Deep Transfer Learning for Single-Channel Automatic Sleep Staging with Channel Mismatch

被引:25
作者
Huy Phan [1 ]
Chen, Oliver Y. [2 ]
Koch, Philipp [3 ]
Mertins, Alfred [3 ]
De Vos, Maarten [2 ]
机构
[1] Univ Kent, Sch Comp, Canterbury, Kent, England
[2] Univ Oxford, Inst Biomed Engn, Oxford, England
[3] Univ Lubeck, Inst Signal Proc, Lubeck, Germany
来源
2019 27TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO) | 2019年
关键词
Automatic sleep staging; deep learning; transfer learning; SeqSleepNet; RESOURCE;
D O I
10.23919/eusipco.2019.8902977
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Many sleep studies suffer from the problem of insufficient data to fully utilize deep neural networks as different labs use different recordings set ups, leading to the need of training automated algorithms on rather small databases, whereas large annotated databases are around but cannot be directly included into these studies for data compensation due to channel mismatch. This work presents a deep transfer learning approach to overcome the channel mismatch problem and transfer knowledge from a large dataset to a small cohort to study automatic sleep staging with single-channel input. We employ the state-of-the-art SeqSleepNet and train the network in the source domain, i.e. the large dataset. Afterwards, the pretrained network is finetuned in the target domain, i.e. the small cohort, to complete knowledge transfer. We study two transfer learning scenarios with slight and heavy channel mismatch between the source and target domains. We also investigate whether, and if so, how finetuning entirely or partially the pretrained network would affect the performance of sleep staging on the target domain. Using the Montreal Archive of Sleep Studies (MASS) database consisting of 200 subjects as the source domain and the Sleep-EDF Expanded database consisting of 20 subjects as the target domain in this study, our experimental results show significant performance improvement on sleep staging achieved with the proposed deep transfer learning approach. Furthermore, these results also reveal the essential of finetuning the feature-learning parts of the pretrained network to be able to bypass the channel mismatch problem.
引用
收藏
页数:5
相关论文
共 24 条
[1]  
Abadi M., 2015, TENSORFLOW LARGE SCA, DOI DOI 10.48550/ARXIV.1603.04467
[2]  
Andreotti F, 2018, IEEE ENG MED BIO, P171, DOI 10.1109/EMBC.2018.8512214
[3]  
[Anonymous], 2017, IEEE T NEUR SYS REH, DOI [DOI 10.1109/TNSRE.2017.2721116, 10.1109/TNSRE.2017.2721116]
[4]  
[Anonymous], 2016, ARXIV161001683
[5]  
Cho Kyunghyun, 2014, C EMPIRICAL METHODS, P1724
[6]  
Cooray N., 2018, ARXIV181104662
[7]   PhysioBank, PhysioToolkit, and PhysioNet - Components of a new research resource for complex physiologic signals [J].
Goldberger, AL ;
Amaral, LAN ;
Glass, L ;
Hausdorff, JM ;
Ivanov, PC ;
Mark, RG ;
Mietus, JE ;
Moody, GB ;
Peng, CK ;
Stanley, HE .
CIRCULATION, 2000, 101 (23) :E215-E220
[8]  
Hobson J. A., 1969, Public Health Service Publication, V26, P644
[9]   SeqSleepNet: End-to-End Hierarchical Recurrent Neural Network for Sequence-to-Sequence Automatic Sleep Staging [J].
Huy Phan ;
Andreotti, Fernando ;
Cooray, Navin ;
Chen, Oliver Y. ;
De Vos, Maarten .
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2019, 27 (03) :400-410
[10]  
Phan H, 2018, IEEE ENG MED BIO, P453, DOI 10.1109/EMBC.2018.8512286