A Hybrid Feature Selection and Extraction Methods for Sleep Apnea Detection Using Bio-Signals

• Published in: Sensors (Basel, Switzerland) Vol. 20; no. 15; p. 4323
• Main Authors: Li, Xilin, Ling, Sai Ho, Su, Steven
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 03.08.2020; MDPI
• Subjects: Abdomen; Air flow; Algorithms; Automation; Classification; Datasets; Decision trees; Electrocardiography; Electroencephalography; Electromyography; Eye movements; feature extraction; Feature selection; Fuzzy sets; Humans; Methods; Neural networks; Oxygen content; Parameter sensitivity; Physiology; Polysomnography; Regularization; Respiration; Sleep apnea; Sleep Apnea Syndromes - diagnosis; Support Vector Machine; Support vector machines; Variance analysis; Wavelet transforms; polysomnography; feature extraction; sleep apnea; feature selection
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s20154323

People with sleep apnea (SA) are at increased risk of having stroke and cardiovascular diseases. Polysomnography (PSG) is used to detect SA. This paper conducts feature selection from PSG signals and uses a support vector machine (SVM) to detect SA. To analyze SA, the Physionet Apnea Database was used to obtain various features. Electrocardiography (ECG), oxygen saturation (SaO ), airflow, abdominal, and thoracic signals were used to provide various frequency-, time-domain and non-linear features ( = 87). To analyse the significance of these features, firstly, two evaluation measures, the rank-sum method and the analysis of variance (ANOVA) were used to evaluate the significance of the features. These features were then classified according to their significance. Finally, different class feature sets were presented as inputs for an SVM classifier to detect the onset of SA. The hill-climbing feature selection algorithm and the -fold cross-validation method were applied to evaluate each classification performance. Through the experiments, we discovered that the best feature set (including the top-five significant features) obtained the best classification performance. Furthermore, we plotted receiver operating characteristic (ROC) curves to examine the performance of the SVM, and the results showed the SVM with Linear kernel (regularization parameter = 1) outperformed other classifiers (area under curve = 95.23%, sensitivity = 94.29%, specificity = 96.17%). The results confirm that feature subsets based on multiple bio-signals have the potential to identify patients with SA. The use of a smaller subset avoids dimensionality problems and reduces the computational load.