The 'Digital Twin' to enable the vision of precision cardiology.

Providing therapies tailored to each patient is the vision of precision medicine, enabled by the increasing ability to capture extensive data about individual patients. In this position paper, we argue that the second enabling pillar towards this vision is the increasing power of computers and algorithms to learn, reason, and build the 'digital twin' of a patient. Computational models are boosting the capacity to draw diagnosis and prognosis, and future treatments will be tailored not only to current health status and data, but also to an accurate projection of the pathways to restore health by model predictions. The early steps of the digital twin in the area of cardiovascular medicine are reviewed in this article, together with a discussion of the challenges and opportunities ahead. We emphasize the synergies between mechanistic and statistical models in accelerating cardiovascular research and enabling the vision of precision medicine.

Comprehensive data integration-Toward a more personalized assessment of diastolic function.

BACKGROUND AND AIM: The main challenge of assessing diastolic function is the balance between clinical utility, in the sense of usability and time-efficiency, and overall applicability, in the sense of precision for the patient under investigation. In this review, we aim to explore the challenges of integrating data in the assessment of diastolic function and discuss the perspectives of a more comprehensive data integration approach. METHODS: Review of traditional and novel approaches regarding data integration in the assessment of diastolic function. RESULTS: Comprehensive data integration can lead to improved understanding of disease phenotypes and better relation of these phenotypes to underlying pathophysiological processes-which may help affirm diagnostic reasoning, guide treatment options, and reduce limitations related to previously unaddressed confounders. The optimal assessment of diastolic function should ideally integrate all relevant clinical information with all available structural and functional whole cardiac cycle echocardiographic data-envisioning a personalized approach to patient care, a high-reaching future goal in medicine. CONCLUSION: Complete data integration seems to be a long-lasting goal, the way forward in diastology, and machine learning seems to be one of the tools suited for the challenge. With perpetual evidence that traditional approaches to complex problems may not the optimal solution, there is room for a steady and cautious, and inherently very exciting paradigm shift toward novel diagnostic tools and workflows to reach a more personalized, comprehensive, and integrated assessment of cardiac function.

Preponderance of microbial isolates among heart transplantation recipients requiring renal replacement therapy: a propensity score-adjusted analysis.

AIM: To assess the association between renal replacement therapy (RRT) and post-transplant infection incidence. METHODS: This single-center retrospective cohort study included 158 patients who underwent heart transplantation (HTx) in our center from 2008 to 2016, survived beyond the first post-procedural day, and had available microbial data. The patients were dichotomized according to the need for periprocedural RRT. Twenty-seven patients in RRT group had lower preoperative creatinine clearance, greater body mass index, and higher likelihood of having diabetes. Propensity score adjustment was used to account for multiple covariates. The primary outcome measure was the presence of bacteremia in patients with and without the need for RRT. The secondary outcome measures were the presence of microbial isolates from any culture and clinical outcome data. RESULTS: Unadjusted analysis showed that the RRT group had higher incidence of any positive microbial isolate (93% vs 73%; odds ratio [OR] 4.77, 95% confidence interval [CI] 1.01-30.53; P=0.026) and an increased susceptibility to bacteremia (50% vs 22%; OR 3.50, 95% CI 1.28-9.67; P=0.012). Propensity score-adjusted analysis corroborated the between-group difference in positive blood cultures (OR 3.97, 95% CI 1.28-12.32; P=0.017). There was no difference in the incidence of total microbial isolates between the groups (OR 4.55, 95% CI 0.90-23.05; P=0.067). CONCLUSIONS: Patients requiring RRT after HTx had an increased susceptibility to infections via various portals of entry, predominantly due to an increase in blood-borne infections. Understanding the underlying conditions leading to infection-related morbidity is important for infection control and prevention.

Integration of artificial intelligence into clinical patient management: focus on cardiac imaging.

