An Empirical Mode Decomposition-Based Method for Feature Extraction and Classification of Sleep Apnea

摘要: Background: Sleep apnea is a breathing disorder found among thirty percentage of the total population. Polysomnography (PSG) analysis standard method used for identification sleep apnea. laboratories are conducting this test. Unavailability in rural areas makes detection difficult ordinary people. There different methods detecting Past researches show that electrocardiogram-based more accurate other signals. This paper investigates idea electrocardiogram (ECG) signals recognition Methods: In paper, classification healthy and subjects performed using The proper feature extraction from these signal segments executed with help empirical mode decomposition (EMD). EMD algorithm decomposes incoming into intrinsic functions (IMFs). Four morphological features extracted IMF levels. These include characteristics QRS complex, T P waves. done machine learning technique called support vector machine. Result: All experiments carried out by St. Vincents University Hospital/University College Dublin Apnea Database (UCD database). database available online physionet. It observed results decomposition; it could be possible to extract ECG segments. also enhances accuracy classifier. overall sensitivity, specificity, achieved proposed work 90, 85, 93.33%, respectively.