnetic resonance imaging (MRI). The infarct region is composed of ischemic myocytes and non-conducting cells (fibrosis, for instance). Electrophysiology is captured using an established cardiac myocyte model of the human ventricle modified to describe ischemia. The simulation results clearly reveal that ectopic beats emerge from micro-reentries that are sustained by the heterogeneous structure of the infarct regions. Because microscopic information about the heterogeneous structure of the infarct regions is not available, Monte-Carlo simulations are used to identify the probabilities of an infarct region to behave as an ectopic focus for different levels of ischemia and different percentages of non-conducting cells. From the proposed model, it is observed that ectopic beats are generated when a percentage of non-conducting cells is near a topological metric known as the percolation threshold. Although the mechanism for micro-reentries was proposed half a century ago to be a source of ectopic beats or premature ventricular contractions during myocardial infarction, the present study is the first to reproduce this mechanism in-silico using patient-specific data.

Assessment of wall stresses and mechanical heart power in the left ventricle: Finite element modeling versus Laplace analysis.

INTRODUCTION: Stenotic aortic valve disease (AS) causes pressure overload of the left ventricle (LV) that may trigger adverse remodeling and precipitate progression towards heart failure (HF). As myocardial energetics can be impaired during AS, LV wall stresses and biomechanical power provide a complementary view of LV performance that may aide in better assessing the state of disease. OBJECTIVES: Using a high-resolution electro-mechanical (EM) in silico model of the LV as a reference, we evaluated clinically feasible Laplace-based methods for assessing global LV wall stresses and biomechanical power. METHODS: We used N = 4 in silico finite element (FE) EM models of LV and aorta of patients suffering from AS. All models were personalized with clinical data under pretreatment conditions. Left ventricle wall stresses and biomechanical power were computed accurately from FE kinematic data and compared with Laplace-based estimation methods, which were applied to the same FE model data. RESULTS AND CONCLUSION: Laplace estimates of LV wall stress are able to provide a rough approximation of global mean stress in the circumferential-longitudinal plane of the LV. However, according to FE results, spatial heterogeneity of stresses in the LV wall is significant, leading to major discrepancies between local stresses and global mean stress. Assessment of mechanical power with Laplace methods is feasible, but these are inferior in accuracy compared with FE models. The accurate assessment of stress and power density distribution in the LV wall is only feasible based on patient-specific FE modeling.

Avoidable costs of stenting for aortic coarctation in the United Kingdom: an economic model.

BACKGROUND: Undesirable outcomes in health care are associated with patient harm and substantial excess costs. Coarctation of the aorta (CoA), one of the most common congenital heart diseases, can be repaired with stenting but requires monitoring and subsequent interventions to detect and treat disease recurrence and aortic wall injuries. Avoidable costs associated with stenting in patients with CoA are unknown. METHODS: We developed an economic model to calculate potentially avoidable costs in stenting treatment of CoA in the United Kingdom over 5 years. We calculated baseline costs for the intervention and potentially avoidable complications and follow-up interventions and compared these to the costs in hypothetical scenarios with improved treatment effectiveness and complication rates. RESULTS: Baseline costs were £16 688 ($25 182) per patient. Avoidable costs ranged from £137 ($207) per patient in a scenario assuming a 10% reduction in aortic wall injuries and reinterventions at follow-up, to £1627 ($2455) in a Best-case scenario with 100% treatment success and no complications. Overall costs in the Best-case scenario were 90.2% of overall costs at Baseline. Reintervention rate at follow-up was identified as most influential lever for overall costs. Probabilistic sensitivity analysis showed a considerable degree of uncertainty for avoidable costs with widely overlapping 95% confidence intervals. CONCLUSIONS: Significant improvements in the treatment effectiveness and reductions in complication rates are required to realize discernible cost savings. Up to 10% of total baseline costs could be avoided in the best-case scenario. This highlights the need to pursue patient-specific treatment approaches which promise optimal outcomes.

Assessment of cardiac function and myocardial morphology using small animal Look-Locker inversion recovery (SALLI) MRI in rats.

Small animal magnetic resonance imaging is an important tool to study cardiac function and changes in myocardial tissue. The high heart rates of small animals (200 to 600 beats/min) have previously limited the role of CMR imaging. Small animal Look-Locker inversion recovery (SALLI) is a T1 mapping sequence for small animals to overcome this problem. T1 maps provide quantitative information about tissue alterations and contrast agent kinetics. It is also possible to detect diffuse myocardial processes such as interstitial fibrosis or edema. Furthermore, from a single set of image data, it is possible to examine heart function and myocardial scarring by generating cine and inversion recovery-prepared late gadolinium enhancement-type MR images. The presented video shows step-by-step the procedures to perform small animal CMR imaging. Here it is presented with a healthy Sprague-Dawley rat, however naturally it can be extended to different cardiac small animal models.

Cardiovascular magnetic resonance of myocardial edema using a short inversion time inversion recovery (STIR) black-blood technique: diagnostic accuracy of visual and semi-quantitative assessment.