Cardiac imaging is a crucial component in the management of patients with heart disease, and as such it influences multiple, inter-related parts of the clinical workflow: physician-patient contact, image acquisition, image pre- and postprocessing, study reporting, diagnostics and outcome predictions, medical interventions, and, finally, knowledge-building through clinical research. With the gradual and ubiquitous infiltration of artificial intelligence into cardiology, it has become clear that, when used appropriately, it will influence and potentially improve-through automation, standardization and data integration-all components of the clinical workflow. This review aims to present a comprehensive view of full integration of artificial intelligence into the standard clinical patient management-with a focus on cardiac imaging, but applicable to all information handling-and to discuss current barriers that remain to be overcome before its widespread implementation and integration.

Septal curvature as a robust and reproducible marker for basal septal hypertrophy.

BACKGROUND: Basal septal hypertrophy (BSH) is an asymmetric, localized thickening of the upper interventricular septum and constitutes a marker of an early remodelling in patients with hypertension. This morphological trait has been extensively researched because of its prevalence in hypertension, yet its clinical and prognostic value for individual patients remains undetermined. One of the reasons is the lack of a reliable and reproducible metric to quantify the presence and the extent of BSH. This article proposes the use of the curvature of the left ventricular endocardium as a robust feature for BSH characterization, and as an objective criterion to quantify current subjective 'visual assessment' of the presence of sigmoidal septum. The proposed marker, called average septal curvature, is defined as the inverse of the radius adjacent to each point of the endocardial contour along the basal and mid inferoseptal segments of the left ventricle. METHOD: Robustness and reproducibility were assessed on a cohort of 220 patients, including 161 hypertensive patients (32 with BSH) and 59 healthy controls. RESULTS: The results show that compared with the conventionally used wall thickness metrics, the new marker is more reproducible (relative standard deviation of errors of 7 vs. 13%, and 8 vs. 38% for intra-observer and inter-observer variability, respectively) and better correlates to the functional parameters related to BSH, with main difference (absolute rank correlation 0.417 vs. 0.341) in local deformation changes assessed by longitudinal strain. CONCLUSION: Average septal curvature is a more precisely defined and reproducible metric than thickness ratios, it can be fully automated, and better infers the functional remodelling related to hypertension.

Etiology-Discriminative Multimodal Imaging of Left Ventricular Hypertrophy and Synchrotron-Based Assessment of Microstructural Tissue Remodeling.

Background: Distinguishing the etiology of left ventricular hypertrophy (LVH) is clinically relevant due to patient outcomes and management. Easily obtained, echocardiography-based myocardial deformation patterns may improve standard non-invasive phenotyping, however, the relationship between deformation phenotypes and etiology-related, microstructural cardiac remodeling has not been reported. Synchrotron radiation-based X-ray phase-contrast imaging (X-PCI) can provide high resolution, three-dimensional (3D) information on myocardial microstructure. The aim of this pilot study is to apply a multiscale, multimodality protocol in LVH patients undergoing septal myectomy to visualize in vivo and ex vivo myocardial tissue and relate non-invasive LVH imaging phenotypes to the underlying synchrotron-assessed microstructure. Methods and findings: Three patients (P1-3) undergoing septal myectomy were comprehensively studied. Medical history was collected, and patients were imaged with echocardiography/cardiac magnetic resonance prior to the procedure. Myocardial tissue samples obtained during the myectomy were imaged with X-PCI generating high spatial resolution images (0.65 µm) to assess myocyte organization, 3D connective tissue distribution and vasculature remodeling. Etiology-centered non-invasive imaging phenotypes, based on findings of hypertrophy and late gadolinium enhancement (LGE) distribution, and enriched by speckle-tracking and tissue Doppler echocardiography deformation patterns, identified a clear phenotype of hypertensive heart disease (HTN) in P1, and hypertrophic cardiomyopathy (HCM) in P2/P3. X-PCI showed extensive interstitial fibrosis with normal 3D myocyte and collagen organization in P1. In comparison, in P2/P3, X-PCI showed 3D myocyte and collagen disarray, as well as arterial wall hypertrophy with increased perivascular collagen, compatible with sarcomere-mutation HCM in both patients. The results of this pilot study suggest the association of non-invasive deformation phenotypes with etiology-related myocyte and connective tissue matrix disorganization. A larger patient cohort could enable statistical analysis of group characteristics and the assessment of deformation pattern reproducibility. Conclusion: High-resolution, 3D X-PCI provides novel ways to visualize myocardial remodeling in LVH, and illustrates the correspondence of macrostructural and functional non-invasive phenotypes with invasive microstructural phenotypes, suggesting the potential clinical utility of non-invasive myocardial deformation patterns in phenotyping LVH in everyday clinical practice.

