Title | ||
---|---|---|
U-Sleep: A Deep Neural Network for Automated Detection of Sleep Arousals Using Multiple PSGs. |
Abstract | ||
---|---|---|
Sleep disorders can seriously affect human health. Most of the previous studies focused on sleep disorders of apnea, but few on non-apnea. This type of sleep disorder is complex and difficult to detect by traditional methods. In this paper, a physiological time series segmentation network U-Sleep based on deep learning is proposed to analyze these sleep disorders. U-Sleep is a time series convolution network based on U-Net architecture. U-Sleep uses the sequence to sequence input-output mode to map multiple complete original polysomnograms to a single tag sequence. This enables our model to automatically learn the variable interaction between different signals and any related time dependence, and automatically extract such arousal features from the rich physiological time series. We conducted three-fold cross-validation, and use the ensemble model strategy to get the final detection results. Experiments on the datasets of PhysioNet show that the average performance of the final model is: accuracy(0.886), F1(0.892), AUROC(0.916), AUPRC(0.797), SE(0.804), and AI(6.240). |
Year | DOI | Venue |
---|---|---|
2020 | 10.1007/978-3-030-63836-8_52 | ICONIP (3) |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shenglan Yang | 1 | 0 | 0.34 |
Bijue Jia | 2 | 3 | 2.06 |
Yao Chen | 3 | 15 | 2.36 |
Zhan ao Huang | 4 | 0 | 0.34 |
Xiaoming Huang | 5 | 0 | 0.34 |
Jian Cheng Lv | 6 | 337 | 54.52 |