Sleep apnea (SA) is a common sleep-related breathing disorder, which tends to induce a series of complications such as pediatric intracranial hypertension, psoriasis, and even sudden death. Therefore, early diagnosis and treatment can effectively prevent malignant complications. Portable monitoring (PM) is a widely used tool for people to monitor their sleep conditions out of hospital. In this study, we focus on SA detection based on single-lead ECG signals which are easily collected by PM. We propose a bottleneck attention based fusion of deep neural network named BAFNet, which mainly includes five parts of RRI (R-R intervals) stream network, RPA (R-peak amplitudes) stream network, global query generation, feature fusion, and classifier. To learn the feature representation of RRI/RPA segments, fully convolutional networks (FCN) with cross-learning are proposed. Meanwhile, to control the feature information flow in RRI/RPA stream network, a global query generation with bottleneck attention is proposed. To further improve the SA detection performance, a hard sample scheme with k-means clustering is employed. Experiment results show that BAFNet can achieve competitive results, which is superior to the state-of-the-art SA detection methods. It means that BAFNet has a great potential to be applied in home sleep apnea test (HSAT) for sleep conditions monitoring.
- Get the pkl file
- Download the dataset Apnea-ECG Database
- Run Preprocessing.py to get a file named apnea-ecg.pkl
- Per-segment classification
- Run BAFNet.ipynb
- Per-recording classification
- Run evaluate.py
- The performance is shown in Table.csv
If you have any questions, please email to: xhchen@nyu.edu