Automated Pattern Recognition in Whole-Cardiac Cycle Echocardiographic Data: Capturing Functional Phenotypes with Machine Learning.

BACKGROUND: Echocardiography provides complex data on cardiac function that can be integrated into patterns of dysfunction related to the severity of cardiac disease. The aim of this study was to demonstrate the feasibility of applying machine learning (ML) to automate the integration of echocardiographic data from the whole cardiac cycle and to automatically recognize patterns in velocity profiles and deformation curves, allowing the identification of functional phenotypes. METHODS: Echocardiography was performed in 189 clinically managed patients with hypertension and 97 healthy individuals without hypertension. Speckle-tracking analysis of the left ventricle and atrium was performed, and deformation curves were extracted. Aortic and mitral blood pool pulsed-wave Doppler and mitral annular tissue pulsed-wave Doppler velocity profiles were obtained. These whole-cardiac cycle deformation and velocity curves were used as ML input. Unsupervised ML was used to create a representation of patients with hypertension in a virtual space in which patients are positioned on the basis of the similarity of their integrated whole-cardiac cycle echocardiography data. Regression methods were used to explore patterns of echocardiographic traces within this virtual ML-derived space, while clustering was used to define phenogroups. RESULTS: The algorithm captured different patterns in tissue and blood-pool velocity and deformation profiles and integrated the findings, yielding phenotypes related to normal cardiac function and others to advanced remodeling associated with pressure overload in hypertension. The addition of individuals without hypertension into the ML-derived space confirmed the interpretation of normal and remodeled phenotypes. CONCLUSIONS: ML-based pattern recognition is feasible from echocardiographic data obtained during the whole cardiac cycle. Automated algorithms can consistently capture patterns in velocity and deformation data and, on the basis of these patterns, group patients into interpretable, clinically comprehensive phenogroups that describe structural and functional remodeling. Automated pattern recognition may potentially aid interpretation of imaging data and diagnostic accuracy.

Towards patient-specific prediction of conduction abnormalities induced by transcatheter aortic valve implantation: a combined mechanistic modelling and machine learning approach.

Aims: Post-procedure conduction abnormalities (CA) remain a common complication of transcatheter aortic valve implantation (TAVI), highlighting the need for personalized prediction models. We used machine learning (ML), integrating statistical and mechanistic modelling to provide a patient-specific estimation of the probability of developing CA after TAVI. Methods and results: The cohort consisted of 151 patients with normal conduction and no pacemaker at baseline who underwent TAVI in nine European centres. Devices included CoreValve, Evolut R, Evolut PRO, and Lotus. Preoperative multi-slice computed tomography was performed. Virtual valve implantation with patient-specific computer modelling and simulation (CM&S) allowed calculation of valve-induced contact pressure on the anatomy. The primary composite outcome was new onset left or right bundle branch block or permanent pacemaker implantation (PPI) before discharge. A supervised ML approach was applied with eight models predicting CA based on anatomical, procedural and mechanistic data. CA occurred in 59% of patients (n = 89), more often after mechanical than first or second generation self-expanding valves (68% vs. 60% vs. 41%). CM&S revealed significantly higher contact pressure and contact pressure index in patients with CA. The best model achieved 83% accuracy (area under the curve 0.84) and sensitivity, specificity, positive predictive value, negative predictive value, and F1-score of 100%, 62%, 76%, 100%, and 82%. Conclusion: ML, integrating statistical and mechanistic modelling, achieved an accurate prediction of CA after TAVI. This study demonstrates the potential of a synergetic approach for personalizing procedure planning, allowing selection of the optimal device and implantation strategy, avoiding new CA and/or PPI.