BACKGROUND: The short inversion time inversion recovery (STIR) black-blood technique has been used to visualize myocardial edema, and thus to differentiate acute from chronic myocardial lesions. However, some cardiovascular magnetic resonance (CMR) groups have reported variable image quality, and hence the diagnostic value of STIR in routine clinical practice has been put into question. The aim of our study was to analyze image quality and diagnostic performance of STIR using a set of pulse sequence parameters dedicated to edema detection, and to discuss possible factors that influence image quality. We hypothesized that STIR imaging is an accurate and robust way of detecting myocardial edema in non-selected patients with acute myocardial infarction. METHODS: Forty-six consecutive patients with acute myocardial infarction underwent CMR (day 4.5, +/- 1.6) including STIR for the assessment of myocardial edema and late gadolinium enhancement (LGE) for quantification of myocardial necrosis. Thirty of these patients underwent a follow-up CMR at approximately six months (195 +/- 39 days). Both STIR and LGE images were evaluated separately on a segmental basis for image quality as well as for presence and extent of myocardial hyper-intensity, with both visual and semi-quantitative (threshold-based) analysis. LGE was used as a reference standard for localization and extent of myocardial necrosis (acute) or scar (chronic). RESULTS: Image quality of STIR images was rated as diagnostic in 99.5% of cases. At the acute stage, the sensitivity and specificity of STIR to detect infarcted segments on visual assessment was 95% and 78% respectively, and on semi-quantitative assessment was 99% and 83%, respectively. STIR differentiated acutely from chronically infarcted segments with a sensitivity of 95% by both methods and with a specificity of 99% by visual assessment and 97% by semi-quantitative assessment. The extent of hyper-intense areas on acute STIR images was 85% larger than those on LGE images, with a larger myocardial salvage index in reperfused than in non-reperfused infarcts (p = 0.035). CONCLUSIONS: STIR with appropriate pulse sequence settings is accurate in detecting acute myocardial infarction (MI) and distinguishing acute from chronic MI with both visual and semi-quantitative analysis. Due to its unique technical characteristics, STIR should be regarded as an edema-weighted rather than a purely T2-weighted technique.

Assessment of diffuse myocardial fibrosis in rats using small-animal Look-Locker inversion recovery T1 mapping.

BACKGROUND: The concentration of gadopentetate dimeglumine in myocardium and blood can be assessed from T1 measurements and can be used to calculate the extracellular volume (ECV) of the myocardium. We hypothesized that diffuse myocardial fibrosis in a small-animal model could be quantitatively assessed by measuring myocardial ECV using small-animal Look-Locker inversion recovery T1 mapping. METHODS AND RESULTS: Sprague-Dawley rats (n=10) were subjected to continuous angiotensin-2 (AT2) infusion for 2 weeks via a subcutaneously implanted minipump system. Magnetic resonance imaging (MRI) was performed both before and after AT2 infusion. The MRI protocol included multislice cine imaging and before-and-after contrast small-animal Look-Locker inversion recovery T1 mapping and late gadolinium enhancement imaging. Myocardial ECV was calculated from hematocrit and T1 values of blood and myocardium. During the course of AT2 infusion, the mean±SD systolic blood pressure increased from 122±10.9 to 152±27.5 mm Hg (P=0.003). Normalized heart weight was significantly higher in AT2-treated animals than in control littermates (P=0.033). Cine MRI documented concentric left ventricular hypertrophy. Postcontrast myocardial T1 times were shortened after treatment (median [interquartile range], 712 [63] versus 820 [131] ms; P=0.002). Myocardial ECV increased from 17.2% (4.3%) before to 23.0% (6.2%) after AT2 treatment (P=0.031), which was accompanied by perivascular fibrosis and microscarring on myocardial histological analysis. There was a moderate level of correlation between ECV and collagen volume fraction, as assessed by histological analysis (r=0.69, P=0.013). CONCLUSIONS: In a small-animal model of left ventricular hypertrophy, contrast-enhanced T1 mapping can be used to detect diffuse myocardial fibrosis by quantification of myocardial ECV.

Comprehensive four-dimensional phase-contrast flow assessment in hemi-Fontan circulation: systemic-to-pulmonary collateral flow quantification.

Precise lung perfusion quantification is essential for evaluation of patients with hemi-Fontan surgery. It is possible for two-dimensional cardiac magnetic resonance phase contrast flow (two-dimensional flow) to evaluate non-invasively the systemic-to-pulmonary collateral blood flow. This case report intends to illustrate the benefit of four-dimensional flow over the current two-dimensional flow in the comprehensive assessment of hemi-Fontan patients.

Sex-specific pediatric percentiles for ventricular size and mass as reference values for cardiac MRI: assessment by steady-state free-precession and phase-contrast MRI flow.

