English

Deep Learning for ECG Segmentation

Signal Processing 2020-01-15 v1 Machine Learning Quantitative Methods Machine Learning

Abstract

We propose an algorithm for electrocardiogram (ECG) segmentation using a UNet-like full-convolutional neural network. The algorithm receives an arbitrary sampling rate ECG signal as an input, and gives a list of onsets and offsets of P and T waves and QRS complexes as output. Our method of segmentation differs from others in speed, a small number of parameters and a good generalization: it is adaptive to different sampling rates and it is generalized to various types of ECG monitors. The proposed approach is superior to other state-of-the-art segmentation methods in terms of quality. In particular, F1-measures for detection of onsets and offsets of P and T waves and for QRS-complexes are at least 97.8%, 99.5%, and 99.9%, respectively.

Keywords

Cite

@article{arxiv.2001.04689,
  title  = {Deep Learning for ECG Segmentation},
  author = {Viktor Moskalenko and Nikolai Zolotykh and Grigory Osipov},
  journal= {arXiv preprint arXiv:2001.04689},
  year   = {2020}
}

Comments

10 pages, 7 figures

R2 v1 2026-06-23T13:10:35.600Z