Self-Supervised Representation Learning-Based OSA Detection Method Using Single-Channel ECG Signals

被引:3
|
作者
Kumar, Chandra Bhushan
Mondal, Arnab Kumar [1 ,2 ]
Bhatia, Manvir [3 ]
Panigrahi, Bijaya Ketan [4 ]
Gandhi, Tapan Kumar [4 ]
机构
[1] Indian Inst Technol, Bharti Sch Telecommun & Management, New Delhi 110016, India
[2] Indian Inst Technol, Sch Informat Technol, New Delhi 110016, India
[3] Neurol & Sleep Ctr, New Delhi 110016, India
[4] Indian Inst Technol, Dept Elect Engn, New Delhi 110016, India
关键词
Sleep apnea; Training; Self-supervised learning; Representation learning; Electrocardiography; Deep learning; Data models; 1-D convolutional neural network (CNN); apnea-hypopnea index (AHI); contrastive learning; obstructive sleep apnea (OSA); self-supervised learning; sleep apnea (SA); SLEEP-APNEA; NEURAL-NETWORK; CLASSIFICATION; OXIMETRY; EVENTS;
D O I
10.1109/TIM.2023.3261931
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
Sleep apnea (SA) is a pervasive and highly prevalent sleep disorder identified by recurrent breathing-related problems such as respiratory pauses for almost 10 s (called apnea events) during sleep. It is a strongly underdiagnosed problem because the person suffering from this disease is not aware of this situation. It may cause serious health issues and badly affect the quality of life. Therefore, the diagnosis of sleep is crucial to cure disease. Polysomnography (PSG) is a golden technique for diagnosing sleep disorders. In this technique, multiple sensors are used to collect specific physiological signals such as electroencephalogram (EEG), electromyogram (EMG), electrooculogram (EOG), and many more. In regular clinical practice, medical experts need to manually analyze the signals of sleep hours which is a tedious process. Therefore, the automatic diagnosis tool is needed to simplify this process. Recently, many research groups have proposed deep learning models for the automatic diagnosis of SA using physiological signals with good accuracy. However, all these models require a large amount of annotated data in the supervised training process, which limits the use of those models in real-time scenarios. However, annotating a huge amount of biomedical signals is challenging and requires lots of time and domain expertise. This study proposes a self-supervised representation learning (SSRL) method for detecting hypopnea events from single-channel electrocardiography (ECG) signals. The proposed model is trained in two phases. In the first training phase, an encoder is trained to learn signal representation from the unlabeled data. In the second training phase, the classifier and the encoder are fine-tuned for the classification. Our proposed model performed well on the test dataset with a per-segment classification accuracy of 85%, 89%, and 92% using only 1%, 10%, and 100% of the training data with labels, respectively, for fine-tuning encoder along with the classifier. Also, our proposed model can identify a person suffering from the obstructive SA (OSA) with the accuracy of 100%, even when the encoder and classifier are fine-tuned using only 1% of the training data with the label. The proposed model outperformed the state-of-the-art techniques and can be implemented offline or online for rapid and accurate diagnosis of the problem.
引用
收藏
页数:15
相关论文
共 50 条
  • [21] Contrastive Self-Supervised Learning-Based Background Reconstruction for Hyperspectral Anomaly Detection
    Sun, Xiaoming
    Zhang, Yuxiang
    Dong, Yanni
    Du, Bo
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2025, 63
  • [22] SELF-SUPERVISED LEARNING FOR ECG-BASED EMOTION RECOGNITION
    Sarkar, Pritam
    Etemad, Ali
    2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 3217 - 3221
  • [23] Self-Supervised Video Representation Learning by Video Incoherence Detection
    Cao, Haozhi
    Xu, Yuecong
    Mao, Kezhi
    Xie, Lihua
    Yin, Jianxiong
    See, Simon
    Xu, Qianwen
    Yang, Jianfei
    IEEE TRANSACTIONS ON CYBERNETICS, 2024, 54 (06) : 3810 - 3822
  • [24] Self-Supervised Graph Representation Learning Method Based on Data and Feature Augmentation
    Xu, Yunfeng
    Fan, Hexun
    Computer Engineering and Applications, 2024, 60 (17) : 148 - 157
  • [25] A Novel Self-Supervised Learning-Based Anomalous Node Detection Method Based on an Autoencoder for Wireless Sensor Networks
    Ye, Miao
    Zhang, Qinghao
    Xue, Xingsi
    Wang, Yong
    Jiang, Qiuxiang
    Qiu, Hongbing
    IEEE SYSTEMS JOURNAL, 2024, 18 (01): : 256 - 267
  • [26] Contrastive Self-Supervised Learning for Stress Detection from ECG Data
    Rabbani, Suha
    Khan, Naimul
    BIOENGINEERING-BASEL, 2022, 9 (08):
  • [27] Channel randomisation: Self-supervised representation learning for reliable visual anomaly detection in speciality crops
    Choi, Taeyeong
    Would, Owen
    Salazar-Gomez, Adrian
    Liu, Xin
    Cielniak, Grzegorz
    COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2024, 226
  • [28] Self-supervised learning representation for abnormal acoustic event detection based on attentional contrastive learning
    Wei, Juan
    Zhang, Qian
    Ning, Weichen
    DIGITAL SIGNAL PROCESSING, 2023, 142
  • [29] Automated Classification of Sleep Stages Using Single-Channel EEG: A Machine Learning-Based Method
    Satapathy, Santosh Kumar
    Loganathan, D.
    INTERNATIONAL JOURNAL OF INFORMATION RETRIEVAL RESEARCH, 2022, 12 (02)
  • [30] Medical Image Classification Using Self-Supervised Learning-Based Masked Autoencoder
    Fan, Zong
    Wang, Zhimin
    Gong, Ping
    Lee, Christine U.
    Tang, Shanshan
    Zhang, Xiaohui
    Hao, Yao
    Zhang, Zhongwei
    Song, Pengfei
    Chen, Shigao
    Li, Hua
    MEDICAL IMAGING 2024: IMAGE PROCESSING, 2024, 12926