BACKGROUND: Cardiac MRI is important in the treatment of children with congenital heart disease, but sufficient normative data are lacking. For ventricular volumes and mass, we sought to deliver reference centiles and to investigate sex effects. METHODS AND RESULTS: We included 114 healthy children and adolescents, uniformly distributed spanning an age range of 4 to 20 years, as required by the Lambda-Mu-Sigma method to achieve a percentile distribution, thus avoiding arbitrary age categories. Subjects underwent axial volumetry (1.5-T scanner) using standardized 2D steady-state free-precession and flow protocols. Percentiles were computed for age 8 to 20 years (99 subjects) because breath-holds were more consistent in this group. When indexed for body surface area or height, the centile curves of ventricular volumetric parameters showed allometric increase until adolescence, when a plateau was reached, with values comparable to published adult reference data. In contrast, ventricular mass centiles increased without plateau. There was a significant sex difference, with centiles reflecting larger values in boys than in girls (P<0.05) when ventricular volumes were indexed to body surface area or height but not when indexed to weight (exception: mass). There was excellent agreement of axial and short-axis volumetry and of volumetric and flow-derived stroke volumes. CONCLUSIONS: Percentiles for ventricular volumes and mass in healthy children have been established to serve as reference values in pediatric heart disease. Significant sex differences were noted when indexing volumes to body surface area or height. Unisex centiles related to weight may be considered for chamber volumes albeit not for mass.

Integrated assessment of diastolic and systolic ventricular function using diagnostic cardiac magnetic resonance catheterization: validation in pigs and application in a clinical pilot study.

OBJECTIVES: This study sought to develop and validate a method for the integrated analysis of systolic and diastolic ventricular function. BACKGROUND: An integrated approach to assess ventricular pump function, myocontractility (end-systolic pressure-volume relationship [ESPVR]), and diastolic compliance (end-diastolic pressure-volume relation [EDPVR]) is of high clinical value. Cardiac magnetic resonance (CMR) is well established for measuring global pump function, and catheterization-combined CMR was previously shown to accurately measure ESPVR, but not yet the EDPVR. METHODS: In 8 pigs, the CMR technique was compared with conductance catheter methods (gold standard) for measuring the EDPVR in the left and right ventricle. Measurements were performed at rest and during dobutamine administration. For CMR, the ESPVR was estimated with a single-beat approach by synchronizing invasive ventricular pressures with cine CMR-derived ventricular volumes. The EDPVR was determined during pre-load reduction from additional volume data that were obtained from real-time velocity-encoded CMR pulmonary/aortic blood flow measurements. Pre-load reduction was achieved by transient balloon occlusion of the inferior vena cava. The stiffness coefficient beta was calculated by an exponential fit from the EDPVR. After validation in the animal experiments, the EDPVR was assessed in a pilot study of 3 patients with a single ventricle using identical CMR and conductance catheter techniques. RESULTS: Bland-Altman tests showed good agreement between conductance catheter-derived and CMR-derived EDPVR. In both ventricles of the pigs, dobutamine enhanced myocontractility (p < 0.01), increased stroke volume (p < 0.01), and improved diastolic function. The latter was evidenced by shorter early relaxation (p < 0.05), a downward shift of the EDPVR, and a decreased stiffness coefficient beta (p < 0.05). In contrast, in the patients, early relaxation was inconspicuous but the EDPVR shifted left-upward and the stiffness constant remained unchanged. The observed changes in diastolic function were not significantly different when measured with conductance catheter and CMR. CONCLUSIONS: This novel CMR method provides differential information about diastolic function in conjunction with parameters of systolic contractility and global pump function.

Endovascular stents in pulmonary valve and artery in swine: feasibility study of MR imaging-guided deployment and postinterventional assessment.

PURPOSE: To assess the feasibility of using magnetic resonance (MR) imaging to guide stent deployment in the pulmonary valve and artery and evaluate, after stent deployment, the position and morphology of and blood flow through the stent. MATERIALS AND METHODS: Angiography and 1.5-T MR imaging were performed in a dual-imaging suite. Nitinol stents were placed in the pulmonary valve and main pulmonary artery in five pigs by using MR imaging guidance. For interactive MR imaging monitoring of catheter manipulation and stent delivery, balanced fast field-echo and T1-weighted turbo field-echo sequences were used. Visualization of the delivery system was based on T2* (with air as the contrast material) or T1 (with gadodiamide as the contrast material). After stent deployment, the position and morphology of and flow through the stent were verified with multiphase multisection balanced fast field-echo and velocity-encoded cine MR imaging. Findings at angiography and postmortem examination also helped verify stent placement. The paired Student t test was used for data analysis. RESULTS: The stent was successfully deployed in all animals. The stent was placed distal to the pulmonary valve in four animals and across the pulmonary valve in one animal. The position and morphology of the stent were clearly depicted on balanced fast field-echo images. In the animal with the stent placed across the pulmonary valve, the pulmonary regurgitant fraction was 37%; this was not seen in the animals with stents placed distal to the pulmonary valve. No complication (eg, stent migration, intramural injury, or vascular perforation) was noted during the intervention. Findings at angiography and postmortem examination confirmed the position of the stents. CONCLUSION: MR imaging has the potential to guide stent placement in the pulmonary valve or artery and to evaluate flow volume within the stent lumen after the intervention.