Leveraging Liquid Time-Constant Neural Networks for ECG Classification: A Focus on Pre-Processing Techniques

Lisa-Maria Beneke, Michell Boerger, Philipp Lämmel, Helene Knof, Andrei Aleksandrov, Nikolay Tcholtchev, Nikolay Tcholtchev

2025

Abstract

Neural networks have become pivotal in timeseries classification due to their ability to capture complex temporal relationships. This paper presents an evaluation of Liquid Time-Constant Neural Networks (LTCs), a novel approach inspired by recurrent neural networks (RNNs) that introduces a unique mechanism to adaptively manage temporal dynamics through time-constant parameters. Specifically, we explore the applicability and effectiveness of LTC in the context of classifying myocardial infarctions in electrocardiogram data by benchmarking the performance of LTCs against RNN and LSTM models utilzing the PTB-XL dataset. Moreover, our study focuses on analyzing the impact of various pre-processing methods, including baseline wander removal, Fourier transformation, Butterworth filtering, and a novel x-scaling method, on the efficacy of these models. The findings provide insights into the strengths and limitations of LTCs, enhancing the understanding of their applicability in multivariate time series classification tasks.

Paper Citation

in Harvard Style

Beneke L., Boerger M., Lämmel P., Knof H., Aleksandrov A. and Tcholtchev N. (2025). Leveraging Liquid Time-Constant Neural Networks for ECG Classification: A Focus on Pre-Processing Techniques. In Proceedings of the 14th International Conference on Data Science, Technology and Applications - Volume 1: DATA; ISBN 978-989-758-758-0, SciTePress, pages 234-245. DOI: 10.5220/0013648000003967

in Bibtex Style

@conference{data25,
author={Lisa-Maria Beneke and Michell Boerger and Philipp Lämmel and Helene Knof and Andrei Aleksandrov and Nikolay Tcholtchev},
title={Leveraging Liquid Time-Constant Neural Networks for ECG Classification: A Focus on Pre-Processing Techniques},
booktitle={Proceedings of the 14th International Conference on Data Science, Technology and Applications - Volume 1: DATA},
year={2025},
pages={234-245},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013648000003967},
isbn={978-989-758-758-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Data Science, Technology and Applications - Volume 1: DATA
TI - Leveraging Liquid Time-Constant Neural Networks for ECG Classification: A Focus on Pre-Processing Techniques
SN - 978-989-758-758-0
AU - Beneke L.
AU - Boerger M.
AU - Lämmel P.
AU - Knof H.
AU - Aleksandrov A.
AU - Tcholtchev N.
PY - 2025
SP - 234
EP - 245
DO - 10.5220/0013648000003967
PB - SciTePress