Back

Detection of Atrial Fibrillation with a Hybrid Deep Learning Model and Time-Frequency Representations

Luo, Y.; Huang, B.; Zhu, J.; Zeng, X.; Zhang, Q.

2025-06-04 cardiovascular medicine
10.1101/2025.06.03.25328929 medRxiv
Show abstract

Atrial fibrillation (AF), a common cardiac arrhythmia, can lead to severe complications, emphasizing the urgent need for effective detection methods. This study proposes an automated algorithm for AF detection that combines time-frequency analysis with deep learning techniques, achieving exceptional performance across multiple public ECG datasets. The proposed system applies variational mode decomposition (VMD) to decompose non-stationary ECG signals, followed by Hilbert transform (HT) to generate time-frequency representations. These 2D maps are then processed by a deep learning model for classification. We introduce a novel architecture, SwinMobileNet, which integrates the strengths of the Swin Transformer and MobileNetV2 to effectively model both spatial and temporal features in ECG signals. An adaptive attention mechanism ensures efficient and accurate classification. The implementation of this algorithm is available at: https://anonymous.4open.science/r/SwinMobileNet1-0C3C.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
IEEE Journal of Biomedical and Health Informatics
● 37 papers in training set
Top 0.1%
18.1%
Similar papers in this journal
2
IEEE Access
● 35 papers in training set
Top 0.1%
12.4%
Similar papers in this journal
3
Biomedical Signal Processing and Control
○ 22 papers in training set
Top 0.1%
9.6%
Similar papers in this journal
4
IEEE Transactions on Biomedical Engineering
● 40 papers in training set
Top 0.1%
7.1%
Similar papers in this journal
5
PLOS ONE
● 5266 papers in training set
Top 24%
6.6%
Similar papers in this journal
50% of probability mass above
6
Computer Methods and Programs in Biomedicine
○ 28 papers in training set
Top 0.1%
6.1%
7
Artificial Intelligence in Medicine
○ 17 papers in training set
Top 0.2%
3.4%
8
Scientific Reports
○ 3612 papers in training set
Top 34%
3.2%
9
Computers in Biology and Medicine
○ 128 papers in training set
Top 2%
2.7%
10
Medical Image Analysis
○ 35 papers in training set
Top 0.3%
2.4%
11
Patterns
○ 78 papers in training set
Top 0.9%
2.3%
12
Bioengineering
○ 29 papers in training set
Top 0.4%
1.7%
13
PLOS Digital Health
● 106 papers in training set
Top 3%
1.7%
14
Science Bulletin
○ 21 papers in training set
Top 0.2%
1.7%
15
European Heart Journal - Digital Health
◐ 18 papers in training set
Top 0.5%
1.7%
16
PLOS Computational Biology
● 1863 papers in training set
Top 17%
1.1%
17
iScience
○ 1154 papers in training set
Top 28%
1.1%
18
IEEE Transactions on Neural Systems and Rehabilitation Engineering
● 49 papers in training set
Top 0.8%
1.1%
19
Physiological Measurement
○ 14 papers in training set
Top 0.3%
1.1%
20
Frontiers in Physiology
○ 106 papers in training set
Top 2%
1.1%
21
Mathematical Biosciences and Engineering
● 23 papers in training set
Top 0.9%
0.8%
22
Methods
○ 34 papers in training set
Top 0.7%
0.8%
23
Frontiers in Artificial Intelligence
○ 20 papers in training set
Top 0.8%
0.8%
24
Nature Communications
○ 5641 papers in training set
Top 57%
0.8%
25
Biomedical Optics Express
● 95 papers in training set
Top 1%
0.6%
26
BMC Medical Informatics and Decision Making
○ 43 papers in training set
Top 2%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.