Enhancing AI-Based ECG Delineation with Deep Learning Denoising Techniques

TL;DR

This paper introduces an autoencoder-based deep learning method to denoise noisy canine ECG signals, improving the accuracy of ECG delineation by effectively removing diverse noise while preserving important features.

Contribution

The study presents a novel neural network model and training strategy specifically designed for ECG denoising, enhancing robustness and accuracy in canine cardiac signal analysis.

Findings

Effective noise reduction without degrading ECG features

Robust performance across noisy and clean ECG recordings

Improved ECG delineation accuracy

Abstract

Evaluating canine electrocardiograms (ECGs) is challenging due to noise that can obscure clinically relevant cardiac electrical activity. Common sources of interference include respiration, muscle activity, poor lead contact, and external electrical artifacts. Classical signal denoising techniques, such as filtering and wavelet-based methods, struggle to suppress diverse noise patterns while preserving morphological features critical for accurate ECG delineation. We propose an autoencoder-based neural network model and training strategy for ECG denoising as a preprocessing step for canine ECG analysis. The model is trained to reconstruct clean cardiac signals from noisy inputs, enabling effective noise reduction without degrading diagnostically important waveforms. Our approach demonstrates strong performance across both noisy and clean ECG recordings, indicating robustness to varying signal conditions and suitability for downstream delineation tasks.