Obstructive sleep apnea event prediction using recurrence plots and convolutional neural networks (RP-CNNs) from polysomnographic signals

• Published in: Biomedical signal processing and control Vol. 69; p. 102928
• Main Authors: Taghizadegan, Yashar, Jafarnia Dabanloo, Nader, Maghooli, Keivan, Sheikhani, Ali
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2021
• Subjects: Convolutional neural network (CNN); Fusion; Obstructive sleep apnea (OSA); Polysomnography; Prediction; Recurrence plots (RPs); Weighted majority voting (WMV); Recurrence plots (RPs); Obstructive sleep apnea (OSA); Prediction; Convolutional neural network (CNN); Polysomnography; Fusion; Weighted majority voting (WMV)
• DOI: 10.1016/j.bspc.2021.102928
• ISSN: 1746-8094

•Representation of dynamic behavior of EEG, ECG and respiration channels of OSA patients using the recurrence plot (RP) were used to predict OSA events.•A novel system based on fusion of Recurrence Plot (RP), pre-trained convolutional neural networks and majority voting method was proposed to predict OSA.•Fusion of fine-tuned ShuffleNets on recurrence plots from EEG, ECG and respiration channels improves prediction rate of OSA compare to single channels. The prediction of Obstructive Sleep Apnea (OSA) through common polysomnographic signals before stop breathing triggers the ventilation-aided machines such as Continuous Positive Airway Pressure (CPAP). In this paper, a novel schema is proposed based on the representation of the dynamical behavior of polysomnographic signals. This procedure is accomplished using a combination of the Recurrence Plots (RPs) and Convolutional Neural Networks (CNNs), called RP-CNNs. In this regard, the OSA events of 30, 60, 90, and 120 s are predicted before the occurrence. The first phase was to create RP images via Electroencephalogram (EEG), Electrocardiogram (ECG), and respiration signals at a single level. Then, the RP images were independently fed into two fast and robust pre-trained CNNs, naming ResNet-18 and ShuffleNet. Thus, the networks were fine-tuned, and the mentioned events were classified. In the second phase, the classification results were fused using the Weighted Majority Voting (WMV) method to make the final decision. Finally, subject-dependent and subject-independent evaluation criteria were utilized for the MIT-BIH polysomnographic and Dublin sleep apnea databases. The RP-ShuffleNet and 10-fold cross-validation were employed to attain the highest average accuracy and Area Under the Curve (AUC) through 30-second intervals before the OSA events at fusion-level in MIT-BIH polysomnographic and Dublin sleep apnea databases. The achieved results were 90.72%, 0.8937, 90.45%, and 0.9010, respectively. Predicting the OSA events using representation of the dynamical behavior of polysomnographic signals and the fusion of results of the fine-tuned CNNs have been led to the enhancement of the results compared to the state-of-the-art studies.