BRAVEHEART: Open-source software for automated electrocardiographic and vectorcardiographic analysis.

BACKGROUND AND OBJECTIVES: Electrocardiographic (ECG) and vectorcardiographic (VCG) analyses are used to diagnose current cardiovascular disease and for risk stratification for future adverse cardiovascular events. With increasing use of digital ECGs, research into novel ECG/VCG parameters has increased, but widespread computer-based ECG/VCG analysis is limited because there are no currently available, open-source, and easily customizable software packages designed for automated and reproducible analysis. METHODS AND RESULTS: We present BRAVEHEART, an open-source, modular, customizable, and easy to use software package implemented in the MATLAB programming language, for scientific analysis of standard 12-lead ECGs acquired in a digital format. BRAVEHEART accepts a wide variety of digital ECG formats and provides complete and automatic ECG/VCG processing with signal denoising to remove high- and low-frequency artifact, non-dominant beat identification and removal, accurate fiducial point annotation, VCG construction, median beat construction, customizable measurements on median beats, and output of measurements and results in numeric and graphical formats. CONCLUSIONS: The BRAVEHEART software package provides easily customizable scientific analysis of ECGs and VCGs. We hope that making BRAVEHART available will allow other researchers to further the field of ECG/VCG analysis without having to spend significant time and resources developing their own ECG/VCG analysis software and will improve the reproducibility of future studies. Source code, compiled executables, and a detailed user guide can be found at http://github.com/BIVectors/BRAVEHEART. The source code is distributed under the GNU General Public License version 3.

The spatial ventricular gradient is associated with adverse outcomes in acute pulmonary embolism.

BACKGROUND: The spatial ventricular gradient (SVG) is a vectorcardiographic measurement that reflects cardiac loading conditions via electromechanical coupling. OBJECTIVES: We hypothesized that the SVG is correlated with right ventricular (RV) strain and is prognostic of adverse events in patients with acute pulmonary embolism (PE). METHODS: Retrospective, single-center study of patients with acute PE. Electrocardiogram (ECG), imaging, and outcome data were obtained. SVG components were regressed on tricuspid annular plane systolic excursion (TAPSE), qualitative RV dysfunction, and RV/left ventricular (LV) ratio. Odds of adverse outcomes (30-day mortality, vasopressor requirement, or advanced therapy) after PE were regressed on demographics, RV/LV ratios, traditional ECG signs of RV dysfunction, and SVG components using a logit model. RESULTS: ECGs from 317 patients (48% male, age 63.1 ± 16.6 years) with acute PE were analyzed; 36 patients (11.4%) experienced an adverse event. Worse RV hypokinesis, larger RV/LV ratio, and smaller TAPSE were associated with smaller SVG X and Y components, larger SVG Z components, and smaller SVG vector magnitude (p < .001 for all). In multivariable logistic regression, odds of adverse events after PE decreased with increasing SVG magnitude and TAPSE (OR 0.32 and 0.54 per standard deviation increase; p = .03 and p = .004, respectively). Receiver operating characteristic (ROC) analysis showed that, when combined with imaging, replacing traditional ECG criteria with the SVG significantly improved the area under the ROC from 0.70 to 0.77 (p = .01). CONCLUSION: The SVG is correlated with RV dysfunction and adverse outcomes in acute PE and has a better prognostic value than traditional ECG markers.

The Spatial Ventricular Gradient Is an Independent Predictor of Anthracycline-Associated Cardiotoxicity.

