Current diagnosis of heart failure with preserved ejection fraction (HFpEF) is suboptimal. We tested the hypothesis that comprehensive machine learning (ML) of left ventricular function at rest and ...exercise objectively captures differences between HFpEF and healthy subjects.
One hundred fifty-six subjects aged >60 years (72 HFpEF+33 healthy for the initial analyses; 24 hypertensive+27 breathless for independent evaluation) underwent stress echocardiography, in the MEDIA study (Metabolic Road to Diastolic Heart Failure). Left ventricular long-axis myocardial velocity patterns were analyzed using an unsupervised ML algorithm that orders subjects according to their similarity, allowing exploration of the main trends in velocity patterns. ML identified a continuum from health to disease, including a transition zone associated to an uncertain diagnosis. Clinical validation was performed (1) to characterize the main trends in the patterns for each zone, which corresponded to known characteristics and new features of HFpEF; the ML-diagnostic zones differed for age, body mass index, 6-minute walk distance, B-type natriuretic peptide, and left ventricular mass index (
<0.05) and (2) to evaluate the consistency of the proposed groupings against diagnosis by current clinical criteria; correlation with diagnosis was good (κ, 72.6%; 95% confidence interval, 58.1-87.0); ML identified 6% of healthy controls as HFpEF. Blinded reinterpretation of imaging from subjects with discordant clinical and ML diagnoses revealed abnormalities not included in diagnostic criteria. The algorithm was applied independently to another 51 subjects, classifying 33% of hypertensive and 67% of breathless controls as mild-HFpEF.
The analysis of left ventricular long-axis function on exercise by interpretable ML may improve the diagnosis and understanding of HFpEF.
Objective Management and follow-up of chronic aortic dissections continue to be a clinical challenge due to progressive dilatation and subsequent rupture. To predict complications, guidelines suggest ...follow-up of aortic diameter. However, dilatation is triggered by hemodynamic parameters (pressures/wall shear stresses) and geometry of false (FL) and true lumen (TL), information not captured by diameter alone. Therefore, we aimed at better understanding the influence of dissection anatomy on TL and FL hemodynamics. Methods In vitro studies were performed using pulsatile flow in realistic dissected latex/silicone geometries with varying tear number, size, and location. We assessed three different conformations: (1) proximal tear only; (2) distal tear only; (3) both proximal and distal tears. All possible combinations (n = 8) of small (10% of aortic diameter) and large (25% of aortic diameter) tears were considered. Pressure, velocity, and flow patterns were analyzed within the lumina (at proximal and distal sections) and at the tears. We also computed the FL mean pressure index (FPImean %) as a percentage of the TL mean pressure, to compare pressures among models. Results The presence of large tears equalized FL/TL pressures compared with models with only small tears (proximal FPImean % 99.85 ± 0.45 vs 92.73 ± 3.63; distal FPImean % 99.51 ± 0.80 vs 96.35 ± 1.96; P < .001). Thus, large tears resulted in slower velocities through the tears (systolic velocity <180 cm/s) and complex flows within the FL, whereas small tears resulted in lower FL pressures, higher tear velocities (systolic velocity >290 cm/s), and a well-defined flow. Additionally, both proximal and distal tears act as entry and exit. During systole, flow enters the FL through all tears simultaneously, while during diastole, flow leaves through all communications. Flow through the FL, from proximal to distal tears or vice versa, is minimal. Conclusions Our results suggest that FL hemodynamics heavily depends on cumulative tear size, and thus, it is an important parameter to take into account when clinically assessing chronic aortic dissections.
In fetal cardiology, imaging (especially echocardiography) has demonstrated to help in the diagnosis and monitoring of fetuses with a compromised cardiovascular system potentially associated with ...several fetal conditions. Different ultrasound approaches are currently used to evaluate fetal cardiac structure and function, including conventional 2-D imaging and M-mode and tissue Doppler imaging among others. However, assessment of the fetal heart is still challenging mainly due to involuntary movements of the fetus, the small size of the heart, and the lack of expertise in fetal echocardiography of some sonographers. Therefore, the use of new technologies to improve the primary acquired images, to help extract measurements, or to aid in the diagnosis of cardiac abnormalities is of great importance for optimal assessment of the fetal heart. Machine leaning (ML) is a computer science discipline focused on teaching a computer to perform tasks with specific goals without explicitly programming the rules on how to perform this task. In this review we provide a brief overview on the potential of ML techniques to improve the evaluation of fetal cardiac function by optimizing image acquisition and quantification/segmentation, as well as aid in improving the prenatal diagnoses of fetal cardiac remodeling and abnormalities.
