Machine learning-based detection of cardiovascular disease using ECG signals: performance vs. complexity
Frontiers in Cardiovascular Medicine · 31.07.2023 · 10
DOI: 10.3389/fcvm.2023.1229743 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Cardiovascular Medicine, online erschienen am 31.07.2023. Verfasst von Huy Pham und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, HSE University, Moscow, Russia. Die Arbeit wurde bisher 28-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Cardiovascular disease remains a significant problem in modern society. Among non-invasive techniques, the electrocardiogram (ECG) is one of the most reliable methods for detecting cardiac abnormalities. However, ECG interpretation requires expert knowledge and it is time-consuming. Developing a novel method to detect the disease early improves the quality and efficiency of medical care.
Methods
The paper presents various modern approaches for classifying cardiac diseases from ECG recordings. The first approach suggests the Poincaré representation of ECG signal and deep-learning-based image classifiers. Additionally, the raw signals were processed with the one-dimensional convolutional model while the XGBoost model was facilitated to predict based on the time-series features.
Results
The Poincaré-based methods showed decent performance in predicting AF (atrial fibrillation) but not other types of arrhythmia. XGBoost model gave an acceptable performance in long-term data but had a long inference time due to highly-consuming calculations within the pre-processing phase. Finally, the 1D convolutional model, specifically the 1D ResNet, showed the best results in both studied CinC 2017 and CinC 2020 datasets, reaching the F1 score of 85% and 71%, respectively, and they were superior to the first-ranking solution of each challenge. The 1D models also presented high specificity. Additionally, our paper investigated efficiency metrics including power consumption and equivalent CO 2 emissions, with one-dimensional models like 1D CNN and 1D ResNet being the most energy efficient. Model interpretation analysis showed that the DenseNet detected AF using heart rate variability while the 1D ResNet assessed the AF patterns in raw ECG signals.
Discussion
Despite the under-performed results, the Poincaré diagrams are still worth studying further because of the accessibility and inexpensive procedure. In the 1D convolutional models, the residual connections are useful to keep the model simple but not decrease the performance. Our approach in power measurement and model interpretation helped understand the numerical complexity and mechanism behind the model decision.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Computer Science, HSE University, Moscow, Russia.
- AI for Medicine, Sber AI Lab, Moscow, Russia.
- Applied Research Center, Sber AI Lab, Moscow, Russia.
- New Materials Discovery Group, Artificial Intelligence Research Institute (AIRI), Moscow, Russia.
Zitieren
APA
Pham, H., Egorov, K., Kazakov, A., & Budennyy, S. (2023). Machine learning-based detection of cardiovascular disease using ECG signals: performance vs. complexity. Frontiers in Cardiovascular Medicine, 10. https://doi.org/10.3389/fcvm.2023.1229743
Vancouver
Pham H, Egorov K, Kazakov A, Budennyy S. Machine learning-based detection of cardiovascular disease using ECG signals: performance vs. complexity. Frontiers in Cardiovascular Medicine. 2023;10. doi:10.3389/fcvm.2023.1229743
BibTeX
@article{Pham2023,
author = {Pham, Huy and Egorov, Konstantin and Kazakov, Alexey and Budennyy, Semen},
title = {Machine learning-based detection of cardiovascular disease using ECG signals: performance vs. complexity},
journal = {Frontiers in Cardiovascular Medicine},
year = {2023},
volume = {10},
doi = {10.3389/fcvm.2023.1229743}
}
EUVIMED