Background: Anthracyclines are effective chemotherapies that are limited by cardiotoxicity. The spatial ventricular gradient (SVG) is a marker of electrical heterogeneity linked to adverse cardiovascular outcomes, including sudden cardiac death and heart failure (HF). Objectives: The purpose of this study was to assess if SVG values before chemotherapy are associated with the risk of anthracycline-associated HF or cardiomyopathy (CM). Methods: We analyzed 12-lead electrocardiograms obtained within 6 months before initiation of anthracyclines in a retrospective cohort treated for cancer between 1992 and 2019 at a single academic medical center. Incident HF and CM were defined by ICD-9/10 codes and confirmed by chart review. Vectorcardiograms were constructed from baseline electrocardiograms, and the SVG was calculated. The cumulative incidence of anthracycline-associated HF or CM was regressed on SVG vector orientation and magnitude with death as a competing risk. Results: In 889 patients (47% male; mean age 58 ± 16 years; 71% hematologic malignancies), larger SVG magnitude prechemotherapy was associated with decreased risk of HF or CM after multivariable adjustment, with a subhazard ratio of 0.76 per 1 SD increase (95% CI: 0.59-0.96; P = 0.024). SVG vector orientation, specifically a more leftward oriented VGx, was associated with decreased risk of HF or CM with a subhazard ratio of 0.77 per 1 SD increase (95% CI: 0.61-0.96; P = 0.023). Conclusions: Larger SVG magnitude and more leftward SVG orientation were associated with a decreased risk of anthracycline cardiotoxicity in a large retrospective cohort. Improved cardiac risk stratification algorithms incorporating the SVG could personalize both cancer and cardioprotective therapy.

ECGAug: A novel method of generating augmented annotated electrocardiogram QRST complexes and rhythm strips.

Applications of neural networks (NNs) in medicine have increased dramatically in recent years. In order to train a NN that performs ECG segmentation, it can be very time consuming, or even completely prohibitive, to manually annotate fiducial points on enough QRST complexes to reach a high level of performance. Existing methods for time series data augmentation risk creating non-physiological ECG signals that may hamper NN training, and are unable to provide accurate fiducial point locations in the augmented data. We therefore developed ECGAug, a new method which generates an augmented training set of QRST signals (single beats or rhythm strips) with accurate fiducial point annotations. Our algorithm recombines a library of existing, annotated QRS complexes and T waves in physiologic ways, and then performs additional physiological transformations to generate a set of new annotated QRST complexes or rhythm strips to be used for NN training or validation of ECG annotation algorithms. In experiments where we trained NNs to annotate QRST complexes with a limited training dataset, QRST complexes added to the training dataset by ECGAug significantly improved NN performance. We present the ECGAug process, demonstrate its efficacy, and provide links for downloading the open source ECGAug software.

Changes in global electrical heterogeneity associated with dofetilide, quinidine, ranolazine, and verapamil.

BACKGROUND: Electrocardiographic (ECG) markers of antiarrhythmic drug (AAD) activity could be used to optimize efficacy and minimize toxicity. Vectorcardiographic global electrical heterogeneity (GEH) is associated with ventricular arrhythmias and sudden death, but it is unclear how GEH measurements change in response to AADs. OBJECTIVE: The purpose of this study was to characterize acute effects of AADs on GEH measurements. METHODS: We analyzed double-blind placebo-controlled trial data from healthy volunteers given 1 dose of placebo, dofetilide, quinidine, ranolazine, or verapamil on subsequent visits. Serial ECGs and plasma drug concentrations were collected. Vectorcardiographic GEH parameters (spatial ventricular gradient [SVG], spatial QRST angle, sum absolute QRST integral, and SVG-QRS peak angle) were measured. Placebo-corrected change from baseline was regressed on drug concentration stratified by sex using linear mixed effects models. RESULTS: Among 22 persons (11 (50%) male median age 27 ± 5 years), 5232 ECGs were analyzed. Dofetilide and quinidine were associated with significant changes in more GEH parameters (5) compared with verapamil (2) and ranolazine (1). The most notable change occurred in SVG azimuth, with largest changes (degrees per unit normalized drug concentration) in dofetilide (6.1; 95% confidence interval [CI] 4.2-8.0) and quinidine (9.4; 95% CI 6.7-12.0), and smaller effects in verapamil (4.4; 95% CI 2.9-5.9) and ranolazine (5.4; 95% CI 3.5-7.3). AAD-induced GEH changes significantly differed in men and women. CONCLUSION: AADs change GEH measurements. These changes, which differ by sex, are likely driven by alterations in ion channel function and dispersion of depolarization or repolarization. GEH measurement may allow early assessment of favorable or adverse AAD effects.