Aortic wall stiffness, tear size and location and the presence of abdominal side branches arising from the false lumen (FL) are key properties potentially involved in FL enlargement in chronic aortic ...dissections (ADs). We hypothesize that temporal variations on FL flow patterns, as measured in a cross-section by phase-contrast magnetic resonance imaging (PC-MRI), could be used to infer integrated information on these features. In 33 patients with chronic descending AD, instantaneous flow profiles were quantified in the FL at diaphragm level by PC-MRI. We used a lumped-parameter model to assess the changes in flow profiles induced by wall stiffness, tear size/location, and the presence of abdominal side branches arising from the FL. Four characteristic FL flow patterns were identified in 31/33 patients (94%) based on the direction of flow in systole and diastole: BA = systolic biphasic flow and primarily diastolic antegrade flow (n = 6); BR = systolic biphasic flow and primarily diastolic retrograde flow (n = 14); MA = systolic monophasic flow and primarily diastolic antegrade flow (n = 9); MR = systolic monophasic flow and primarily diastolic retrograde flow (n = 2). In the computational model, the temporal variation of flow directions within the FL was highly dependent on the position of assessment along the aorta. FL flow patterns (especially at the level of the diaphragm) showed their characteristic patterns due to variations in the cumulative size and the spatial distribution of the communicating tears, and the incidence of visceral side branches originating from the FL. Changes in wall stiffness did not change the temporal variation of the flows whereas it importantly determined intraluminal pressures. FL flow patterns implicitly codify morphological information on key determinants of aortic expansion in ADs. This data might be taken into consideration in the imaging protocol to define the predictive value of FL flows.
Aims
We tested the hypothesis that a machine learning (ML) algorithm utilizing both complex echocardiographic data and clinical parameters could be used to phenogroup a heart failure (HF) cohort and ...identify patients with beneficial response to cardiac resynchronization therapy (CRT).
Methods and results
We studied 1106 HF patients from the Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy (MADIT‐CRT) (left ventricular ejection fraction ≤ 30%, QRS ≥ 130 ms, New York Heart Association class ≤ II) randomized to CRT with a defibrillator (CRT‐D, n = 677) or an implantable cardioverter defibrillator (ICD, n = 429). An unsupervised ML algorithm (Multiple Kernel Learning and K‐means clustering) was used to categorize subjects by similarities in clinical parameters, and left ventricular volume and deformation traces at baseline into mutually exclusive groups. The treatment effect of CRT‐D on the primary outcome (all‐cause death or HF event) and on volume response was compared among these groups. Our analysis identified four phenogroups, significantly different in the majority of baseline clinical characteristics, biomarker values, measures of left and right ventricular structure and function and the primary outcome occurrence. Two phenogroups included a higher proportion of known clinical characteristics predictive of CRT response, and were associated with a substantially better treatment effect of CRT‐D on the primary outcome hazard ratio (HR) 0.35; 95% confidence interval (CI) 0.19–0.64; P = 0.0005 and HR 0.36; 95% CI 0.19–0.68; P = 0.001 than observed in the other groups (interaction P = 0.02).
Conclusions
Our results serve as a proof‐of‐concept that, by integrating clinical parameters and full heart cycle imaging data, unsupervised ML can provide a clinically meaningful classification of a phenotypically heterogeneous HF cohort and might aid in optimizing the rate of responders to specific therapies.
•Multiple myocardial velocity patterns from a stress protocol are jointly analyzed.•Unsupervised Multiple Kernel Learning is used to reduce the complexity of the data.•The variability analysis on the ...learnt space unravels healthy/diseased differences.•The joint analysis of multiple patterns notably improves the characterization.
Display omitted
We propose an independent objective method to characterize different patterns of functional responses to stress in the heart failure with preserved ejection fraction (HFPEF) syndrome by combining multiple temporally-aligned myocardial velocity traces at rest and during exercise, together with temporal information on the occurrence of cardiac events (valves openings/closures and atrial activation). The method builds upon multiple kernel learning, a machine learning technique that allows the combination of data of different nature and the reduction of their dimensionality towards a meaningful representation (output space). The learning process is kept unsupervised, to study the variability of the input traces without being conditioned by data labels. To enhance the physiological interpretation of the output space, the variability that it encodes is analyzed in the space of input signals after reconstructing the velocity traces via multiscale kernel regression. The methodology was applied to 2D sequences from a stress echocardiography protocol from 55 subjects (22 healthy, 19 HFPEF and 14 breathless subjects). The results confirm that characterization of the myocardial functional response to stress in the HFPEF syndrome may be improved by the joint analysis of multiple relevant features.
