Automatic screening method for atrial fibrillation based on lossy compression of the electrocardiogram signal.

Researchers

Journal

Modalities

Models

Abstract

Compressed sensing is a low-complexity compression technology that has recently been proposed. It can be applied to long-term electrocardiogram (ECG) monitoring using wearable devices. In this study, an automatic screening method for atrial fibrillation based on lossy compression of the electrocardiogram signal is proposed.
The proposed method combines the compressed sensing with the convolutional neural network. The sparse binary sensing matrix is first used to project the raw ECG signal randomly, transformed the raw ECG data from high-dimensional space to low-dimensional space to complete compression, and then uses CNN to classify the compressed ECG signal involving AF. Our proposed model is validated on the MIT-BIH Atrial Fibrillation Database.
The experimental results show that the model only needs about 1s to complete the 24-hour ECG recording of AF, which is 3.41%, 69.84% and 67.56% less than the time required by AlexNet, VGGNet and GoogLeNet. Under different compression ratios of 10% to 90%, the maximum and minimum F1 scores reach 96.25% and 88.17%, respectively.
The CS-CNN model has high computational efficiency while ensuring prediction accuracy, and is a promising method for AF screening in wearable application scenarios.
© 2020 Institute of Physics and Engineering in Medicine.