Non-linear Analysis Based ECG Classification of Cardiovascular Disorders

TL;DR

This work tackles multi-class ECG-based cardiovascular disorder classification under nonlinear signal conditions by marrying Recurrence plots with autoencoder latent-space embeddings and Recurrence Quantification Analysis. By transforming 15-channel ECG into 224×224 recurrence images, reducing them to a 14×14 latent space, and extracting 10 RQA features, the authors deploy two classifiers—a CNN on latent embeddings and a stacked ensemble on RQA features—achieving 100% and 97.05% accuracy, respectively, on four disorders plus healthy controls using the PTB dataset. The approach is validated with statistical tests (Wilcoxon) and visualizations (t-SNE) showing well-separated clusters, and it outperforms recent state-of-the-art methods on the same data. The results suggest a robust, non-linear framework for rapid and reliable ECG-based CVD screening with potential clinical impact and directions for future enhancements including transformers and graph-based models.

Abstract

Multi-channel ECG-based cardiac disorders detection has an impact on cardiac care and treatment. Limitations of existing methods included variation in ECG waveforms due to the location of electrodes, high non-linearity in the signal, and amplitude measurement in millivolts. The present study reports a non-linear analysis-based methodology that utilizes Recurrence plot visualization. The patterned occurrence of well-defined structures, such as the QRS complex, can be exploited effectively using Recurrence plots. This Recurrence-based method is applied to the publicly available Physikalisch-Technische Bundesanstalt (PTB) dataset from PhysioNet database, where we studied four classes of different cardiac disorders (Myocardial infarction, Bundle branch blocks, Cardiomyopathy, and Dysrhythmia) and healthy controls, achieving an impressive classification accuracy of 100%. Additionally, t-SNE plot visualizations of the latent space embeddings derived from Recurrence plots and Recurrence Quantification Analysis features reveal a clear demarcation between the considered cardiac disorders and healthy individuals, demonstrating the potential of this approach.

Figures (9)

• Figure 1: ECG signal for a representative subjects of different CVDs. The pattern of ECG signal shows clear differences among different CVDs (Figure \ref{['fig:a']}, \ref{['fig:b']}, \ref{['fig:c']}, \ref{['fig:d']}) as compared to healthy control (Figure \ref{['fig:e']})
• Figure 2: Overview of the proposed methodology
• Figure 3: Recurrence plot visualization by converting an Recurrence matrix of size $K\times K$ to uniformly-sized ($224\times224$) Recurrence image
• Figure 4: Autoencoder architecture
• Figure 5: Visualization of autoencoder 2D latent space embedding from Recurrence plots of ECG signal across 15 channels. As seen from this figure, the latent space of the autoencoder showed a clear and distinctive frequency patterns among different CVDs and HC.