Improving the clinical understanding of hypertrophic cardiomyopathy by combining patient data, machine learning and computer simulations: A case study.

Most patients with hypertrophic cardiomyopathy (HCM), the most common genetic cardiac disease, remain asymptomatic, but others may suffer from sudden cardiac death. A better identification of those patients at risk, together with a better understanding of the mechanisms leading to arrhythmia, are crucial to target high-risk patients and provide them with appropriate treatment. However, this currently remains a challenge. In this paper, we present a successful example of implementing computational techniques for clinically-relevant applications. By combining electrocardiogram and imaging data, machine learning and high performance computing simulations, we identified four phenotypes in HCM, with differences in arrhythmic risk, and provided two distinct possible mechanisms that may explain the heterogeneity of HCM manifestation. This led to a better HCM patient stratification and understanding of the underlying disease mechanisms, providing a step further towards tailored HCM patient management and treatment.

ECG-based estimation of dispersion of APD restitution as a tool to stratify sotalol-induced arrhythmic risk.

BACKGROUND: Increased spatial dispersion of restitution properties has been associated to arrhythmic risk. An ECG-based index quantifying restitution dispersion, DRest, is evaluated in patients who experienced Torsades de Pointes (TdP) under sotalol challenge and compared with the response in healthy subjects. METHODS AND RESULTS: ECG recordings were analyzed for quantification of DRest and QTc, among others biomarkers. DRest provides improved discrimination following sotalol administration between TdP and healthy subjects ([min-max]: [0.18-0.22] vs [0.02-0.12]), compared to other biomarkers including QTc ([436-548ms] vs [376-467ms]). Results in healthy subjects are in agreement with simulations of sotalol effects on a human tissue electrophysiological model. CONCLUSIONS: This case study supports the potential of DRest for improved arrhythmia risk stratification even with QTc values below 450ms.

Modeling and quantification of repolarization feature dependency on heart rate.

INTRODUCTION: This article is part of the Focus Theme of Methods of Information in Medicine on "Biosignal Interpretation: Advanced Methods for Studying Cardiovascular and Respiratory Systems". OBJECTIVES: This work aims at providing an efficient method to estimate the parameters of a non linear model including memory, previously proposed to characterize rate adaptation of repolarization indices. METHODS: The physiological restrictions on the model parameters have been included in the cost function in such a way that unconstrained optimization techniques such as descent optimization methods can be used for parameter estimation. The proposed method has been evaluated on electrocardiogram (ECG) recordings of healthy subjects performing a tilt test, where rate adaptation of QT and Tpeak-to-Tend (Tpe) intervals has been characterized. RESULTS: The proposed strategy results in an efficient methodology to characterize rate adaptation of repolarization features, improving the convergence time with respect to previous strategies. Moreover, Tpe interval adapts faster to changes in heart rate than the QT interval. CONCLUSIONS: In this work an efficient estimation of the parameters of a model aimed at characterizing rate adaptation of repolarization features has been proposed. The Tpe interval has been shown to be rate related and with a shorter memory lag than the QT interval.

Assessing real-time RR-QT frequency-domain measures of coupling and causality through inhomogeneous point-process bivariate models.

Ventricular repolarization instability is known to be related to arrhythmogenesis and increased risk of sudden cardiac death. These repolarization dynamics are linked to the distance between T-wave and Q-wave occurrences (QT) on the ECG, and they are coupled with R-wave to R-wave interval variability (RRV). Several efforts have been dedicated to the analysis of QT-RR interactions in order to provide both a quantification of the coupling and estimates of intrinsic repolarization dynamics. However, a methodology able to quantify dynamic changes in repolarization variability unrelated to RRV dynamics is still needed. In this study, we propose a bivariate model embedded within a multiple inhomogeneous point-process framework to obtain time-varying tracking of (causal) interactions between QT variability (QTV), a marker of repolarization variability, and RRV. Data from 15 healthy subjects undergoing a tilt table test were analyzed. Our results demonstrate that the model effectively captures the time-varying mutual QTV-RRV interactions. The analysis of time-varying coherence confirms that head-up tilt is associated with a decrease in linear QTV-RRV coupling, while time-varying directed coherence shows that intrinsic QTV becomes more prominent during head-up tilt.