Distribution of myocardial work in arterial hypertension: insights from non-invasive left ventricular pressure-strain relations.

A index of non-invasive myocardial work (MWI) can account for pressure during the assessment of cardiac function, potentially separating the influence of loading conditions from the influence of the underlying tissue remodelling. The aim is to assess LV function accounted for loading and explore hypertensive MWI distribution by comparing healthy individuals to hypertensive patients without and with localized basal septal hypertrophy (BSH). An echocardiogram was performed in 170 hypertensive patients and 20 healthy individuals. BSH was defined by a basal-to-mid septal wall thickness ratio ≥ 1.4. LV speckle-tracking was performed, and the MWI calculated globally and regionally for the apical, mid and basal regions. An apex-to-base gradient, seen in regional strain values, was preserved in the distribution of myocardial work, with the apical region compensating for the impairment of the basal segments. This functional redistribution was further pronounced in patients with localized BSH. In these patients, segmental MWI analysis revealed underlying impairment of regional work unrelated to acute loading conditions. Non-invasive MWI analysis offers the possibility to compare LV function regardless of blood pressure at the time of observation. Changes in MWI distribution can be seen in hypertension unrelated to the load-dependency of strain. Accentuated functional changes affirm the role of BSH as an echocardiographic marker in hypertension.

Variability in the Assessment of Myocardial Strain Patterns: Implications for Adequate Interpretation.

Variability in global and regional peak strain has been thoroughly studied, but variability in spatiotemporal myocardial strain patterns has not been studied as well. This study reports on such variability and its implications for adequate disease interpretation. Forty in-training operators, distributed on 20 workstations, analyzed six cases with representative deformation patterns with commercial speckle tracking. Inter-operator differences were quantified through the variability in myocardial delineations, spatiotemporal longitudinal strain patterns and peak longitudinal strain. Intra-operator differences were assessed similarly using 10 repeated measurements from a single clinician expert. Delineations varied mainly along the lateral wall and at the valve level. Peak longitudinal strain variability was low to moderate. The spatiotemporal strain patterns were consistent despite high variability at the apex and near the valve. The results indicate that relevant pattern assessment is possible despite heterogeneous experience with speckle tracking and that careful interpretation of pattern abnormalities should be recommended before a more systematic quantitative analysis.

Basal Ventricular Septal Hypertrophy in Systemic Hypertension.

Basal septal hypertrophy (BSH) is commonly seen in patients with systemic hypertension and has been associated with increased afterload. The impact of localized hypertrophy on left ventricular (LV) and left atrial (LA) function is still unclear. Our aim is to investigate if BSH is a marker of a more pronounced impact of hypertension on cardiac function in the early stages of hypertensive heart disease. An echocardiogram was performed in 163 well-controlled hypertensive patients and 22 healthy individuals. BSH was defined by a basal-to-mid septal thickness ratio ≥1.4. LV dimensions and mass were evaluated. LV global and regional deformation was assessed by 2-dimensional (2D) speckle tracking echocardiography, and LV diastolic function by 2D and Doppler imaging. LA function was evaluated with phasic volume indices calculated from 2D and 3-dimensional volumes, as well as speckle tracking echocardiography. The population was 54% men, mean age 57 (53 to 60) years. BSH was seen in 20% (n = 32) of the hypertensive cohort. Patients with BSH showed decreased regional LV systolic deformation, impaired LV relaxation with a higher proportion of indeterminate LV diastolic function, and LA functional impairment defined by a reduction of reservoir strain and a change in LA functional dynamics. In conclusion, in well-controlled hypertension impairment of LV and LA function is present in patients with early LV remodeling and localized hypertrophy. BSH might be useful as an early marker of the burden of hypertensive heart disease.