Detection of Atrial Fibrillation Using 1D Convolutional Neural Network

• Published in: Sensors (Basel, Switzerland) Vol. 20; no. 7; p. 2136
• Main Authors: Hsieh, Chaur-Heh, Li, Yan-Shuo, Hwang, Bor-Jiunn, Hsiao, Ching-Hua
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 10.04.2020; MDPI
• Subjects: Accuracy; Algorithms; atrial fibrillation (AF); Atrial Fibrillation - diagnosis; Cardiac arrhythmia; Classification; convolutional neural network (CNN); Datasets; Deep Learning; electrocardiogram (ECG); Electrocardiography; Heart failure; Humans; Methods; Neural networks; Neural Networks, Computer; Wavelet transforms; atrial fibrillation (AF); deep learning; electrocardiogram (ECG); convolutional neural network (CNN)
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s20072136

The automatic detection of atrial fibrillation (AF) is crucial for its association with the risk of embolic stroke. Most of the existing AF detection methods usually convert 1D time-series electrocardiogram (ECG) signal into 2D spectrogram to train a complex AF detection system, which results in heavy training computation and high implementation cost. This paper proposes an AF detection method based on an end-to-end 1D convolutional neural network (CNN) architecture to raise the detection accuracy and reduce network complexity. By investigating the impact of major components of a convolutional block on detection accuracy and using grid search to obtain optimal hyperparameters of the CNN, we develop a simple, yet effective 1D CNN. Since the dataset provided by PhysioNet Challenge 2017 contains ECG recordings with different lengths, we also propose a length normalization algorithm to generate equal-length records to meet the requirement of CNN. Experimental results and analysis indicate that our method of 1D CNN achieves an average F1 score of 78.2%, which has better detection accuracy with lower network complexity, as compared with the existing deep learning-based methods.