Optimizing sleep staging on multimodal time series: Leveraging borderline synthetic minority oversampling technique and supervised convolutional contrastive learning

• Published in: Computers in biology and medicine Vol. 166; p. 107501
• Main Authors: Huang, Xinyu, Schmelter, Franziska, Irshad, Muhammad Tausif, Piet, Artur, Nisar, Muhammad Adeel, Sina, Christian, Grzegorzek, Marcin
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.11.2023; Elsevier Limited
• Subjects: Adult; Annotations; B-SMOTE; Biosignal processing; Classification; Cross-domain adaption; Datasets; Electroencephalography - methods; Electromyography - methods; Electrooculography - methods; Female; Humans; Internal Medicine; Learning; Male; Multimodal polysomnography; Optimization; Other; Oversampling; Polysomnography - methods; Precision medicine; Residual network; Signal Processing, Computer-Assisted; Sleep; Sleep disorders; Sleep stage classification; Sleep Stages - physiology; Supervised contrastive learning; Supervised Machine Learning; Time series; Truncated cross-entropy; Well being; B-SMOTE; Sleep stage classification; Truncated cross-entropy; Cross-domain adaption; Biosignal processing; Supervised contrastive learning; Multimodal polysomnography; Residual network
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2023.107501

Sleep is an important research area in nutritional medicine that plays a crucial role in human physical and mental health restoration. It can influence diet, metabolism, and hormone regulation, which can affect overall health and well-being. As an essential tool in the sleep study, the sleep stage classification provides a parsing of sleep architecture and a comprehensive understanding of sleep patterns to identify sleep disorders and facilitate the formulation of targeted sleep interventions. However, the class imbalance issue is typically salient in sleep datasets, which severely affects classification performances. To address this issue and to extract optimal multimodal features of EEG, EOG, and EMG that can improve the accuracy of sleep stage classification, a Borderline Synthetic Minority Oversampling Technique (B-SMOTE)-Based Supervised Convolutional Contrastive Learning (BST-SCCL) is proposed, which can avoid the risk of data mismatch between various sleep knowledge domains (varying health conditions and annotation rules) and strengthening learning characteristics of the N1 stage from the pair-wise segments comparison strategy. The lightweight residual network architecture with a novel truncated cross-entropy loss function is designed to accommodate multimodal time series and boost the training speed and performance stability. The proposed model has been validated on four well-known public sleep datasets (Sleep-EDF-20, Sleep-EDF-78, ISRUC-1, and ISRUC-3) and its superior performance (overall accuracy of 91.31–92.34%, MF1 of 88.21–90.08%, and Cohen’s Kappa coefficient k of 0.87–0.89) has further demonstrated its effectiveness. It shows the great potential of contrastive learning for cross-domain knowledge interaction in precision medicine. •B-SMOTE addresses class imbalance issue to reinforce N1 stage sleep pattern learning.•Supervised contrastive learning for similarity understanding in random feature pairs.•Contrastive learning for adaptable cross-domain matching in inter- and intra-datasets.•Lightweight ResNet & truncated cross-entropy to increase application feasibility.•Validation on four well-known public datasets to prove the model’s generalizability.