Arrhythmogenic right ventricular (RV) remodeling has been reported in response to regular training, but it remains unclear how exercise intensity affects the presence and extent of such remodeling. ...We aimed to assess the relationship between RV remodeling and exercise load in a long-term endurance training model. Wistar rats were conditioned to run at moderate (MOD; 45 min, 30 cm/s) or intense (INT; 60 min, 60 cm/s) workloads for 16 wk; sedentary rats served as controls. Cardiac remodeling was assessed with standard echocardiographic and tissue Doppler techniques, sensor-tip pressure catheters, and pressure-volume loop analyses. After MOD training, both ventricles similarly dilated (~16%); the RV apical segment deformation, but not the basal segment deformation, was increased apical strain rate (SR): -2.9 ± 0.5 vs. -3.3 ± 0.6 s
, SED vs. MOD. INT training prompted marked RV dilatation (~26%) but did not further dilate the left ventricle (LV). A reduction in both RV segments' deformation in INT rats (apical SR: -3.3 ± 0.6 vs. -3.0 ± 0.4 s
and basal SR: -3.3 ± 0.7 vs. -2.7 ± 0.6 s
, MOD vs. INT) led to decreased global contractile function (maximal rate of rise of LV pressure: 2.53 ± 0.15 vs. 2.17 ± 0.116 mmHg/ms, MOD vs. INT). Echocardiography and hemodynamics consistently pointed to impaired RV diastolic function in INT rats. LV systolic and diastolic functions remained unchanged in all groups. In conclusion, we showed a biphasic, unbalanced RV remodeling response with increasing doses of exercise: physiological adaptation after MOD training turns adverse with INT training, involving disproportionate RV dilatation, decreased contractility, and impaired diastolic function. Our findings support the existence of an exercise load threshold beyond which cardiac remodeling becomes maladaptive.
Exercise promotes left ventricular eccentric hypertrophy with no changes in systolic or diastolic function in healthy rats. Conversely, right ventricular adaptation to physical activity follows a biphasic, dose-dependent, and segmentary pattern. Moderate exercise promotes a mild systolic function enhancement at the right ventricular apex and more intense exercise impairs systolic and diastolic function.
Structural analysis of biological membranes is important for understanding cell and sub-cellular organelle function as well as their interaction with the surrounding environment. Imaging of whole ...cells in three dimension at high spatial resolution remains a significant challenge, particularly for thick cells. Cryo-transmission soft X-ray microscopy (cryo-TXM) has recently gained popularity to image, in 3D, intact thick cells (∼10μm) with details of sub-cellular architecture and organization in near-native state. This paper reports a new tool to segment and quantify structural changes of biological membranes in 3D from cryo-TXM images by tracking an initial 2D contour along the third axis of the microscope, through a multi-scale ridge detection followed by an active contours-based model, with a subsequent refinement along the other two axes. A quantitative metric that assesses the grayscale profiles perpendicular to the membrane surfaces is introduced and shown to be linearly related to the membrane thickness. Our methodology has been validated on synthetic phantoms using realistic microscope properties and structure dimensions, as well as on real cryo-TXM data. Results demonstrate the validity of our algorithms for cryo-TXM data analysis.
Fetal growth restriction (FGR) affects 5% to 10% of newborns and is associated with increased cardiovascular mortality in adulthood. The most commonly accepted hypothesis is that fetal metabolic ...programming leads secondarily to diseases associated with cardiovascular disease, such as obesity, diabetes mellitus, and hypertension. Our main objective was to evaluate the alternative hypothesis that FGR induces primary cardiac changes that persist into childhood.
Within a cohort of fetuses with growth restriction identified in fetal life and followed up into childhood, we randomly selected 80 subjects with FGR and compared them with 120 normally grown fetuses, matched for gender, birth date, and gestational age at birth. Cardiovascular assessment was performed in childhood (mean age of 5 years). Compared with control subjects, children with FGR had a different cardiac shape, with increased transversal diameters and more globular cardiac ventricles. Although left ejection fraction was similar among the study groups, stroke volume was reduced significantly, which was compensated for by an increased heart rate to maintain output in severe FGR. This was associated with subclinical longitudinal systolic dysfunction (decreased myocardial peak velocities) and diastolic changes (increased E/E' ratio and E deceleration time). Children with FGR also had higher blood pressure and increased intima-media thickness. For all parameters evaluated, there was a linear increase with the severity of growth restriction.
These findings suggest that FGR induces primary cardiac and vascular changes that could explain the increased predisposition to cardiovascular disease in adult life. If these results are confirmed, the impact of strategies with beneficial effects on cardiac remodeling should be explored in children with FGR.
Abstract
Cardiovascular research is in an ongoing quest for a superior imaging method to integrate gross-anatomical information with microanatomy, combined with quantifiable parameters of cardiac ...structure. In recent years, synchrotron radiation-based X-ray Phase Contrast Imaging (X-PCI) has been extensively used to characterize soft tissue in detail. The objective was to use X-PCI to comprehensively quantify ischemic remodeling of different myocardial structures, from cell to organ level, in a rat model of myocardial infarction. Myocardial infarction-induced remodeling was recreated in a well-established rodent model. Ex vivo rodent hearts were imaged by propagation based X-PCI using two configurations resulting in 5.8 µm and 0.65 µm effective pixel size images. The acquired datasets were used for a comprehensive assessment of macrostructural changes including the whole heart and vascular tree morphology, and quantification of left ventricular myocardial thickness, mass, volume, and organization. On the meso-scale, tissue characteristics were explored and compared with histopathological methods, while microstructural changes were quantified by segmentation of cardiomyocytes and calculation of cross-sectional areas. Propagation based X-PCI provides detailed visualization and quantification of morphological changes on whole organ, tissue, vascular as well as individual cellular level of the ex vivo heart, with a single, non-destructive 3D imaging modality.