ECG Classification on PTB-XL: A Data-Centric Approach with Simplified CNN-VAE

TL;DR

This paper shows that a simplified CNN-VAE model, combined with careful data preprocessing and class balancing, can effectively classify ECG signals from the PTB-XL dataset with high accuracy and fewer parameters.

Contribution

The study demonstrates that data-centric practices and simplified neural network architectures can achieve competitive ECG classification performance, challenging the trend of increasing model complexity.

Findings

Achieved 87.01% binary accuracy on PTB-XL dataset.

Used only 197,093 trainable parameters in the model.

Highlighted the importance of data preprocessing and class balancing.

Abstract

Automated electrocardiogram (ECG) classification is essential for early detection of cardiovascular diseases. While recent approaches have increasingly relied on deep neural networks with complex architectures, we demonstrate that careful data preprocessing, class balancing, and a simplified convolutional neural network combined with a variational autoencoder (CNN-VAE) architecture can achieve competitive performance with significantly reduced model complexity. Using the publicly available PTB XL dataset, we achieve 87.01% binary accuracy and 0.7454 weighted F1-score across five diagnostic classes (CD, HYP, MI, NORM, STTC) with only 197,093 trainable parameters. Our work emphasises the importance of data-centric machine learning practices over architectural complexity, demonstrating that systematic preprocessing and balanced training strategies are critical for medical signal classification. We identify challenges in minority class detection (particularly hypertrophy) and provide insights for future improvements in handling imbalanced ECG datasets. Index Terms: ECG classification, convolutional neural networks, class balancing, data preprocessing, variational autoencoders, PTB-XL dataset