Symmetric Projection Attractor Reconstruction: Sex Differences in the ECG

Abstract

Background: The electrocardiogram (ECG) is a key tool in patient management. Automated ECG analysis supports clinical decision-making, but traditional fiducial point identification discards much of the time-series data that captures the morphology of the whole waveform. Our Symmetric Projection Attractor Reconstruction (SPAR) method uses all the available data to provide a new visualization and quantification of the morphology and variability of any approximately periodic signal. We therefore applied SPAR to ECG signals to ascertain whether this more detailed investigation of ECG morphology adds clinical value.Methods: Our aim was to demonstrate the accuracy of the SPAR method in discriminating between two biologically distinct groups. As sex has been shown to influence the waveform appearance, we investigated sex differences in normal sinus rhythm ECGs. We applied the SPAR method to 9,007 10 second 12-lead ECG recordings from Physionet, which comprised; Dataset 1: 104 subjects (40% female), Dataset 2: 8,903 subjects (54% female).Results: SPAR showed clear visual differences between female and male ECGs (Dataset 1). A stacked machine learning model achieved a cross-validation sex classification accuracy of 86.3% (Dataset 2) and an unseen test accuracy of 91.3% (Dataset 1). The mid-precordial leads performed best in classification individually, but the highest overall accuracy was achieved with all 12 leads. Classification accuracy was highest for young adults and declined with older age.Conclusions: SPAR allows quantification of the morphology of the ECG without the need to identify conventional fiducial points, whilst utilizing of all the data reduces inadvertent bias. By intuitively re-visualizing signal morphology as two-dimensional images, SPAR accurately discriminated ECG sex differences in a small dataset. We extended the approach to a machine learning classification of sex for a larger dataset, and showed that the SPAR method provided a means of visualizing the similarities of subjects given the same classification. This proof-of-concept study therefore provided an implementation of SPAR using existing data and showed that subtle differences in the ECG can be amplified by the attractor. SPAR's supplementary analysis of ECG morphology may enhance conventional automated analysis in clinically important datasets, and improve patient stratification and risk management.

Detecting paroxysmal atrial fibrillation from normal sinus rhythm in equine athletes using Symmetric Projection Attractor Reconstruction and machine learning

“…The r density distribution can be further split into 2 regions, the high-density core at the center of the attractor and the lower-density attractor “arms.” Further details on the construction of a density profile can be found in Aston and colleagues 21 and Lyle and colleagues. 26 …”

“…Overall, the restitution and complexity studies used less than the 3 minutes (3 60-second strips) of data applied in this study, but SPAR was applied to individual 20-second sub-strips, which would be too short for the alternative techniques, and still achieved a ROC AUC of 0.94 as separate records. Our human ECG studies have typically used 10-second diagnostic ECGs, 25 , 26 , 39 and extension of this study should evaluate a similar approach. We also propose that we should build on our ensemble machine learning approach using the 3 classifiers ( k -NN, linear SVM, and SVM with RBF kernel) by incorporating features drawn from the 3 techniques (SPAR, restitution, and complexity) to assess their complementary attributes in the classification of PAF.…”

“…Recent studies applying SPAR analysis to ECG signals include the detection of a genetic mutation associated with increased risk of arrhythmias in murine ECG, 24 the capture of changes induced by dofetilide in human ECG, 25 and the discrimination of sex in human ECG. 26 …”

Sex Recognition through ECG Signals aiming toward Smartphone Authentication

“…It is also noteworthy that the contributions of Attia et al [2], Siegersma et al [4], and Lyle et al [47] reached accuracies of 90.4%, 92.2%, and [91.3%, 86.3%], respectively. In Table 1, more indicators can be found regarding the mentioned work in this section.…”

Cardiovascular waveforms - can we extract more from routine signals?

“…In addition to beat-to-beat variability, increasing evidence highlights the importance of appraising ECG morphology beyond routine interval and amplitude measures. Deep learning data science approaches have demonstrated it is possible to classify between biologically distinct groups (including age and sex) based on utilising all the ECG waveform data, 22,23 and errors in classification may be associated with an increased mortality risk. 24 The appraisal of T wave morphology has also garnered particular interest, for example the T-peak to T-end (Tpe) interval is a metric of the spatial dispersion of repolarization in the left ventricle which, if prolonged, can increase vulnerability to arrythmias, providing a marker of arrhythmic risk and mortality.…”