Detection of OSAHS using only time-domain property of snoring signal

Polysomnography (PSG) is the standard method for obstructive sleep apnea hypopnea syndrome (OSAHS) diagnosis in medical, but the shortcomings in it have pushed many researchers to find a more convenient and effective way to detect OSAHS. With this purpose, a new technique for OSAHS assessment is proposed in this paper. It combined the properties of respiratory events with an ingenious dynamic threshold endpoint detection method to realize the automatic detection of respiratory events, and then obtained the apnea hypopnea index (AHI) to determine the patient´s condition. The method was evaluated on the whole night snoring signal of 34 patients. Compared the results with those monitored by PSG, it has shown that the accuracy of this method can reach above 90%, at the same time, it has the advantages in low cost, high speed, easy to operate and so on.