Obstructive Sleep Apnea Detection based on ECG Signal using Statistical Features of Wavelet Subband

One of the respiratory disorders is obstructive sleep apnea (OSA). OSA occurs when a person sleeps. causes breathing to stop momentarily due obstruction in airways. In this condition, people with will be deprived oxygen, awake and short breath. Diagnosis by doctor can done confirming patient's complaints during sleep, patterns, other symptoms that point OSA. Another way diagnosing polysomnography (PSG) examination laboratory analyze hypopnea. However, tends high cost time consuming. An alternative diagnostic tool an electrocardiogram (ECG) referring changes mechanism ECG-derived respiration (EDR). So digital ECG signal analysis potential for detection. Therefore, study, it proposed classify based on signals using wavelets statistical parameters. Statistical parameters include mean, variance, skewness kurtosis entropy calculated decomposition results. The validation performance method carried out support vector machine, k-nearest neighbor (k-NN), ensemble classifier. produces highest accuracy 89.2% bagged tree where all features are used as predictors. From hoped complete clinical diagnosis detecting