Deep-Learning Premature Contraction Localization in 12-lead ECG From Whole Signal Annotations

Deep-Learning Premature Contraction Localization in 12-lead ECG From Whole Signal Annotations

Paper in proceedings (conference paper)

Since common electrocardiography (ECG) diagnostics approaches are time-consuming and arrhythmia-type sensitive, deep-learning methods are state-of-the-art for their detection accuracy. However, premature ventricular contractions' (PVC) localization via common deep-learning approaches requires large training set, therefore Multiple Instance Learning (MIL) framework was applied, where model is trained from whole-signal annotations. Proposed MIL framework is based on 1D Convolutional Neural Network (CNN), with global max-pooling in the last layer. The detection of PVCs' positions was done by the peak detector with specified parameters - threshold, minimal distance and peak prominence. Our method was tested on database containing 1590 ECGs, including 672 signals with PVCs. Dice coefficient reaches 0.947. This simple deep-learning method for the localization of PVC achieves a promising performance while being trainable from the whole-signal annotations instead of positional labels.

Since common electrocardiography (ECG) diagnostics approaches are time-consuming and arrhythmia-type sensitive, deep-learning methods are state-of-the-art for their detection accuracy. However, premature ventricular contractions' (PVC) localization via common deep-learning approaches requires large training set, therefore Multiple Instance Learning (MIL) framework was applied, where model is trained from whole-signal annotations. Proposed MIL framework is based on 1D Convolutional Neural Network (CNN), with global max-pooling in the last layer. The detection of PVCs' positions was done by the peak detector with specified parameters - threshold, minimal distance and peak prominence. Our method was tested on database containing 1590 ECGs, including 672 signals with PVCs. Dice coefficient reaches 0.947. This simple deep-learning method for the localization of PVC achieves a promising performance while being trainable from the whole-signal annotations instead of positional labels.

ECG, electrocardiogram, arrhythmia, localization, global, annotation, PVC, premature ventricular contractions

ECG, electrocardiogram, arrhythmia, localization, global, annotation, PVC, premature ventricular contractions

NOVOTNÁ, P.; VIČAR, T.; RONZHINA, M.; HEJČ, J.; KOLÁŘOVÁ, J.

2021

30.09.2020

IEEE

NEW YORK

978-1-7281-7382-5

Computing in Cardiology 2020

47

2325-8861

Compuing in Cardiology 2013

Kingdom of Spain

1

4

4

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9344059